Question
This module introduced an array of challenges and problems associated with globalization, including weaponized interdependence, the erosion of social cohesion within national polities, the digitization of globalization, and the diffusion & transformation of power in the global system. What do you see as the
biggest challenge and how should we address it?
Make an argument for why it is a serious problem, then make your case for how to address it. This essay will be submitted in the format of a policy memo – please see document for detailed
guidelines, rubric, and sample memo
.
12 point Times New Roman
1” margins
Single spaced
1-2 full pages
Note: you do not have to choose one of the challenges discussed in the module, you may also come up with your own (just run it by the professor first if you go “off-book”).
Rubric
Guidelines – GSC 512 Essay #1 [Policy Memo]
This module introduced an array of challenges and problems associated with globalization, including weaponized interdependence, the erosion of social cohesion within national polities, and the diffusion & transformation of power in the global system. What do you see as the biggest challenge and how should we address it?
Note: you do not have to choose one of the challenges discussed in the module, you may also come up with your own if you run it by the professor first.
Your memo should be 12-point Times New Roman, 1” margins, SINGLE spaced. This is an exercise in clear, concise writing, which takes time, but is a very valuable skill for an analyst to have; DO NOT go over the 2-page limit. It should be formatted in accordance with the 2nd page of this guide. You may use section headers (such as background, summary, recommendations, etc. – whatever makes sense to you), however, make sure you hit all the points outlined and organize it in the given order.
You will be graded on how clearly and concisely you state your case, how well organized and easy to read the memo is, and whether your recommended courses of action clearly align with the primary goal you identify. Please see below for the grading rubric:
GSC 512 – Essay
Criteria
Pts
Structure
Essay has a clear structure. Well-organized sentences and paragraphs flow logically from one to the next and are connected into a coherent overall argument.
30 pts
Writing
Clear, concise prose; persuasively written; correct grammar, punctuation, sentence structure, word choice.
30 pts
Content
– answers the prompt/question, drawing on material from the course
– makes a case that this is a serious problem or challenge
– clearly identifies an overall goal for the entity being asked to address it
– makes policy recommendations that clearly support the overall goal
100 pts
Total Points: 160 pts
If you struggle with grammar, syntax, sentence structure, or writing style, please consult the writing center (or ask a trusted person to check it) before submission. Make sure you leave yourself time to review your work (in other words, don’t wait until the day it’s due to write it).
TO: Office of the President of the United States (you may also substitute this with another office/state/organization if you prefer)
FROM: Your Real Name
Your (Pretend) Job Title – just come up with something that makes sense
DATE: Today’s Date
SUBJECT: Policy Recommendation on [_____fill in blank here with your topic_____]
Introduction: Briefly summarize the issue. This should be as concise as possible – just hit the big picture/most important points that are necessary to make your argument. Explain why this is a major problem/challenge worthy of the president’s attention. End the introductory section by identifying what the primary goal(s) of the U.S. (or whatever entity you chose to situate this in) should be.
Body (tip: the body of your memo will be several paragraphs – make sure you choose your paragraph breaks appropriately- it should be well organized).
Identify 1-3 specific policy recommendations that can be taken to meet this goal. For each one, describe it, explain how it will help meet this goal, make an argument in favor of it!
If you are recommending multiple action items, start your next paragraph by identifying another specific action that can be taken to meet this goal and repeat the exercise: describe, explain, make a case! Repeat this for as many specific actions you are planning to identify to support the overall goal you identified at the outset.
Alternatively, you may choose to identify only one policy action and spedn the rest of the memo making a strong case for that one. In this case, you should use the space to identify any potential pitfalls or new problems that may arise from these recommendations and suggest mitigation strategies or counter arguments. Do not be dismissive of the other side if you choose this route – show that you fully understand it and have taken the drawbacks seriously.
Conclusion: Link the recommended course(s) of action back to how it supports the big picture goal and your overall argument that this is a worthwhile goal to pursue. This is your last chance to make a compelling case for policy change. Keep it brief, direct, get right to the point… remember, they’re incredibly busy and scanning everything to see if it’s even worth their time first. If you ramble on, you lose the reader. Don’t waste their time reiterating what you already said (it’s a short memo and they just read it). Think about the biggest takeaway or the implications of this issue and/or policy change, and leave the reader wanting to take action on it (but don’t just say “we must take action”- that doesn’t count).
1
1
Introduction
Labor strikes in France at the end of 1995, which were aimed at reversing
the French government’s efforts to bring its budget in line with the Maas-
tricht criteria, threw the country into its worst crisis since 1968. Around
the same time in the United States, a prominent Republican was running a
vigorous campaign for the presidency on a plank of economic nationalism,
promising to erect trade barriers and tougher restrictions on immigration.
In the countries of Eastern Europe and in Russia, former communists
have won most of the parliamentary elections held since the fall of the
Berlin Wall, and communist candidate Gennady Zyuganov garnered 40
percent of the vote in the second round of the Russian presidential election
held in July 1996.
These apparently disparate developments have one common element:
the international integration of markets for goods, services, and capital
is pressuring societies to alter their traditional practices, and in return
broad segments of these societies are putting up a fight.1 The pressures
for change are tangible and affect all societies: In Japan, large corporations
have started to dismantle the postwar practice of lifetime employment,
one of Japan’s most distinctive social institutions. In Germany, the federal
government has been fighting union opposition to cuts on pension benefits
aimed at improving competitiveness and balancing the budget. In South
1. See the perceptive column by Thomas L. Friedman (1996). Friedman stresses that the
recent salience of such apparently diverse political movements as that of Patrick Buchanan
in the United States, Communists in Russia, and the Islamists in Turkey may be due to a
common root: a backlash against globalization. I thank Robert Wade for bringing Friedman’s
piece to my attention.
Institute for International Economics | http://www.iie.com
http://www.iie.com
2 HAS GLOBALIZATION GONE TOO FAR?
Korea, trade unions have gone on nationwide strikes to protest new
legislation making it easier for firms to lay off workers. Developing coun-
tries in Latin America have been competing with each other in opening
up to trade, deregulating their economies, and privatizing public enter-
prises. Ask business executives or government officials why these changes
are necessary, and you will hear the same mantra repeatedly: ‘‘We need
to remain (or become) competitive in a global economy.’’
The opposition to these changes is no less tangible and sometimes
makes for strange bedfellows. Labor unions decrying unfair competition
from underage workers overseas and environmentalists are joined by
billionaire businessmen Ross Perot and Sir James Goldsmith in railing
against the North American Free Trade Agreement (NAFTA) and the
World Trade Organization (WTO). In the United States, perhaps the most
free-market-oriented of advanced industrial societies, the philosophical
foundations of the classical liberal state have come under attack not only
from traditional protectionists but also from the new communitarian
movement, which emphasizes moral and civic virtue and is inherently
suspicious of the expansion of markets (see, e.g., Etzioni 1994; Sandel
1996).2
The process that has come to be called ‘‘globalization’’ is exposing a
deep fault line between groups who have the skills and mobility to flourish
in global markets and those who either don’t have these advantages or
perceive the expansion of unregulated markets as inimical to social stabil-
ity and deeply held norms. The result is severe tension between the market
and social groups such as workers, pensioners, and environmentalists,
with governments stuck in the middle.3
This book argues that the most serious challenge for the world economy
in the years ahead lies in making globalization compatible with domestic
social and political stability—or to put it even more directly, in ensuring
that international economic integration does not contribute to domestic
social disintegration.
Attuned to the anxieties of their voters, politicians in the advanced
industrial countries are well aware that all is not well with globalization.
The Lyon summit of the Group of Seven, held in June 1996, gave the
issue central billing: its communiqué was titled ‘‘Making a Success of
Globalization for the Benefit of All.’’ The communiqué opened with a
2. The cheerleaders on the side of globalization sometimes make for strange bedfellows
too. Consider, for example, the philosophy of an organization called the Global Awareness
Society International: ‘‘Globalization has made possible what was once merely a vision: the
people of our world united together under the roof of one Global Village.’’
3. See also Kapstein (1996) and Vernon (forthcoming). Kapstein argues that a backlash from
labor is likely unless policymakers take a more active role in managing their economies.
Vernon argues that we might be at the threshold of a global reaction against the pervasive
role of multinational enterprises.
Institute for International Economics | http://www.iie.com
http://www.iie.com
INTRODUCTION 3
discussion of globalization—its challenges as well as its benefits. The
leaders recognized that globalization raises difficulties for certain groups,
and they wrote:
In an increasingly interdependent world we must all recognize that we have an
interest in spreading the benefits of economic growth as widely as possible and
in diminishing the risk either of excluding individuals or groups in our own
economies or of excluding certain countries or regions from the benefits of global-
ization.
But how are these objectives to be met?
An adequate policy response requires an understanding of the sources
of the tensions generated by globalization. Without such an understand-
ing, the reactions are likely to be of two kinds. One is of the knee-jerk
type, with proposed cures worse than the disease. Such certainly is the
case with blanket protectionism à la Patrick Buchanan or the abolition of
the WTO à la Sir James Goldsmith. Indeed, much of what passes as
analysis (followed by condemnation) of international trade is based on
faulty logic and misleading empirics.4 To paraphrase Paul Samuelson,
there is no better proof that the principle of comparative advantage is the
only proposition in economics that is at once true and nontrivial than the
long history of misunderstanding that has attached to the consequences
of trade. The problems, while real, are more subtle than the terminology
that has come to dominate the debate, such as ‘‘low-wage competition,’’
or ‘‘leveling the playing field,’’ or ‘‘race to the bottom.’’ Consequently,
they require nuanced and imaginative solutions.
The other possible response, and the one that perhaps best characterizes
the attitude of much of the economics and policy community, is to down-
play the problem. Economists’ standard approach to globalization is to
emphasize the benefits of the free flow of goods, capital, and ideas and
to overlook the social tensions that may result.5 A common view is that
the complaints of nongovernmental organizations or labor advocates rep-
resent nothing but old protectionist wine in new bottles. Recent research
on trade and wages gives strength to this view: the available empirical
evidence suggests that trade has played a somewhat minor role in generat-
ing the labor-market ills of the advanced industrial countries—that is, in
increasing income inequality in the United States and unemployment
in Europe.6
4. Jagdish Bhagwati and Paul Krugman are two economists who have been tireless in
exposing common fallacies in discussions on international trade. See in particular Bhagwati
(1988) and Krugman (1996).
5. When I mention ‘‘economists’’ here, I am, of course, referring to mainstream economics,
as represented by neoclassical economists (of which I count myself as one).
6. Cline (1997) provides an excellent review of the literature. See also Collins (1996).
Institute for International Economics | http://www.iie.com
http://www.iie.com
4 HAS GLOBALIZATION GONE TOO FAR?
While I share the idea that much of the opposition to trade is based on
faulty premises, I also believe that economists have tended to take an
excessively narrow view of the issues. To understand the impact of global-
ization on domestic social arrangements, we have to go beyond the ques-
tion of what trade does to the skill premium. And even if we focus more
narrowly on labor-market outcomes, there are additional channels, which
have not yet come under close empirical scrutiny, through which
increased economic integration works to the disadvantage of labor, and
particularly of unskilled labor. This book attempts to offer such a broad-
ened perspective. As we shall see, this perspective leads to a less benign
outlook than the one economists commonly adopt. One side benefit, there-
fore, is that it serves to reduce the yawning gap that separates the views
of most economists from the gut instincts of many laypeople.
Sources of Tension
I focus on three sources of tension between the global market and social
stability and offer a brief overview of them here.
First, reduced barriers to trade and investment accentuate the asymme-
try between groups that can cross international borders (either directly
or indirectly, say through outsourcing7) and those that cannot. In the
first category are owners of capital, highly skilled workers, and many
professionals, who are free to take their resources where they are most
in demand. Unskilled and semiskilled workers and most middle managers
belong in the second category. Putting the same point in more technical
terms, globalization makes the demand for the services of individuals in
the second category more elastic—that is, the services of large segments
of the working population can be more easily substituted by the services
of other people across national boundaries. Globalization therefore funda-
mentally transforms the employment relationship.
The fact that ‘‘workers’’ can be more easily substituted for each other
across national boundaries undermines what many conceive to be a post-
war social bargain between workers and employers, under which the
former would receive a steady increase in wages and benefits in return
for labor peace. This is because increased substitutability results in the
following concrete consequences:
n Workers now have to pay a larger share of the cost of improvements
in work conditions and benefits (that is, they bear a greater incidence
of nonwage costs).
7. Outsourcing refers to companies’ practice of subcontracting part of the production pro-
cess—typically the most labor-intensive and least skill-intensive parts—to firms in other
countries with lower costs.
Institute for International Economics | http://www.iie.com
http://www.iie.com
INTRODUCTION 5
n They have to incur greater instability in earnings and hours worked in
response to shocks to labor demand or labor productivity (that is,
volatility and insecurity increase).
n Their bargaining power erodes, so they receive lower wages and bene-
fits whenever bargaining is an element in setting the terms of employ-
ment.
These considerations have received insufficient attention in the recent
academic literature on trade and wages, which has focused on the down-
ward shift in demand for unskilled workers rather than the increase in
the elasticity of that demand.
Second, globalization engenders conflicts within and between nations
over domestic norms and the social institutions that embody them. As the
technology for manufactured goods becomes standardized and diffused
internationally, nations with very different sets of values, norms, institu-
tions, and collective preferences begin to compete head on in markets for
similar goods. And the spread of globalization creates opportunities for
trade between countries at very different levels of development.
This is of no consequence under traditional multilateral trade policy of
the WTO and the General Agreement on Tariffs and Trade (GATT): the
‘‘process’’ or ‘‘technology’’ through which goods are produced is immate-
rial, and so are the social institutions of the trading partners. Differences
in national practices are treated just like differences in factor endowments
or any other determinant of comparative advantage. However, introspec-
tion and empirical evidence both reveal that most people attach values
to processes as well as outcomes. This is reflected in the norms that shape
and constrain the domestic environment in which goods and services
are produced—for example, workplace practices, legal rules, and social
safety nets.
Trade becomes contentious when it unleashes forces that undermine
the norms implicit in domestic practices. Many residents of advanced
industrial countries are uncomfortable with the weakening of domestic
institutions through the forces of trade, as when, for example, child labor
in Honduras displaces workers in South Carolina or when pension benefits
are cut in Europe in response to the requirements of the Maastricht treaty.
This sense of unease is one way of interpreting the demands for ‘‘fair
trade.’’ Much of the discussion surrounding the ‘‘new’’ issues in trade
policy—that is, labor standards, environment, competition policy, corrup-
tion—can be cast in this light of procedural fairness.
We cannot understand what is happening in these new areas until we
take individual preferences for processes and the social arrangements that
embody them seriously. In particular, by doing so we can start to make
sense of people’s uneasiness about the consequences of international eco-
nomic integration and avoid the trap of automatically branding all con-
Institute for International Economics | http://www.iie.com
http://www.iie.com
6 HAS GLOBALIZATION GONE TOO FAR?
cerned groups as self-interested protectionists. Indeed, since trade policy
almost always has redistributive consequences (among sectors, income
groups, and individuals), one cannot produce a principled defense of free
trade without confronting the question of the fairness and legitimacy of
the practices that generate these consequences. By the same token, one
should not expect broad popular support for trade when trade involves
exchanges that clash with (and erode) prevailing domestic social arrange-
ments.
Third, globalization has made it exceedingly difficult for governments
to provide social insurance—one of their central functions and one that
has helped maintain social cohesion and domestic political support for
ongoing liberalization throughout the postwar period. In essence, govern-
ments have used their fiscal powers to insulate domestic groups from
excessive market risks, particularly those having an external origin. In
fact, there is a striking correlation between an economy’s exposure to
foreign trade and the size of its welfare state. It is in the most open
countries, such as Sweden, Denmark, and the Netherlands, that spending
on income transfers has expanded the most. This is not to say that the
government is the sole, or the best, provider of social insurance. The
extended family, religious groups, and local communities often play simi-
lar roles. My point is that it is a hallmark of the postwar period that
governments in the advanced countries have been expected to provide
such insurance.
At the present, however, international economic integration is taking
place against the background of receding governments and diminished
social obligations. The welfare state has been under attack for two decades.
Moreover, the increasing mobility of capital has rendered an important
segment of the tax base footloose, leaving governments with the unappe-
tizing option of increasing tax rates disproportionately on labor income.
Yet the need for social insurance for the vast majority of the population
that remains internationally immobile has not diminished. If anything,
this need has become greater as a consequence of increased integration.
The question therefore is how the tension between globalization and the
pressures for socialization of risk can be eased. If the tension is not man-
aged intelligently and creatively, the danger is that the domestic consensus
in favor of open markets will ultimately erode to the point where a
generalized resurgence of protectionism becomes a serious possibility.
Each of these arguments points to an important weakness in the manner
in which advanced societies are handling—or are equipped to handle—
the consequences of globalization. Collectively, they point to what is
perhaps the greatest risk of all, namely that the cumulative consequence
of the tensions mentioned above will be the solidifying of a new set of
class divisions—between those who prosper in the globalized economy
and those who do not, between those who share its values and those who
Institute for International Economics | http://www.iie.com
http://www.iie.com
INTRODUCTION 7
would rather not, and between those who can diversify away its risks
and those who cannot. This is not a pleasing prospect, even for individuals
on the winning side of the divide who have little empathy for the other
side. Social disintegration is not a spectator sport—those on the sidelines
also get splashed with mud from the field. Ultimately, the deepening of
social fissures can harm all.
Globalization: Now and Then
This is not the first time we have experienced a truly global market. By
many measures, the world economy was possibly even more integrated
at the height of the gold standard in the late 19th century than it is now.
Figure 1.1 charts the ratio of exports to national income for the United
States, Western Europe, and Japan since 1870. In the United States and
Europe, trade volumes peaked before World War I and then collapsed
during the interwar years. Trade surged again after 1950, but none of the
three regions is significantly more open by this measure now than it was
under the late gold standard. Japan, in fact, has a lower share of exports
in GDP now than it did during the interwar period.
Institute for International Economics | http://www.iie.com
http://www.iie.com
8 HAS GLOBALIZATION GONE TOO FAR?
Other measures of global economic integration tell a similar story.
As railways and steamships lowered transport costs and Europe moved
toward free trade during the late 19th century, a dramatic convergence
in commodity prices took place (Williamson 1996). Labor flows were
considerably higher then as well, as millions of immigrants made their
way from the old world to the new. In the United States, immigration
was responsible for 24 percent of the expansion of the labor force during
the 40 years before World War I (Williamson 1996, appendix table 1). As
for capital mobility, the share of net capital outflows in GNP was much
higher in the United Kingdom during the classical gold standard than it
has been since.
Does this earlier period of globalization hold any lessons for our current
situation? It well might. There is some evidence, for example, that trade
and migration had significant consequences for income distribution.
According to Jeffrey Williamson, ‘‘[G]lobalization . . . accounted for more
than half of the rising inequality in rich, labor-scarce countries [e.g., the
United States, Argentina, and Australia] and for a little more than a
quarter of the falling inequality in poor, labor-abundant countries [e.g.,
Sweden, Denmark, and Ireland]’’ in the period before World War I (1996,
19). Equally to the point are the political consequences of these changes:
There is a literature almost a century old that argues that immigration hurt
American labor and accounted for much of the rise in inequality from the 1890s
to World War I, so much so that a labor-sympathetic Congress passed immigration
quotas. There is a literature even older that argues that a New World grain
invasion eroded land rents in Europe, so much so that landowner-dominated
Continental Parliaments raised tariffs to help protect them from the impact of
globalization. (Williamson 1996, 1)
Williamson (1996, 20) concludes that ‘‘the inequality trends which global-
ization produced are at least partly responsible for the interwar retreat
from globalization [which appeared] first in the rich industrial trading
partners.’’
Moreover, there are some key differences that make today’s global
economy more contentious. First, restrictions on immigration were not as
common during the 19th century, and consequently labor’s international
mobility was more comparable to that of capital. Consequently, the asym-
metry between mobile capital (physical and human) and immobile ‘‘natu-
ral’’ labor, which characterizes the present situation, is a relatively recent
phenomenon. Second, there was little head-on international competition
in identical or similar products during the previous century, and most
trade consisted of the exchange of noncompeting products, such as pri-
mary products for manufactured goods. The aggregate trade ratios do
not reflect the ‘‘vast increase in the exposure of tradable goods industries
to international competition’’ that is now taking place compared with the
situation in the 1890s (Irwin 1996, 42). Third, and perhaps most important,
Institute for International Economics | http://www.iie.com
http://www.iie.com
INTRODUCTION 9
governments had not yet been called on to perform social-welfare func-
tions on a large scale, such as ensuring adequate levels of employment,
establishing social safety nets, providing medical and social insurance,
and caring for the poor. This shift in the perceived role of government is
also a relatively recent transformation, one that makes life in an interde-
pendent economy considerably more difficult for today’s policymakers.
At any rate, the lesson from history seems to be that continued globaliza-
tion cannot be taken for granted. If its consequences are not managed
wisely and creatively, a retreat from openness becomes a distinct possi-
bility.
Implications
So has international economic integration gone too far? Not if policymak-
ers act wisely and imaginatively.
We need to be upfront about the irreversibility of the many changes
that have occurred in the global economy. Advances in communications
and transportation mean that large segments of national economies are
much more exposed to international trade and capital flows than they
have ever been, regardless of what policymakers choose to do. There is
only limited scope for government policy to make a difference. In addition,
a serious retreat into protectionism would hurt the many groups that
benefit from trade and would result in the same kind of social conflicts
that globalization itself generates. We have to recognize that erecting
trade barriers will help in only a limited set of circumstances and that
trade policies will rarely be the best response to the problems that will
be discussed here. Transfer and social insurance programs will generally
dominate. In short, the genie cannot be stuffed back into the bottle, even
if it were desirable to do so. We will need more imaginative and more
subtle responses. I will suggest some guidelines in the concluding chapter.
Even so, my primary purpose in this book is not prescriptive; it is to
broaden the debate on the consequences of globalization by probing
deeper into some of the dimensions that have received insufficient atten-
tion and ultimately recasting the debate so as to facilitate a more produc-
tive dialogue between opposing groups and interests. It is only through
greater understanding of what is at stake that we can hope to develop
appropriate public policies.
One final introductory note. I hope the reader will soon realize that
this book is not a one-sided brief against globalization. Indeed, the major
benefit of clarifying and adding rigor to some of the arguments against
trade is that it helps us draw a distinction between objections that are
valid (or at least logically coherent) and objections that aren’t. From this
perspective, what I end up doing, at least on occasion, is strengthening
the arsenal of arguments in favor of free trade. If this book is viewed as
Institute for International Economics | http://www.iie.com
http://www.iie.com
10 HAS GLOBALIZATION GONE TOO FAR?
controversial, it will have done its job; I have failed if it is perceived
as polemical.
The chapters that follow will elaborate on the three sources of tension
between globalization and society identified above and will review the
relevant empirical evidence. The objectives will be to cast the debate in
terms that both sides—economists and populists alike—can join, marshal
evidence on the likely significance of the tension in question, and where
there is evidence for serious concern, open the debate on possible
remedies.
Institute for International Economics | http://www.iie.com
http://www.iie.com
1 Introduction
Sources of Tension
Globalization: Now and Then
Figure 1.1 Japan, the United States, and Western Europe: merchandise exports as a share of GDP, 1870-1992
Implications
69
5
Implications
If the arguments this book makes are right, there are two dangers arising
from complacency toward the social consequences of globalization. The
first of these, and the more obvious, is a political backlash against trade.
The candidacy of Patrick Buchanan in the Republican primaries of the
1996 presidential election revealed that protectionism can be a rather easy
sell when broad segments of American society are experiencing anxieties
linked, at least in part, to globalization. One wonders how much greater
Buchanan’s support would have been had the unemployment rate been,
say, 10 percent rather than 5.6 percent. Economists may complain that
protectionism is mere snake oil and argue that the ailments require alto-
gether different medicine. But intellectual arguments will not win hearts
and minds unless there are concrete solutions on offer. Trade protection,
for all its faults, has the benefit of concreteness.
Perhaps future Buchanans will be ultimately defeated, as Buchanan
himself was, by the common sense of the public. Even so, there is a second
and perhaps even more serious danger: that globalization will contribute
to social disintegration, as nations are split along lines of economic status,
mobility, region, or social norms. Even without a protectionist backlash,
a victory for globalization that comes at the price of social disintegration
will be a very hollow victory indeed.
Social Disintegration as the Price of
Economic Integration?
If not handled well, then, the social pressures unleashed by global eco-
nomic integration will likely result in bad economics and bad governance.
Institute for International Economics | http://www.iie.com
http://www.iie.com
70 HAS GLOBALIZATION GONE TOO FAR?
This is not only because globalization highlights and exacerbates tensions
among groups, which it does. It is also because it reduces the willingness
of internationally mobile groups to cooperate with others in resolving
disagreements and conflicts.
Far-sighted companies will tend to their own communities as they
globalize. But an employer that has an ‘‘exit’’ option is one that is less
likely to exercise the ‘‘voice’’ option. It is so much easier to outsource
than to enter a debate on how to revitalize the local economy. This means
that owners of internationally mobile factors become disengaged from
their local communities and disinterested in their development and pros-
perity—just as suburban flight in an earlier era condemned many urban
areas to neglect.
‘‘[D]iverse groups [in society] hold together,’’ wrote Bernard Crick
(1962, 24), ‘‘because they practice politics—not because they agree about
‘fundamentals,’ or some such concept too vague, too personal, or too
divine ever to do the job of politics for it. The moral consensus of a free
state is not something mysteriously prior to or above politics: it is the
activity (the civilizing activity) of politics itself.’’ Or as Albert Hirschman
(1994, 25) put it, ‘‘The community spirit that is normally needed in a
democratic market society tends to be spontaneously generated through
the experience of tending the conflicts that are typical of that society.’’
But what if globalization reduces the incentives to ‘‘tend’’ to these con-
flicts? What if, by reducing the civic engagement of internationally mobile
groups, globalization loosens the civic glue that holds societies together
and exacerbates social fragmentation?1
Hence globalization delivers a double blow to social cohesion—first by
exacerbating conflict over fundamental beliefs regarding social organiza-
tion and second by weakening the forces that would normally militate for
the resolution of these conflicts through national debate and deliberation.
These developments are afflicting all societies exposed to globalization,
with many developing countries perhaps even more exposed than the
advanced industrial countries. A recent analysis of Mexican society by
Jorge Castañeda (1996) is worth quoting extensively. Castañeda speaks
of ‘‘a new cleavage that is rapidly cutting across Mexican society’’:
This split separates those Mexicans plugged into the US economy from those
who are not. . . . It divides Mexicans who are highly sensitive to government
1. Here the debate on globalization joins the debate on social capital (Putnam 1996). Putnam
documents a significant decline in civic participation in the United States, and attributes it,
in large part, to television. There is now considerable empirical evidence that suggests social
fragmentation is detrimental to economic performance. Alesina and Rodrik (1994), among
others, show that income inequality reduces subsequent economic growth, Knack and Keefer
(1996) find levels of social trust to be positively correlated with investment, and Easterly
and Levine (1996) find a strong negative correlation between an index of ethnolinguistic
fragmentation and subsequent levels of economic growth.
Institute for International Economics | http://www.iie.com
http://www.iie.com
IMPLICATIONS 71
macroeconomic policy from those who are indifferent to it. It separates those who
correctly believe that politics and events in Mexico still determine their destiny
from those who just as rightly understand that the decisions most critical to their
lives are made in Washington and New York. It parts Mexicans who remain on
the margins of global flows of capital, goods, and services, even if they are not
on the margins of Mexican society, from those who are steadily being integrated
into those flows. This growing group of Mexicans oriented toward the United
States is isolated from much of the country’s economic tribulation and relatively
complacent about its political travails. (95)
With between one-fifth and one-quarter of the Mexican population tied
into the world economy in this fashion, Castañeda doubts that a social
explosion will happen. But, as he emphasizes, the presence of this group
also makes meaningful reform less likely: ‘‘[W]ithout a stake in political
change, [the segments of Mexican society linked to the world economy]
also have little reason to foster it’’ (1996, 100). Castañeda’s account vividly
describes an extreme form of the syndromes associated with globalization
cum social disintegration.
Markets are a social institution, and their continued existence is predi-
cated on the perception that their processes and outcomes are legitimate.
As Karl Polanyi (1944) pointed out more than 50 years ago, the interna-
tional market is the only market that is not regulated by an overarching
political authority. Consequently, transactions undertaken in the interna-
tional marketplace carry the least inherent legitimacy. This in itself is an
ongoing source of tension between globalization and society. The problem
becomes much worse when segments of society are perceived as having
broken their links with their local communities and become footloose.
Institutions that lose their legitimacy can no longer function, and markets
are no different.
As John Ruggie put it (1995, 508), ‘‘In some respects . . . the world
[today] finds itself faced with a challenge which is not unlike the one it
faced in 1945: devising compatible forms of international liberalization
and domestic stability.’’ That challenge is augmented by some key diffi-
culties. The United States is neither willing nor able to play the kind of
leadership role it did in the immediate aftermath of World War II, and
there is no alternative leader. Perhaps more seriously, there is a lack of
clear strategies on which to proceed, even if the United States or another
country were to provide the leadership.
Policy Implications
As emphasized in the introductory chapter, many of the underlying
changes that have occurred in the global economy are now irreversible.
Advances in transportation and communications technologies render
national borders more porous to foreign competition than they have ever
Institute for International Economics | http://www.iie.com
http://www.iie.com
72 HAS GLOBALIZATION GONE TOO FAR?
been, and nothing short of drastic government restrictions can alter that.
Protectionism is not a solution because it would likely generate its own
set of social conflicts, even if one were to discount its costs in terms of
economic efficiency. There are no easy fixes. We need to think imagina-
tively and creatively, without being blinded by ideologies that lead us to
overlook problems and/or their potential solutions.
John Maynard Keynes, one of the architects of the postwar international
economic system, once argued that the lack of intelligent alternatives to
free trade and economic liberalism was a key obstacle to implementing
a more desirable social system. ‘‘It must be admitted,’’ he wrote, ‘‘that
[the principles of laissez-faire] have been confirmed in the minds of sound
thinkers and reasonable public by the poor quality of the opponent pro-
posals—protectionism on the one hand, and Marxian socialism on the
other’’ (Keynes 1972 [1926], 285). Keynes, of course, was hardly an unadul-
terated free trader.2 What we need today is the same kind of pragmatic
approach to public policy problems that Keynes offered in his own time.
The Role of Economists
There is a big role in this for economists. International economists in
particular have been too Panglossian about the consequences of globaliza-
tion. Their approach on the labor-market consequences of trade, one area
in which they have actually engaged in the debate, has been too narrow,
resulting in a tendency to downplay the role of trade. They have been
too quick to paint those who have taken a more concerned stance as
ignorant of economics or as closet protectionists (and sometimes both).
Largely as a consequence, they have shut themselves out of the broader
policy debate. This is a pity because economics has much to contribute
here.
For example, there is much thinking to do on how to design appropriate
policies and institutions that can best address the need to provide social
insurance, which I have argued is a critical complement to the expansion
of global markets. As a general principle, the better targeted the policies
2. In particular, he saw a potential role for import tariffs—for, well, Keynesian reasons
having to do with aggregate demand and employment. In an oft-quoted article entitled
‘‘National Self-Sufficiency’’ (published in 1933), he went so far as to argue that international
economic relations among nations could be a source of international conflict. Hence the
famous passage: ‘‘I sympathise, therefore, with those who would minimise, rather than
those who would maximise, economic entanglement between nations. Ideas, knowledge,
art, hospitality, travel—these are things which should of their nature be international. But
let goods be homespun whenever it is reasonable and conveniently possible; and, above
all, let finance be primarily national.’’ The rest of the paragraph is not quoted as often: ‘‘Yet,
at the same time, those who seek to disembarrass a country from its entanglements should
be slow and wary. It should not be a matter of tearing up roots but of slowly training a
plant to grow a different direction’’ (Keynes 1982 [1933], 236).
Institute for International Economics | http://www.iie.com
http://www.iie.com
IMPLICATIONS 73
are to the sources of the problem, the less cost they will entail. If the
external risks that buffet national economies and workers were fully
observable, a set of transfers contingent on the realization of the shocks
would work best. But the world is obviously too complicated for first-
best solutions, and realistically we will have to sacrifice some efficiency.
So economists can help develop practical alternatives that provide some
insulation for the most affected groups without blunting market incentives
entirely. It is not entirely clear what role trade policy should play in this,
if any at all.3
Similarly, the mobility of capital and of employers both aggravate the
risks immobile groups face and render it more difficult to generate the
public resources needed to finance social insurance schemes. If this results
in globalization coming up against social and political constraints and a
backlash against trade, the mobility of employers creates a worldwide
negative externality. A logical implication is that some taxation of foot-
loose factors at the global level, with revenue sharing among nations,
may be worth considering. There is a parallel here with the ongoing
discussion of the Tobin tax, with the difference that the current idea
applies to physical (rather than financial) capital. Once again, there is
much thinking to do about the rationale and design of such a policy.
Finally, economists can draw on the literatures on institutional econom-
ics and political economy to formulate designs for a new system of global
safeguards. As I will argue further below, addressing the concerns dis-
cussed here will likely require a mixture of greater multilateral discipline
and broader access to an escape clause. The challenge is to enable countries
that are willing to engage in greater harmonization of domestic policies
to do so, while also allowing them to selectively delink from international
obligations when these obligations come into conflict with domestic norms
or institutions. How best to achieve this is an exciting intellectual chal-
lenge, with potentially large practical payoffs. I will discuss my own ideas
on this later in this chapter.
Hence economists could play a much more constructive role if they were
to recognize that the tensions between social stability and globalization are
real. They could help develop the conceptual frameworks needed for
rethinking the roles of governments and of international institutions in
this new phase of the global economy. They could assist policymakers in
finding the tools and instruments needed to achieve policy objectives
rather than taking issue with the objectives or denying that the prob-
lems exist.
By becoming engaged in this broader debate, economists can establish
greater credibility with the public as they attempt to clear up the misunder-
3. As expressed by Avinash Dixit in his comments on an earlier draft, ‘‘Designing more
efficient systems of social insurance, tailored to the particular shocks that matter to particular
countries, which will allow them to secure more of the benefits of integration and suffer
less of the social costs of it, is just the right kind of task for economists.’’
Institute for International Economics | http://www.iie.com
http://www.iie.com
74 HAS GLOBALIZATION GONE TOO FAR?
standings that opponents of trade often propagate. Keeping the debate
honest and grounded on solid empirical evidence is a natural role for
economists.
William Greider’s recent book, One World, Ready or Not—The Manic
Logic of Global Capitalism (1997), illustrates the appeal of many popular
misconceptions for some commentators. One of the main themes of Greid-
er’s book—that the global expansion of markets is undermining social
cohesion and moving the world inexorably toward a major economic and
political crisis—might be viewed simply as a bolder expression of the
potential danger I have highlighted here. Certainly, I am in sympathy with
many of Greider’s concerns—the consequences for low-skilled workers in
the advanced industrial countries, the weakening of social safety nets,
and the repression of political rights in some leading exporters such as
China and Indonesia. However, the book’s disregard for sound economic
analysis and for systematic empirical evidence makes it a very unreliable
treatise on what is happening and a faulty manual for setting things right.
The misconceptions that crop up in Greider’s book are easy ones for
economists to correct. Greider is wrong, for example, in thinking that low
wages are the driving force behind today’s global commerce. If that were
so, the world’s most formidable exporters would be Bangladesh and a
smattering of African countries. What he fails to take into account is the
importance of differences among countries in labor productivity. Further-
more, Greider (1997, 205) is wrong to attribute the US trade deficit to
the ‘‘unbalanced behavior’’ of US trade partners. If commercial policies
determined trade imbalances, India, until recently one of the world’s most
protectionist countries, would have been running large trade surpluses.
It is a mistake to claim that ‘‘the global economy [is] now a losing transac-
tion for the nation as a whole’’ because the United States’ net factor
payments abroad are positive (202). It is far from true that outward-
oriented industrialization in Southeast Asian countries has made life
worse rather than better for the former farmers who now toil in factories.
It is generally not the case that foreign-owned companies in developing
countries provide inferior working conditions to those that are available
elsewhere in the economy; in fact, the reverse is more often true.
Greider is particularly wrong in thinking that global capitalism inevita-
bly generates excess supply. This is the book’s key argument and ulti-
mately the main reason Greider believes the system will self-destruct.
Consider his discussion of Boeing’s outsourcing of some of its components
to the Xian Aircraft Company in China (155):
[W]hen new production work was moved to Xian from places like the United
States, the global system was, in effect, swapping highly paid industrial workers
for very cheap ones. To put the point more crudely, Boeing was exchanging a
$50,000 American machinist for a Chinese machinist who earned $600 or $700 a
year. Which one could buy the world’s goods? Thus, even though incomes and
purchasing power were expanding robustly among the new consumers of China,
Institute for International Economics | http://www.iie.com
http://www.iie.com
IMPLICATIONS 75
the overall effect was an erosion of the world’s potential purchasing power. If
one multiplied the Xian example across many factories and industrial sectors, as
well as other aspiring countries, one could begin to visualize why global consump-
tion was unable to keep up with global production.
The argument makes little sense, as any economist could point out.
The Chinese worker who earns only a tiny fraction of his American
counterpart will likely demonstrate commensurately lower productivity.
Even if this were not the case and the Chinese workers’ wages were
repressed below what their productivity ought to bring them, the result is
a transfer in purchasing power—to Boeing’s shareholders and the Chinese
employers—and not a diminution of purchasing power. Perhaps Greider
is thinking that Boeing’s shareholders and the Chinese employers have
a lower propensity to consume than the Chinese workers. But if this is
Greider’s reasoning, where is the argument and the evidence? Where is
the global surplus in savings and the secular decline in real interest rates
that we would surely have observed if income was shifting from low
savers to high savers?
It may be unfair to pick on Greider, especially as some of his other
conclusions are worth taking seriously. But the misunderstandings that
his book displays are commonplace in the globalization debate and do
not advance it. Professional economists have a duty to expose these misun-
derstandings and explicate them to a broader audience. But to become
true honest brokers, economists must demonstrate more modesty, less
condescension, and a willingness to broaden their focus.
The Role of Labor Advocates
There should be little doubt in the reader’s mind by now that I am
sympathetic to the difficulties experienced by workers in a globalized
economy. Indeed, much of this book is devoted to arguing that, where
low-skilled or less-educated workers are concerned, trade operates in a
less benign fashion than most trade economists concede. Policymakers
have to be cognizant of this and design their trade and other policies
accordingly. But there is a major responsibility here for labor groups
as well.
The political salience of labor’s voice in the United States (and to a
lesser extent in Europe) is currently diminished by at least three forces.
First, the same pressures that reduce the bargaining power of labor in
the workplace also reduce its power in the political marketplace. As
governments increasingly compete for footloose enterprises and capital,
the interests of workers (who after all have nowhere else to go) are
relegated to second place. ‘‘Competitiveness’’ becomes another word for
labor costs, something that can be enhanced by slashing benefits and
wages. Second, the excessive attachment of labor to a single political party
Institute for International Economics | http://www.iie.com
http://www.iie.com
76 HAS GLOBALIZATION GONE TOO FAR?
in the United States (and the United Kingdom) diminishes its political
power. Political parties are naturally more responsive to the interests of
those who are ready to shift their allegiances to competing parties than
they are to the interests of captive groups.4 Third, the receptivity of the
general public to the ideas of labor advocates is greatly reduced by the
protectionist tenor that too often characterizes these ideas.
Labor groups cannot do much about the first of these factors. The
second may also be difficult to change. Where labor advocates can make
the greatest difference is in distancing themselves from protectionist ideas.
This would result in the advocacy of a more pragmatic and therefore more
productive approach to trade policy. It would also lay the groundwork for
the political repositioning needed to make political parties from both
ends of the spectrum compete for the support of labor. Hence jettisoning
protectionist ideas would not only serve labor interests better, it would
also enhance labor’s political power.
These protectionist ideas find expression most frequently in complaints
about ‘‘low-wage, low-cost competition’’ from developing countries. But
such broad condemnations of trade miss the mark. They ignore the fact
that much of the difference in labor costs is typically due to lower levels
of labor productivity in the exporting countries. Wages in a poor exporting
country that are one-tenth the US level do not disadvantage workers in
the United States when labor productivity in that country is also lower
by a factor of 10.5 More broadly, gaps in labor costs that are due to
differences in the relative abundance of labor across countries are the
foundation for the gains from trade. It makes as little sense to restrict
trade for this reason alone as it does to restrict technological progress.
Consider, for example, the following statement by an AFL-CIO repre-
sentative:
We spend a great deal of time talking about free trade and comparative advantages,
and so forth, and I am sure these are important concepts and certainly we in the
US labor movement subscribe to them. Labor has benefited greatly from freeness
and free trade, not only internationally, but domestically, from the comparative
advantages that result from having a productive society as large and as diverse
as we do in the United States, but the American labor movement has always
taken the position that, to the maximum extent possible, labor costs should be removed
from that equation, because labor is more than just a cost of production. Labor
involves human dignity; it involves another whole dimension than does capital
or interest or the other factors of production, and it therefore has to be treated
very differently from them. (cited by Leebron 1996, note 67, emphasis added)
‘‘Removing labor costs from the equation,’’ as this statement calls for,
would remove the primary source of comparative advantage for develop-
4. See Dixit and Londregan (forthcoming) for a theoretical model that explicates this outcome.
5. Freeman (1994b) finds that approximately 80 percent of the difference in hourly pay
between the United States and Mexico is accounted for by differences in the skill mix of
labor in the two countries and by differences in the purchasing power of wages.
Institute for International Economics | http://www.iie.com
http://www.iie.com
IMPLICATIONS 77
ing countries and would deprive the US economy of the gains from trade
arising from it.
To the extent that there are concerns about fairness, the issue is not
labor costs per se, but how they are determined. Popular discussions of
trade often gloss over this distinction. In any case, the main competitive
threat to US labor, except in some highly labor-intensive sectors, comes
from workers in other advanced countries, many of which have labor
standards and benefit levels that are superior to those in the United States.
Hence the labor movement cannot afford to be—or to be perceived as
being—against trade. This requires recognition by labor unions that
imports go together with exports. One cannot be in favor of exports and
against imports without committing mercantilist fallacies. Similarly, labor
advocates have to accept that trade deficits are the consequence of macro-
economic realities; they have little to do with trade policies abroad, and
they cannot be corrected by trade restrictions at home. The sooner the
labor movement sheds such misconceptions, the sooner it will find allies
in the economics and policy community.
Labor should advocate a global economy that carries a more humane
face—one that recognizes national diversity and leaves room for national
differences in institutions. Domestically, it should work toward labor-
market institutions that enhance the mobility of workers and reduce the
risks they face (some ideas toward this end are suggested below).
The Role of National Governments
Policymakers have to steer a difficult middle course between responding
to the concerns discussed here and sheltering groups from foreign compe-
tition through protectionism. I can offer no hard-and-fast rules here, only
some guiding principles.
Strike a Balance between Openness and Domestic Needs
This book has argued that there is often a trade-off between maintaining
open borders to trade and maintaining social cohesion. When the conflict
arises—when new liberalization initiatives are under discussion, for
example—it makes little sense to sacrifice social concerns completely for
the sake of liberalization. Put differently, as policymakers sort out eco-
nomic and social objectives, free trade policies are not automatically enti-
tled to first priority.
Thanks to many rounds of multilateral trade liberalization, tariff and
nontariff restrictions on goods and many services are now at extremely
low levels in the industrial countries. Most major developing countries
have also slashed their trade barriers, often unilaterally and in conformity
with their own domestic reforms. Most economists would agree that
Institute for International Economics | http://www.iie.com
http://www.iie.com
78 HAS GLOBALIZATION GONE TOO FAR?
the efficiency benefits of further reductions in these existing barriers are
unlikely to be large. Indeed, the dirty little secret of international econom-
ics is that a tiny bit of protection reduces efficiency only a tiny bit. A logical
implication is that the case for further liberalization in the traditional area
of manufactured goods is rather weak.6
Moreover, there is a case for taking greater advantage of the World
Trade Organization’s existing escape clause, which allows countries to
institute otherwise-illegal trade restrictions under specified conditions, as
well as for broadening the scope of these multilateral safeguard actions
(see discussion below). In recent years, trade policy in the United States
and the European Union has gone in a rather different direction, with
increased use of antidumping measures and limited recourse to escape
clause actions. This is likely because WTO rules and domestic legislation
make the petitioning industry’s job much easier in antidumping cases:
there are lower evidentiary hurdles than in escape clause actions,7 no
determinate time limit, and no requirement for compensation for affected
trade partners, as the escape clause provides. Also, escape clause actions,
unlike antidumping duties, require presidential approval in the United
States. This is an undesirable situation because antidumping rules are, on
the whole, consistent neither with economics principles nor, as discussed
below, with fairness. Tightening the rules on antidumping in conjunction
with a reconsideration and reinvigoration of the escape clause mechanism
would make a lot of sense.8
Do Not Neglect Social Insurance
Policymakers have to bear in mind the important role that the provision
of social insurance, through social programs, has played historically in
enabling multilateral liberalization and an explosion of world trade. As
the welfare state is being pruned, there is a real danger that this contribu-
tion will be forgotten.
This does not mean that fiscal policy has to be profligate and budget
deficits large. Nor does it mean a bigger government role. Enhanced levels
of social insurance, for better labor-market outcomes, can be provided in
most countries within existing levels of spending. This can be done, for
6. Of course, since trade barriers are still higher elsewhere than in the United States, multilat-
eral liberalization would generate relatively greater trade opportunities for the United States.
See Bergsten (1996) for an argument that emphasizes this ‘‘asymmetric’’ nature of the benefits.
7. In the United States, escape clause action requires demonstration of ‘‘serious injury’’
rather than ‘‘material injury,’’ the latter being the lower threshold, which applies to anti-
dumping. WTO rules also require that escape clause actions be nondiscriminatory, unlike
antidumping, which can apply to any particular exporting country. Of course, an antidump-
ing action requires a demonstration that there is dumping, but in practice US Commerce
Department criteria for what constitutes ‘‘dumping’’ are not at all restrictive.
8. This was one of the options considered by Schott (1990).
Institute for International Economics | http://www.iie.com
http://www.iie.com
IMPLICATIONS 79
example, by shifting the composition of income transfers from old-age
insurance (i.e., social security) to labor-market insurance (i.e., unemploy-
ment compensation, trade adjustment assistance, training programs).
Because pensions typically constitute the largest item of social spending
in the advanced industrial countries, better targeting of this sort is highly
compatible with responsible fiscal policies. Gearing social insurance more
directly toward labor markets, without increasing the overall tax burden,
would be one key step toward alleviating the insecurities associated with
globalization.
There is a widespread feeling in many countries that, in the words
of Tanzi and Schuknecht (1995, 17), ‘‘[s]ocial safety nets have . . . been
transformed into universal benefits with widespread free-riding behavior,
and social insurance has frequently become an income support system
with special interests making any effective reform very difficult.’’ Further,
‘‘various government performance indicators suggest that the growth in
spending after 1960 may not have brought about significantly improved
economic performance or greater social progress’’ (1995, 20). However,
this book has suggested that social spending has had the important func-
tion of buying social peace. Without disagreeing about the need to elimi-
nate waste and reform in the welfare state more broadly, I would argue
that the need for social insurance does not decline but rather increases
as global integration increases. So the message to reformers of the social
welfare system is, don’t throw the baby out with the bath water.9
Do Not Use ‘‘Competitiveness’’ as an Excuse for Domestic Reform
One of the reasons globalization gets a bad rap is that policymakers often
fall into the trap of using ‘‘competitiveness’’ as an excuse for needed
domestic reforms. Large fiscal deficits or lagging domestic productivity
are problems that drag living standards down in many industrial countries
and would do so even in closed economies. Indeed, the term ‘‘competitive-
ness’’ itself is largely meaningless when applied to whole economies,
unless it is used to refer to things that already have a proper name—such
as productivity, investment, and economic growth. Too often, however,
the need to resolve fiscal or productivity problems is presented to the
electorate as the consequence of global competitive pressures. This not
only makes the required policies a harder sell—why should we adjust
just for the sake of becoming better competitors against the Koreans
or the Mexicans?—it also erodes the domestic support for international
9. Many economists would agree that the amount of resources needed to keep the most
disadvantaged from falling through the cracks is actually not that big. Krugman (1996) cites
a figure of 2 percent of GDP. In absolute terms, this is, of course, a lot of money, but it is
less than half of what an average OECD country spends on servicing the public debt
each year.
Institute for International Economics | http://www.iie.com
http://www.iie.com
80 HAS GLOBALIZATION GONE TOO FAR?
trade—if we have to do all these painful things because of trade, maybe
trade isn’t such a wonderful thing anyhow!
The French strikes of 1995 are a good case in point. What made the
opposition to the proposed fiscal and pension reforms particularly salient
was the perception that fundamental changes in the French way of life
were being imposed for the sake of international economic integration. The
French government presented the reforms as required by the Maastricht
criteria, which they were. But presumably, the Maastricht criteria them-
selves reflected the policymakers’ belief that a smaller welfare state would
serve their economies better in the longer run. By and large, the French
government did not make the case for reform on its own strengths. By
using the Maastricht card, it turned the discussion into a debate on Euro-
pean economic integration. Hence the widespread public reaction, which
extended beyond just those workers whose fates would be immediately
affected.
The lesson for policymakers is, do not sell reforms that are good for
the economy and the citizenry as reforms that are dictated by international
economic integration.
Do Not Abuse ‘‘Fairness’’ Claims in Trade
The notion of fairness in trade is not as vacuous as many economists
think. Consequently, nations have the right—and should be allowed—
to restrict trade when it conflicts with widely held norms at home or
undermines domestic social arrangements that enjoy broad support.
But there is much that is done in the name of ‘‘fair trade’’ that falls far
short of this criterion. There are two sets of practices in particular that
should be immediately suspect. One concerns complaints made against
other nations when very similar practices abound at home. Antidumping
proceedings are a clear example: standard business practices, such as
pricing over the life of a product or pricing over the business cycle, can
result in duties being imposed on an exporting firm. There is nothing
‘‘unfair’’ about these business practices, as is made abundantly clear by
the fact that domestic firms engage in them as well.
The second category concerns cases in which other nations are unilater-
ally asked to change their domestic practices so as to equalize competitive
conditions. Japan is frequently at the receiving end of such demands
from the United States and the European Union. A more recent example
concerns the declaration by the US Trade Representative that corruption
in foreign countries will henceforth be considered as unfair trade. While
considerations of fairness and legitimacy will guide a country’s own social
arrangements, even by restricting imports if need be, such considerations
should not allow one country to impose its own institutions on others.
Proponents of fair trade must bear this key distinction in mind. Thus, it
Institute for International Economics | http://www.iie.com
http://www.iie.com
IMPLICATIONS 81
is perfectly legitimate for the United States to make it illegal for domestic
firms to engage in corrupt practices abroad (as was done with the Foreign
Corrupt Practices Act of 1977). It is also legitimate to negotiate a multilat-
eral set of principles with other countries in the Organization for Economic
Cooperation and Development (OECD) with broadly similar norms. It
may also be legitimate to restrict imports from a country whose labor
practices broad segments of the domestic population deem offensive.
But it is not acceptable to unilaterally threaten retaliation against other
countries because their business practices do not comply with domestic
standards at home in order to force these countries to alter their own standards.10
Using claims of fairness to advance competitive aims is coercive and
inherently contradictory. Trying to ‘‘export’’ norms by asking other coun-
tries to alter their social arrangements to match domestic ones is inappro-
priate for the same reason.
The Role of International Institutions
One area in which international cooperation can be helpful has already
been mentioned: the ability of firms to play national tax authorities off
each other is a source of negative cross-border externality, as it undercuts
the revenue sources needed to maintain social and political cohesion and
ultimately erodes support for free trade. Greater exchange of information
among tax authorities would be one small step in the right direction.
Negotiating an international convention to restrict international firms’
ability to evade taxation via foreign investment would constitute a more
ambitious effort, but one that would have a greater chance of making
a difference.
There is a growing realization among governments that something
along these lines may need to be done. Concern about the revenue conse-
quences of tax competition recently led the OECD to set up a task force
(with priority funding) on curbing such competition among its member
states. As the OECD statement recognizes, globalization ‘‘opens up the
risk of competitive bidding between countries for mobile business.’’ The
task force’s first task is to ‘‘examine criteria for distinguishing between
fair and harmful tax competition.’’11 To be fully effective, such an effort
10. It may be that restricting imports will cause the exporting country to alter its practices,
irrespective of whether that was the stated goal of the policy. But that does not make the
distinction any less valid. The motives that drive trade policy in the advanced industrial
countries are usually transparent. There is little doubt that the Foreign Corrupt Practices
Act of 1977, for example, was motivated by domestic ethical considerations, while many
US and European complaints against Japan and some developing countries are clearly
driven by a desire to make ‘‘them’’ more like ‘‘us.’’ How foreign trade partners choose to
react to policies of the first kind (the ‘‘legitimate’’ actions, that is) is their own business.
11. The quotes are from the OECD’s Internet statement on the project (see www.oecd.org/
daf/fa/taxcomp.htm; Financial Times, 13 January 1997, 16).
Institute for International Economics | http://www.iie.com
http://www.oecd.org
http://www.iie.com
82 HAS GLOBALIZATION GONE TOO FAR?
has to enlist the cooperation of non-OECD countries as well. This is
implicit in the OECD’s approach, as its task force is slated to look at
practices in such tax havens as the Cayman Islands as well as the more
modest preferential tax regimes of countries such as Ireland, the Nether-
lands, and Belgium (Financial Times, 13 January 1997, 16).
More broadly, the arguments made in this book have two somewhat
conflicting implications for multilateral institutions. On the one hand,
these institutions must encourage greater convergence of policies and
standards (‘‘deep integration’’) among willing countries to help reduce ten-
sions arising from differences in national practices.12 On the other, they
must make room for selective disengagement from multilateral disci-
plines, under well-specified contingencies, for countries that need breath-
ing room to satisfy domestic requirements that are in conflict with liberal-
izing trade.
The apparent tension between these two objectives is partly reconciled
by the caveat in the previous sentence. These organizations will need to
set up a well-defined and multilaterally agreed set of hurdles that must
be cleared before a nation can exercise the selective disengagement in
question—be it higher tariffs, a quota, or an exemption from harmoniza-
tion requirements. In other words, there need to be multilateral rules on
how one can depart from multilateral rules!13
Of course, this is what the escape clause mechanism under the WTO,
and before it the General Agreement on Tariffs and Trade (GATT), is in
principle all about. But the mechanism has not served its purpose well.
Governments have preferred other measures to the GATT’s safeguard
mechanism. Hence ‘‘gray area’’ measures, such as voluntary export
restraints (VERs), proliferated before the Uruguay Round ended, and
there has been an explosion of antidumping cases. There were only 150
official safeguard actions under the GATT (through 1994) but more than
1,000 antidumping cases at the national level between 1985 and 1992 alone
(Hoekman and Kostecki 1995, chapter 7). Antidumping procedures are
12. Lawrence, Bressand, and Ito (1996) argue for the creation of a series of ‘‘clubs’’ among
partners willing to engage in deeper integration in areas not well covered in the WTO—
such as competition policy or environment. This would be a departure from unconditional
multilateralism and would risk institutionalizing discriminatory treatment of trade partners.
I would prefer to see the WTO used for these new areas, with a more effective escape clause
as a safety valve (see discussion below).
13. It is recognized in the theory of repeated games that sustaining cooperation among
players throughout a long (infinite) horizon when there are shocks to the system may require
periods of ‘‘noncooperation.’’ An appropriate multilateral trading regime will recognize
this and incorporate ‘‘safety valves’’ into its rules—that is, exemptions from its requirements
under specified contingencies. See Bagwell and Staiger (1990) for a formal model that justifies
the escape clause in these terms. In the context of such models, my argument is that social
conflicts result in a greater temptation to ‘‘defect’’ during bad times and hence require a
more accessible escape clause to render long-term cooperation sustainable.
Institute for International Economics | http://www.iie.com
http://www.iie.com
IMPLICATIONS 83
today effectively serving as the safeguard mechanism of choice. This
subverts the trade regime, gives safeguards a bad name, and crowds out
an effective outlet for legitimate concerns.
Hence, a revamped and expanded safeguards clause, along with tighter
restrictions on the use of antidumping, would be well worth considering.14
By broadening the current Agreement on Safeguards, WTO members can
then allay US and EU fears of import surges, which have so far prevented
the reining in of antidumping measures.
Currently, the Agreement on Safeguards allows temporary increases
in trade restrictions under a very narrow set of conditions. It requires a
determination that increased imports ‘‘cause or threaten to cause serious
injury to the domestic industry’’15 and that causality is firmly established.
Furthermore, injury cannot be attributed solely to imports if there are
multiple causes for it.16 Safeguards cannot be applied to developing-coun-
try exporters unless their share of imports of the product concerned is
above a given threshold. A country applying safeguard measures has to
compensate the affected exporters by providing ‘‘equivalent concessions,’’
lacking which the exporter is free to retaliate.
A broader interpretation of safeguards would acknowledge that coun-
tries may legitimately wish to restrict trade for reasons going beyond
competitive threats to their industries. Distributional concerns or conflicts
with domestic norms or social arrangements are among such legitimate
reasons. One could imagine recasting the current agreement into an Agree-
ment on Social Safeguards, which would permit the application of safe-
guard measures under a broader range of circumstances. This would
require recasting the ‘‘serious injury’’ test. I would replace the injury
criterion with another hurdle: the need to demonstrate broad domestic
support, among all concerned parties, for the proposed safeguard measure.
To see how that might work in practice, consider what the current
agreement says:
A Member may apply a safeguard measure only following an investigation by
the competent authorities of that Member pursuant to procedures previously
established and made public in consonance with Article X of the GATT 1994. This
investigation shall include reasonable public notice to all interested parties and
public hearings or other appropriate means in which importers, exporters and other
14. A refurbished escape clause, under which all trade relief would be centered, was pro-
posed in Hufbauer and Rosen (1986). See also Perez-Lopez for a similar approach (1989).
My discussion of the expanded safeguards clause is based on earlier work reported in
Rodrik (1995).
15. Serious injury is defined as ‘‘a significant overall impairment in the position of a domes-
tic industry.’’
16. According to the agreement, ‘‘When factors other than increased imports are causing
injury to the domestic industry at the same time, such injury shall not be attributed to
increased imports.’’
Institute for International Economics | http://www.iie.com
http://www.iie.com
84 HAS GLOBALIZATION GONE TOO FAR?
interested parties could present evidence and their views, including the opportunity
to respond to the presentations of other parties and to submit their views, inter
alia, as to whether or not the application of a safeguard measure would be in the public
interest [emphasis added]. The competent authorities shall publish a report setting
forth their findings and reasoned conclusions reached on all pertinent issues of
fact and law.
The main shortcoming of this clause is that while it allows all relevant
groups, exporters and importers in particular, to air their views, it does
not actually compel them to do so. Consequently, it results in a strong
bias in the domestic investigative process toward the interests of import-
competing groups, which are the petitioners for import relief and its
obvious beneficiaries. Indeed, this is a key problem with hearings in
antidumping proceedings, where testimony from other groups besides
the import-competing industry is not allowed.
A key reform, then, would be to require those conducting the investiga-
tions in each country to gather testimony and views from all relevant
parties, including consumer and public interest groups, importers of the
products concerned, and exporters to the affected country, and to deter-
mine whether there exists broad support among these groups for the appli-
cation of the safeguard measure in question. The requirement that groups
whose incomes would be hurt by imposing trade restrictions—importers
and exporters—be compelled to testify and that the investigative body
determine whether these groups also support the safeguard measure
would ensure that protectionism pure and simple would not have much
chance of success. At the same time, when deeply and widely held social
norms are at stake, these groups are unlikely to oppose safeguards in a
public manner, as this would endanger their standing among the public
at large. Imagine, for example, that slave labor is used in producing goods
for export in a given country. It is difficult to believe that exporters to
that country would publicly defend trade with it.
The main advantage of the proposed procedure is that it would force
a public debate on the legitimacy of trade and on the appropriateness of
restricting it. It ensures that all sides would be heard. This rarely happens
in practice, unless the trade partner in question is an important one.17
This procedure could also be complemented with a strengthened monitor-
ing and surveillance role for the WTO to ensure that domestic procedures
are in compliance with the expanded safeguard clause. An automatic sunset
clause could ensure that trade restrictions do not become entrenched long
after their perceived need has disappeared.
Broadening safeguard actions in this manner would not be without its
risks. One has to take into account the possibility that the new procedures
17. The public debate that surrounds the US president’s annual decision on whether to
extend China’s most-favored nation trade status is a good example. In my view, this debate
serves a useful function.
Institute for International Economics | http://www.iie.com
http://www.iie.com
IMPLICATIONS 85
would be abused for protectionist ends and that the door to unilateral
action on a broad front would be opened, despite the high threshold
envisaged here. But inaction is not without risk either. Absent creative
thinking and novel institutional designs, the tensions created by globaliza-
tion may spark a new set of ‘‘gray area’’ measures entirely outside multilat-
eral disciplines. That would be far worse than the revised safeguard
regime described here.
Concluding Remarks
Globalization is not occurring in a vacuum. It is part of a broader trend
that we may call marketization. Receding government, deregulation, and
the shrinking of social obligations are the domestic counterparts of the
intertwining of national economies. Globalization could not have
advanced this far without these complementary forces. The broader chal-
lenge for the 21st century is to engineer a new balance between market
and society, one that will continue to unleash the creative energies of
private entrepreneurship without eroding the social basis of cooperation.
The tensions between globalization and social cohesion are real, and
they are unlikely to disappear of their own accord. The proposals in this
chapter are little more than a beginning, and perhaps not even that. There
is no magic formula that can be applied. Indeed, part of the difficulty in
thinking prescriptively about these issues is that some of the basic analyti-
cal and empirical work on the consequences of globalization remains
to be done. Contrary to what many economists believe, we lack a full
understanding of how globalization works.
‘‘What is actually required to make progress with novel problems a
society encounters,’’ writes Albert Hirschman, ‘‘is political entrepreneur-
ship, imagination, patience here, impatience there, and other varieties of
virtù and fortuna. . .’’ (1994, 25). We need all of these, plus a good dose
of pragmatism, to make progress on the challenges ahead.
Institute for International Economics | http://www.iie.com
http://www.iie.com
Chapter 5: Implications
Social Disintegration as the Price of Economic Integration?
Policy Implications
The Role of Economists
The Role of Labor Advocates
The Role of National Governments
Strike a Balance between Openness and Domestic Needs
Do Not Neglect Social Insurance
Do Not Use “Competitiveness” as an Excuse for Domestic Reform
Do Not Abuse “Fairness” Claims in Trade
The Role of International Institutions
Concluding Remarks
1 8 4
I
n 1990, Argentina couldn’t have been in a worse economic mess.
In almost perpetual crisis since the seventies, the country reeled
under hyperinflation and a crushing debt burden. Incomes had
shrunk 25 percent from their levels a decade earlier and private
investment had come to a virtual standstill. Prices were rising at
unprecedented rates, even by Argentina’s demanding standards.
In March 1990, inflation climbed to more than 20,000 percent (on
an annualized basis), sowing chaos and confusion. Struggling to
cope, Buenos Aires’ world-weary residents took refuge in gallows
humor. With prices soaring by the minute, they told themselves, at
least it had become cheaper to take a cab than a bus. With the cab
you paid at the end of the ride instead of the beginning!
Can You Save an Economy by Tying It to
the Mast of Globalization?
Domingo Cavallo thought he knew the real problem. For too
long, Argentina’s governments had changed the rules of the game
whenever it suited them. Too much governmental discretion had
resulted in a complete loss of confidence in Argentine policy mak-
9
The Political Trilemma of the
World Economy
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
T he Polit ic al Tr ilemma of the Wor ld Economy 1 8 5
ers. The private sector had responded by withholding its invest-
ment and fleeing the domestic currency. To restore credibility
with domestic and foreign investors, the government needed to
commit itself to a clear set of rules. In particular, strict monetary
discipline was required to prevent governments from printing
money at will.1
Cavallo, an economist with a PhD from Harvard, was foreign
minister in the administration of President Carlos Menem. He
would get the chance to execute his plan when Menem put him
in charge of the economy in February 1991. The linchpin of Cav-
allo’s strategy was the Convertibility Law, which legally anchored
the Argentine currency to the U.S. dollar at 1 peso per dollar and
prohibited restrictions on foreign payments. The Convertibility
Law effectively forced Argentina’s central bank to operate by gold
standard rules. Henceforth the domestic money supply could be
increased and interest rates lowered only if dollars were flowing
into the economy. If dollars were moving the other way, the money
supply would have to be cut and interest rates raised. No more
mucking around with monetary policy.
In addition, Cavallo accelerated the privatization, deregula-
tion, and opening up of the Argentine economy. He believed open
economy rules and deep integration would reinforce business con-
fidence by precluding discretionary interventions and the hijack-
ing of policy by special interests. With policy on automatic pilot,
investors would have little fear that the rules would be changed on
them. By the early 1990s, Argentina’s record in trade liberaliza-
tion, tax reform, privatization, and financial reform was second to
none in Latin America.
Cavallo envisioned globalization as both a harness and an
engine for Argentina’s economy. Globalization provided not just
discipline and an effective shortcut to credibility in economic poli-
cies. It would also unleash powerful forces to propel the economy
forward. With lack of confidence and other transaction costs out
of the way, foreign capital would flow into the country, allowing
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
1 8 6 T he Globaliz at ion Par adox
domestic investment to rise and the economy to take off. Imports
from abroad in turn would force domestic producers to become
more competitive and productive. Deep integration with the world
economy would solve Argentina’s short- and long-term problems.
This was the Washington Consensus taken to an extreme, and it
turned out to be right about the short term, but not the long term.
Cavallo’s strategy worked wonders on the binding constraint of the
moment. The Convertibility Law eliminated hyperinflation and
restored price stability practically overnight. It generated credibil-
ity and confidence—at least for a while—and led to large capital
inflows. Investment, exports, and incomes all rose rapidly. As we
saw in chapter Six, Argentina became a poster child for multi-
lateral organizations and globalization enthusiasts in the mid-
1990s, even though policies like the Convertibility Law had clearly
not been part of the Washington Consensus. Cavallo became the
toast of the international financial community.
By the end of the decade, the Argentine nightmare had returned
with a vengeance. Adverse developments in the world economy set
the stage for an abrupt reversal in investors’ views on Argentina.
The Asian financial crisis hit the country hard by reducing inter-
national money managers’ appetite for emerging markets, but the
real killer was the Brazilian devaluation in early 1999. The devalu-
ation reduced the value of the Brazilian currency by 40 percent
against the dollar, allowing Brazilian exporters to charge much
lower dollar prices on foreign markets. Since Brazil is Argentina’s
chief global competitor, Brazil’s cost advantage left the Argentin-
ean peso looking decidedly overvalued. Doubts about Argentina’s
ability to service its external debt multiplied, confidence collapsed,
and before too long Argentina’s creditworthiness had slid below
some African countries’.
Cavallo’s relations with Menem had soured in the meantime and
he had left office in 1996. President Fernando de la Rúa, who suc-
ceeded Menem, invited Cavallo back to the government in March
2001 in an effort to shore up confidence. Cavallo’s new efforts
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
T he Polit ic al Tr ilemma of the Wor ld Economy 1 8 7
proved ineffective. When his initial tinkering with the trade and
currency regime produced meager results, he was forced to resort
to austerity policies and sharp fiscal cutbacks in an economy where
one worker out of five was already out of a job. He launched a
“zero-deficit” plan in July and enforced it with cuts in government
salaries and pensions of up to 13 percent. The financial panic
went from bad to worse. Fearing that the peso would be deval-
ued, domestic depositors rushed to pull their money out of banks,
which in turn forced the government to limit cash withdrawals.
The fiscal cuts and the restriction on bank withdrawals sparked
mass protests. Unions called for nationwide strikes, rioting envel-
oped major cities, and looting spread. Just before Christmas,
Cavallo and de la Rúa resigned in rapid succession.2 Starved of
funds, the Argentinean government was eventually forced to
freeze domestic bank accounts, default on its foreign debt, reim-
pose capital controls, and devalue the peso. Incomes shrunk by 12
percent in 2002, the worst drop in decades. The experiment with
hyperglobalization had ended in colossal failure.
What went wrong? The short answer is that domestic politics
got in the way of hyperglobalization. The painful domestic eco-
nomic adjustments required by deep integration did not sit well
with domestic constituencies, and politics ultimately emerged
victorious.
The Inevitable Clash Between Politics and Hyperglobalization
The economic story behind Argentina’s economic collapse is
fairly straightforward in hindsight. Argentina’s policy makers
had succeeded in removing one binding constraint—monetary
mismanagement—but eventually ran into another—an uncom-
petitive currency. Had the government abandoned the Convert-
ibility Law or reformed it in favor of a more flexible exchange rate,
say in 1996, the confidence crisis that engulfed the country later
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
1 8 8 T he Globaliz at ion Par adox
might have been averted. But Argentina’s policy makers were too
wedded to the Convertibility Law. They had sold it to their public
as the central plank of their growth strategy, making it virtually
impossible to step back. Pragmatism would have served the coun-
try better than ideological rigidity.
But there is a deeper political lesson in Argentina’s experience,
one that is fundamental to the nature of globalization. The coun-
try had bumped against one of the central truths of the global
economy: National democracy and deep globalization are incom-
patible. Democratic politics casts a long shadow on financial
markets and makes it impossible for a nation to integrate deeply
with the world economy. Britain had learned this lesson in 1931,
when it was forced to get off gold. Keynes had enshrined it in the
Bretton Woods regime. Argentina overlooked it.
The failure of Argentina’s political leaders was ultimately a mat-
ter not of will but of ability. Their commitment to the Convertibil-
ity Law and to financial market confidence could not have been
doubted. Cavallo knew there was little alternative to playing the
game by financial markets’ rules. Under his policies, the Argen-
tine government was willing to abrogate contracts with virtually
all domestic constituencies—public employees, pensioners, pro-
vincial governments, bank depositors—so as not to skip one cent
of its obligations to foreign creditors.
What sealed Argentina’s fate in the eyes of financial markets was
not what Cavallo and de la Rúa were doing, but what the Argen-
tine people were willing to accept. Investors and creditors grew
increasingly skeptical that the Argentine Congress, provinces, and
ordinary people would tolerate austerity policies long discredited
in advanced industrial countries. In the end, the markets were
right. When globalization collides with domestic politics, the
smart money bets on politics.
Remarkably, deep integration cannot sustain itself even when
its requirements and goals are fully internalized by a country’s
political leadership. For Cavallo, Menem, and de la Rúa, global-
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
T he Polit ic al Tr ilemma of the Wor ld Economy 1 8 9
ization was not a constraint to be respected willy-nilly; it was their
ultimate objective. Yet they could not keep domestic political pres-
sure from unraveling their strategy. The lesson for other countries
is sobering. If hyperglobalization could not be made to work in
Argentina, might it ever work in other settings?3
In his ode to globalization, The Lexus and the Olive Tree, Tom Fried-
man famously described how the “electronic herd”—financiers
and speculators who can move billions of dollars around the globe
in an instant—forced all nations to don a “Golden Straitjacket.”
This defining garment of globalization, he explained, stitched
together the fixed rules to which all countries must submit: free
trade, free capital markets, free enterprise, and small govern-
ment. “If your country has not been fitted for one,” he wrote, “it
will soon.” When you put it on, he continued, two things happen:
“your economy grows, and your politics shrink.” Since globaliza-
tion (which to Friedman meant deep integration) does not permit
nations to deviate from the rules, domestic politics is reduced to a
choice between Coke and Pepsi. All other flavors, especially local
ones, are banished.4
Friedman was wrong to presume that deep integration rules
produce rapid economic growth, as we have already seen. He was
also wrong to treat his Golden Straitjacket as an established real-
ity. Few countries’ leaders put on the Golden Straitjacket more
willingly than Argentina’s (who then also threw the keys away for
good measure). As the unraveling of the Argentine experiment
shows, in a democracy, domestic politics win out eventually. The
only exceptions are small nations that are already part of a larger
political grouping such as the European Union; we will look at
the case of Latvia in the next chapter. When push comes to shove,
democracy shrugs off the Golden Straitjacket.
Nevertheless, Friedman’s central insight remains valid. There
is a fundamental tension between hyperglobalization and demo-
cratic politics. Hyperglobalization does require shrinking domestic
politics and insulating technocrats from the demands of popular
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
1 9 0 T he Globaliz at ion Par adox
groups. Friedman erred when he overstated the economic ben-
efits of hyperglobalization and underestimated the power of poli-
tics. He therefore overestimated the long-run feasibility, as well as
desirability, of deep integration.
When Hyperglobalization Impinges on Democratic Choices
We cherish our democracy and national sovereignty, and yet we
sign one trade agreement after another and treat free capital flows
as the natural order of things. This unstable and incoherent state
of affairs is a recipe for disaster. Argentina in the 1990s gave us a
vivid and extreme example. However, one does not have to live in a
badly governed developing country ravaged by speculative capital
flows to experience the tension on an almost daily basis. The clash
between globalization and domestic social arrangements is a core
feature of the global economy. Consider a few illustrations of how
globalization gets in the way of national democracy.
Labor standards. Every advanced economy has detailed regula-
tions that cover employment practices. These regulations dictate
who can work, the minimum wage, the maximum hours of work,
the nature of working conditions, what the employer can ask the
worker to do, and how easily the worker can be fired. They guar-
antee the worker’s freedom to form unions to represent his or her
interests and set the rules under which collective bargaining can
take place over pay and benefits.
From a classical liberal standpoint, most of these regulations
make little sense. They interfere with an individual’s right to enter
into contracts of his or her choosing. If you are willing to work for
70 hours a week below the minimum wage under unsafe condi-
tions and allow the employer to dismiss you at will, why should
the state prevent you from accepting such terms? Similarly, if you
think it is a good thing for your fourteen-year-old daughter to get
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
T he Polit ic al Tr ilemma of the Wor ld Economy 1 9 1
a full-time job in a factory, why should the government tell you
otherwise? According to classical liberal doctrine, people are the
best judge of their own interests (and the interests of their fam-
ily members), and voluntary contracts, entered freely, must leave
both parties better off.
Labor markets were once governed by this doctrine.5 Since the
1930s, however, U.S. legislation and the courts have recognized
that what may be good for an individual worker may not be good
for workers as a whole. Without regulations that enforce societal
norms of decent work, a prospective employee with little bargain-
ing power may be forced to accept conditions that violate those
norms. By accepting such a contract, the employee also makes it
harder for other workers to achieve higher labor standards. Thus
employers must be prohibited from offering odious contracts even
if some workers are willing to accept them. Certain forms of com-
petition have to be ruled out. You may be willing to work for 70
hours a week below the minimum wage. But my employer cannot
take advantage of your willingness to work under these conditions
and offer my job to you.
Consider how international trade affects this understanding.
Thanks to outsourcing, my employer can now do what he previ-
ously could not. Domestic labor laws still prohibit him from hiring
you in my place and putting me to work under conditions that
violate those laws. But this no longer matters. He can now replace
me with a worker in Indonesia or Guatemala who will work will-
ingly under those same substandard conditions or worse. To econ-
omists, this is not just legal; it is a manifestation of the gains from
trade. Yet the consequences for me and my job do not depend on
the citizenship of the worker bidding down my labor standards.
Why do national regulations protect me from downward competi-
tion in employment practices from a domestic worker but not a
foreign one? Why should we allow international markets to erode
domestic labor regulations through the back door when we do not
allow domestic markets to do the same?
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
1 9 2 T he Globaliz at ion Par adox
The inconsistency is further highlighted by considering whether
a society would condone allowing those Indonesian and Guate-
malans to be employed at home as guest workers under the same
labor standards they face in their native countries. Even most free
traders would object to such a practice. There should be a single
set of labor standards in a country, they will say, applied to all
workers regardless of the passport they carry. But why? Outsourc-
ing jobs through trade has exactly the same consequences, for all
concerned, as allowing migrant workers to toil under a lower set
of standards.
How significant are these issues in the real world? Less than
many labor advocates claim, but more than free traders are willing
to admit. Wage levels are determined first and foremost by labor
productivity. Differences in productivity account for between 80 to
90 percent of the variation in wages around the world. This puts a
significant damper on the potential of outsourcing to undermine
employment practices in the advanced countries. An employer’s
threat to outsource my job to someone who earns half my wage
does not pose much danger to me when that foreign worker also
has half my productivity.
But 80 to 90 percent is not 100 percent. The political and social
institutions that frame labor markets exert some independent
influence on labor earnings, quite separate from the powerful
effects of productivity. Labor regulations, unionization levels,
and more broadly the political rights exercised by workers shape
the bargains between workers and their employers and determine
how the economic value created by firms is shared between them.
These arrangements can move wage levels up or down in any
country by 40 percent or more.6 It is here that outsourcing, or
the threat thereof, can play a role. Moving jobs to where workers
enjoy fewer rights—or threatening to do so—can be beneficial to
employers. Within limits, it can be used as a lever for extracting
concessions on wages and employment practices from domestic
workers.
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
T he Polit ic al Tr ilemma of the Wor ld Economy 1 9 3
There aren’t easy solutions to these conundrums. An employer’s
freedom to choose where he wants to operate is a competing value
that surely deserves attention. The interests of the Guatemalan
or Indonesian workers may collide with the interests of domestic
workers. We cannot however pretend that outsourcing does not
create serious difficulties for domestic labor standards.
Corporate tax competition. The international mobility of firms and
of capital also restricts a nation’s ability to choose the tax struc-
ture that best reflects its needs and preferences. In particular, this
mobility puts downward pressure on corporate tax rates and shifts
the tax burden from capital, which is internationally mobile, to
labor, which is much less so.
The logic is obvious and figures regularly in the arguments of
those who push for lower taxes on business. Senator John McCain
invoked it prominently in his pre-election debate with Barack
Obama when he compared America’s corporate tax rate of 35
percent to Ireland’s 11 percent. “Now, if you’re a business person,
and you can locate any place in the world,” McCain noted, then
obviously “you go to the country where it’s 11 percent tax versus
35 percent.”7 McCain got his number for Ireland wrong: the Irish
corporate tax rate is 12.5 percent, not 11 percent; but note that he
accepted (and cherished) the constraint imposed by globalization.
It enabled him to fortify his argument for lower taxes by appealing
to their inevitability, courtesy of globalization.
There has been a remarkable reduction in corporate taxes
around the world since the early 1980s. The average for the mem-
ber countries of the OECD countries, excluding the United States,
has fallen from around 50 percent in 1981 to 30 percent in 2009.
In the United States, the statutory tax on capital has come down
from 50 percent to 39 percent over the same period.8 Competition
among governments for increasingly mobile global firms—what
economists call “international tax competition”—has played a role
in this global shift. The arguments of McCain and countless other
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
1 9 4 T he Globaliz at ion Par adox
conservative politicians who have used globalization to advance
their agendas provide still more evidence of this role.
A detailed economic study on OECD tax policies finds that when
other countries reduce their average statutory corporate tax rate
by 1 percentage point the home country follows by reducing its tax
rate by 0.7 percentage points. You either stand your ground and
risk seeing your corporations depart for lower tax jurisdictions,
or you respond in kind. Interestingly, the same study finds that
international tax competition takes place only among countries
that have removed their capital controls. When such controls are
in place, capital and profits cannot move as easily across national
borders and there is no downward pressure on capital taxes. The
removal of capital controls appears to be the main factor driving
the reduction in corporate tax rates since the 1980s.9
The problem has become a big enough headache for tax agen-
cies that efforts are under way within the OECD and European
Union to identify and roll back instances of so-called “harmful
tax competition.” To date, these activities have only focused on
tax havens in a number of microstates ranging from Andorra to
Vanuatu. The real challenge is to safeguard the integrity of each
nation’s corporate tax regime in a world where enterprises and
their capital are footloose. This challenge remains unaddressed.
Health and safety standards. Most people would subscribe to the
principle that nations ought to be free to determine their own
standards with respect to public health and safety. What happens
when these standards diverge across countries, either by design
or because of differences in their application? How should goods
and services be treated when they cross the boundaries of jurisdic-
tions with varying standards?
WTO jurisprudence on this question continues to evolve. The
WTO allows countries to enact regulations on public health and
safety grounds that may run against their general obligations
under the trade rules. But these regulations need to be applied
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
T he Polit ic al Tr ilemma of the Wor ld Economy 1 9 5
in a way that does not overtly discriminate against imports and
must not smack of disguised protectionism. The WTO’s Agree-
ment on Sanitary and Phytosanitary (SPS) Measures recognizes
the right of nations to apply measures that protect human, ani-
mal, or plant life or health, but these measures must conform to
international standards or be based on “scientific principles.” In
practice, disputes in these areas hang on the interpretation of a
group of judges in Geneva about what is reasonable or practical.
In the absence of bright lines that demarcate national sovereignty
from international obligations, the judges often claim too much
on behalf of the trade regime.
In 1990, for example, a GATT panel ruled against Thailand’s
ban on imported cigarettes. Thailand had imposed the ban as
part of a campaign to reduce smoking, but continued to allow
the sale of domestic cigarettes. The Thai government argued that
imported cigarettes were more addictive and were more likely to
be consumed by young people and by women on account of their
effective advertising. The GATT panel was unmoved. It reasoned
that the Thai government could have attained its public health
objectives at less cost to trade by pursuing alternative policies.
The government might have resorted to restrictions on advertis-
ing, labeling requirements, or content requirements, all of which
could be applied in a non-discriminatory manner.
The GATT panel was surely correct about the impact of the
Thai ban on trade. But in reaching their decisions, the panelists
second-guessed the government about what is feasible and practi-
cal. As the legal scholars Michael Trebilcock and Robert Howse
put it, “the Panel simply ignored the possibility that the alternative
measures might involve high regulatory and compliance costs, or
might be impracticable to implement effectively in a developing
country.”10
The hormone beef case from chapter Four also raises difficult
issues. In this instance, the European Union ban on beef reared
on certain growth hormones was not discriminatory; it applied to
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
1 9 6 T he Globaliz at ion Par adox
imported and domestic beef alike. It was also obvious that there
was no protectionist motive behind the ban, which was pushed by
consumer lobbies and interests in Europe alarmed by the potential
health threats. Nonetheless, the WTO Panel and appellate body
both ruled against the European Union, arguing that the ban vio-
lated the requirement in the SPS Agreement that policies be based
on “scientific evidence.” There was indeed scant positive evidence
to date that growth hormones posed any health threats. Instead,
the European Union had applied a broader principle not explicitly
covered by the WTO, the “precautionary principle,” which permits
greater caution in the presence of scientific uncertainty.11
The precautionary principle reverses the burden of proof.
Instead of asking, “Is there reasonable evidence that growth hor-
mones or GMOs have adverse effects?” it requires policy mak-
ers to ask, “Are we reasonably sure that they do not ?” In many
unsettled areas of scientific knowledge, the answer to both ques-
tions can be no. The precautionary principle makes sense in cases
where adverse effects can be large and irreversible. As the Euro-
pean Commission argued (unsuccessfully), policy here cannot be
made purely on the basis of science. Politics, which aggregates a
society’s risk preferences, must play the determinative role. The
WTO judges did acknowledge a nation’s right to apply its own risk
standards, but ruled that the European Union’s invocation of the
precautionary principle did not satisfy the criterion of “scientific
evidence.” Instead of simply ascertaining whether the science was
taken into account, the rules of the SPS Agreement forced them
to use an international standard on how scientific evidence should
be processed.
If the European Union, with its sophisticated policy machinery,
could not convince the WTO that it should have leeway in deter-
mining its own standards, we can only imagine the difficulties that
developing nations face. For poor nations, even more than rich
ones, the rules imply a single standard.
Ultimately, the question is whether a democracy is allowed to
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
T he Polit ic al Tr ilemma of the Wor ld Economy 1 9 7
determine its own rules—and make its own mistakes. The Euro-
pean Union regulations on beef (and, in a similar case in 2006, on
biotech) did not discriminate against imports, which makes inter-
national discipline designed to promote trade even more prob-
lematic. As I will argue later, international rules can and should
require certain procedural safeguards for domestic regulatory
proceedings (such as transparency, broad representation, and sci-
entific input) in accord with democratic practices. The trouble
occurs when international tribunals contradict domestic proceed-
ings on substantive matters (in the beef case, how to trade off eco-
nomic benefits against uncertain health risks). In this instance,
trade rules clearly trumped democratic decision making within
the European Union.
“Regulatory takings.” There are thousands of bilateral investment
treaties (BITs) and hundreds of bilateral or regional trade agree-
ments (RTAs) currently in force. Governments use them to pro-
mote trade and investment links in ways that go beyond what the
WTO and other multilateral arrangements permit. A key objec-
tive is to provide a higher level of security to foreign investors by
undertaking stronger external commitments.
BITs and RTAs usually allow foreign investors to sue host gov-
ernments in an international tribunal for damages when new
domestic regulations have adverse effects on the investors’ profits.
The idea is that the change in government regulations amounts to
expropriation (it reduces the benefits that were initially granted
to the investors under the BIT or RTA), and therefore requires
compensation. This is similar to the U.S. doctrine of “regulatory
takings,” which however has never been accepted legal practice
within the United States. The treaties include a general excep-
tion to allow governments to pursue policies in the interests of
the public good, but since these cases are judged in international
courts, different standards can apply. Foreign investors may end
up receiving rights that domestic investors do not have.12
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
1 9 8 T he Globaliz at ion Par adox
Such cases have been prominent under the North American
Free Trade Agreement or NAFTA of 1992, particularly in the area
of environmental regulation. Foreign investors have won dam-
ages against the Canadian and Mexican governments in several
instances. In 1997, a U.S. firm challenged a Mexican municipal-
ity’s refusal to grant a construction permit for a toxic waste facility
and was awarded $15.6 million in damages. The same year, a U.S.
chemical company challenged a Canadian ban on a gasoline addi-
tive and received $13 million in a settlement.13
Perhaps the most worrying case to date involves a suit brought
against the South African government in 2007 by three Italian
mining companies. The companies charge that South Africa’s
affirmative action program, called Black Economic Empower-
ment, violates the rights provided to them under existing bilateral
investment treaties. The program aims to reverse South Africa’s
long history of racial discrimination and is an integral element in
the country’s democratic transition. It requires that mining com-
panies alter their employment practices and sell a minority share
to black partners. The Italian companies have asked for $350 mil-
lion in return for what they assert is an expropriation of their
South African operations.14 If they win, they will have achieved an
outcome beyond the reach of any domestic investor.
Industrial policies in developing nations. Probably the most signifi-
cant external constraint that developing nations face as a conse-
quence of hyperglobalization are the restrictions on industrial
policies that make it harder for countries in Latin America, Africa,
and elsewhere to emulate the development strategies that East
Asian countries have employed to such good effect.
Unlike GATT, which left poor nations essentially free to use any
and all industrial policies, the WTO imposes several restrictions.
Export subsidies are now illegal for all but the poorest nations,
denying developing nations the benefit of export-processing zones
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
T he Polit ic al Tr ilemma of the Wor ld Economy 1 9 9
of the type that Mauritius, China, and many Southeast Asian
nations have used.15 Policies that require firms to use more local
inputs (so-called “domestic content requirements”) are also ille-
gal, even though such policies helped China and India develop
into world-class auto parts suppliers. Patent and copyright laws
must now comply with minimum international standards, rul-
ing out the kind of industrial imitation that was crucial to both
South Korea and Taiwan’s industrial strategies during the 1960s
and 1970s (and indeed to many of today’s rich countries in earlier
periods).16 Countries that are not members of the WTO are often
hit with more restrictive demands as part of their negotiations to
join the organization.
The WTO’s Agreement on Intellectual Property Rights (TRIPS)
deserves special mention. This agreement significantly impairs
the ability of developing nations to reverse-engineer and copy the
advanced technologies used in rich countries. As the Columbia
economist and expert on technology policy Richard Nelson notes,
copying foreign technology has long been one of the most impor-
tant drivers of economic catch-up.17 TRIPS has raised considerable
concern because it restricts access to essential medicines and has
adverse effects on public health. Its detrimental effects on techno-
logical capabilities in developing nations have yet to receive simi-
lar attention, though they may be of equal significance.
Regional or bilateral trade agreements typically extend the
external constraints beyond those found in the WTO. These
agreements are in effect a means for the United States and the
European Union to “export their own regulatory approaches” to
developing nations.18 Often they encompass measures which the
United States and the European Union have tried to get adopted
in the WTO or other multilateral forums, but have failed. In par-
ticular in its free trade agreements with developing countries, the
United States aggressively pushes for restrictions on their govern-
ments’ ability to manage capital flows and shape patent regula-
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
2 0 0 T he Globaliz at ion Par adox
tions. And even though the IMF now exercises greater restraint, its
programs with individual developing countries still contain many
detailed requirements on trade and industrial policies.19
Developing nations have not completely run out of room to pur-
sue industrial strategies that promote new industries. Determined
governments can get around many of these restrictions, but few
governments in the developing world are not constantly asking
themselves if this or that proposed policy is WTO-legal.
The Trilemma
How do we manage the tension between national democracy and
global markets? We have three options. We can restrict democracy
in the interest of minimizing international transaction costs, dis-
regarding the economic and social whiplash that the global econ-
omy occasionally produces. We can limit globalization, in the hope
of building democratic legitimacy at home. Or we can globalize
democracy, at the cost of national sovereignty. This gives us a menu
of options for reconstructing the world economy.
The menu captures the fundamental political trilemma of the
world economy: we cannot have hyperglobalization, democracy,
and national self-determination all at once. We can have at most
two out of three. If we want hyperglobalization and democracy,
we need to give up on the nation state. If we must keep the nation
state and want hyperglobalization too, then we must forget about
democracy. And if we want to combine democracy with the nation
state, then it is bye-bye deep globalization. The figure below depicts
these choices.
Why these stark trade-offs? Consider a hypothetical fully global-
ized world economy in which all transaction costs have been elimi-
nated and national borders do not interfere with the exchange of
goods, services, or capital. Can nation states exist in such a world?
Only if they focus exclusively on economic globalization and on
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
T he Polit ic al Tr ilemma of the Wor ld Economy 2 0 1
becoming attractive to international investors and traders. Domes-
tic regulations and tax policies would then be either brought into
alignment with international standards, or structured so that they
pose the least amount of hindrance to international economic
integration. The only services provided by governments would
be those that reinforce the smooth functioning of international
markets.
We can envisage a world of this sort, and it is the one Tom Fried-
man had in mind when he coined the term “Golden Straitjacket.”
In this world, governments pursue policies that they believe will
earn them market confidence and attract trade and capital inflows:
tight money, small government, low taxes, flexible labor markets,
deregulation, privatization, and openness all around. “Golden
Straitjacket” evokes the era of the gold standard before World War
I. Unencumbered by domestic economic and social obligations,
national governments were then free to pursue an agenda that
focused exclusively on strict monetary rules.
Figure 9-1: Pick two, any two
Hyperglobalization
Nation state Democratic politics
Golden
Straitjacket
Bretton Woods compromise
Global
Governance
The Political Trilemma of the World Economy
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
2 0 2 T he Globaliz at ion Par adox
External restraints were even more blatant under mercantilism
and imperialism. We cannot properly speak of nation states before
the nineteenth century, but the global economic system operated
along strict Golden Straitjacket lines. The rules of the game—
open borders, protection of the rights of foreign merchants and
investors—were enforced by chartered trading companies or impe-
rial powers. There was no possibility of deviating from them.
We may be far from the classical gold standard or chartered
trading companies today, but the demands of hyperglobalization
require a similar crowding out of domestic politics. The signs are
familiar: the insulation of economic policy-making bodies (central
banks, fiscal authorities, regulators, and so on), the disappearance
(or privatization) of social insurance, the push for low corporate
taxes, the erosion of the social compact between business and
labor, and the replacement of domestic developmental goals with
the need to maintain market confidence. Once the rules of the
game are dictated by the requirements of the global economy,
domestic groups’ access to, and their control over, national eco-
nomic policy making must inevitably become restricted. You can
have your globalization and your nation state too, but only if you
keep democracy at bay.
Must we give up on democracy if we want to strive for a fully
globalized world economy? There is actually a way out. We can
drop nation states rather than democratic politics. This is the
“global governance” option. Robust global institutions with regu-
latory and standard-setting powers would align legal and political
jurisdictions with the reach of markets and remove the transaction
costs associated with national borders. If they could be endowed
with adequate accountability and legitimacy in addition, politics
need not, and would not, shrink: it would relocate to the global
level.
Taking this idea to its logical conclusion, we can envisage a
form of global federalism—the U.S. model expanded on a global
scale. Within the United States a national constitution, federal
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
T he Polit ic al Tr ilemma of the Wor ld Economy 2 0 3
government, federal judiciary, and large number of nationwide
regulatory agencies ensure that markets are truly national despite
many differences in regulatory and taxation practices among indi-
vidual states. Or we can imagine alternative forms of global gover-
nance, not as ambitious as global federalism and built around new
mechanisms of accountability and representation. A major move
in the direction of global governance, in whatever form, necessar-
ily would entail a significant diminution of national sovereignty.
National governments would not disappear, but their powers
would be severely circumscribed by supranational rulemaking and
enforcing bodies empowered (and constrained) by democratic
legitimacy. The European Union is a regional example of this.
This may sound like pie in the sky, and perhaps it is. The histori-
cal experience of the United States shows how tricky it can be to
establish and maintain a political union in the face of large differ-
ences in the constituent parts. The halting way in which political
institutions within the European Union have developed, and the
persistent complaints about their democratic deficit, also indicate
the difficulties involved—even when the union comprises a group
of nations at similar income levels and with similar historical tra-
jectories. Real federalism on a global scale is at best a century
away.
The appeal of the global governance model, however wishful,
cannot be denied. When I present my students with the trilemma
and ask them to pick one of the options, this one wins hands-
down. If we can simultaneously reap the benefits of globalization
and democracy, who cares that national politicians will be out of
a job? Yes, there are practical difficulties with democratic global
governance, but perhaps these are exaggerated, too. Many politi-
cal theorists and legal scholars suggest that democratic global gov-
ernance can grow out of today’s international networks of policy
makers, as long as these are held in check by new mechanisms of
accountability of the type we shall consider in the next chapter.
I am skeptical about the global governance option, but mostly
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
2 0 4 T he Globaliz at ion Par adox
on substantive rather than practical grounds. There is simply too
much diversity in the world for nations to be shoehorned into
common rules, even if these rules are somehow the product of
democratic processes. Global standards and regulations are not
just impractical; they are undesirable. The democratic legitimacy
constraint virtually ensures that global governance will result in
the lowest common denominator, a regime of weak and ineffective
rules. We then face the big risk of too little governance all around,
with national governments giving up on their responsibilities and
no one else picking up the slack. But more on this in the next
chapter.
The only remaining option sacrifices hyperglobalization. The
Bretton Woods regime did this, which is why I have called it the
Bretton Woods compromise. The Bretton Woods–GATT regime
allowed countries to dance to their own tune as long as they
removed a number of border restrictions on trade and gener-
ally treated all their trade partners equally. They were allowed
(indeed encouraged) to maintain restrictions on capital flows, as
the architects of the postwar economic order did not believe that
free capital flows were compatible with domestic economic stabil-
ity. Developing country policies were effectively left outside the
scope of international discipline.
Until the 1980s, these loose rules left space for countries to fol-
low their own, possibly divergent paths of development. Western
Europe chose to integrate as a region and to erect an extensive wel-
fare state. As we have seen, Japan caught up with the West using its
own distinctive brand of capitalism, combining a dynamic export
machine with large doses of inefficiency in services and agricul-
ture. China grew by leaps and bounds once it recognized the
importance of private initiative, even though it flouted every other
rule in the guidebook. Much of the rest of East Asia generated an
economic miracle by relying on industrial policies that have since
been banned by the WTO. Scores of countries in Latin America,
the Middle East, and Africa generated unprecedented economic
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
T he Polit ic al Tr ilemma of the Wor ld Economy 2 0 5
growth rates until the late 1970s under import-substitution poli-
cies that insulated their economies from the world economy. As
we saw, the Bretton Woods compromise was largely abandoned in
the 1980s as the liberalization of capital flows gathered speed and
trade agreements began to reach behind national borders.
The world economy has since been trapped in an uncomfort-
able zone between the three nodes of the trilemma. We have not
squarely faced up to the tough choices that the trilemma identi-
fies. In particular, we have yet to accept openly that we need to
lower our sights on economic globalization if we want the nation
state to remain the principal locus of democratic politics. We have
no choice but to settle for a “thin” version of globalization—to
reinvent the Bretton Woods compromise for a different era.
We cannot simply bring back wholesale the approaches of the
1950s and 1960s. We will have to be imaginative, innovative, and
willing to experiment. In the last part of the book, I will provide
some ideas on how to move forward. But the first order of business
is getting the big picture right. The necessary sort of policy experi-
mentation will not be unleashed until we change our narrative.
Smart Globalization Can Enhance National Democracy
Each of the cases I discussed previously embodies a trade-off
between removing transaction costs in the international economy
and maintaining domestic differences. The greater the emphasis
on deep economic integration, the less the room for national dif-
ferences in social and economic arrangements, and the smaller
the space for democratic decision making at the national level.
More restrained forms of globalization need not embrace the
assumptions inherent in deep integration. By placing limits on glo-
balization, the Bretton Woods regime allowed the world economy
and national democracies to flourish side by side. Once we accept
restraints on globalization, we can in fact go one step further. We
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
2 0 6 T he Globaliz at ion Par adox
can envisage global rules that actually enhance the operation of
national democracies.
There is indeed nothing inherently contradictory between
having a global rule–based regime and national democracy.
Democracy is never perfect in practice. As the Princeton politi-
cal scientists Robert Keohane, Stephen Macedo, and Andrew
Moravcsik have argued, well-crafted external rules may enhance
both the quality and legitimacy of democratic practices. Democra-
cies, these authors note, do not aim simply to maximize popular
participation. Even when external rules constrain participation
at the national level, they may provide compensating democratic
benefits such as improving deliberation, suppressing factions,
and ensuring minority representation. Democratic practices can
be enhanced by procedural safeguards that prevent capture by
interest groups and ensure the use of relevant economic and sci-
entific evidence as part of the deliberations. Besides, entering into
binding international commitments is a sovereign act. Restricting
it would be like preventing Congress from delegating some of its
rulemaking powers to independent regulatory agencies.20
While international commitments can enhance national
democracy, they will not necessarily do so. The hyperglobaliza-
tion agenda, with its focus on minimizing transaction costs in the
international economy, clashes with democracy for the simple
reason that it seeks not to improve the functioning of democracy
but to accommodate commercial and financial interests seeking
market access at low cost. It requires us to buy into a narrative that
gives predominance to the needs of multinational enterprises,
big banks, and investment houses over other social and economic
objectives.21 Hence this agenda serves primarily those needs.
We have a choice in how we overcome this defect. We can global-
ize democratic governance along with markets; or we can rethink
trade and investment agreements to expand space for democratic
decision making at the national level. I discuss each of these strat-
egies in turn in the following two chapters.
Rodrik, Dani. The Globalization Paradox:Why Global Markets, States, and Democracy Can’t Coexist : Why
Global Markets, States, and Democracy Can’t Coexist, Oxford University Press USA – OSO, 2011. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/asulib-ebooks/detail.action?docID=990527.
Created from asulib-ebooks on 2021-03-15 11:08:06.
C
op
yr
ig
ht
©
2
01
1.
O
xf
or
d
U
ni
ve
rs
ity
P
re
ss
U
S
A
–
O
S
O
. A
ll
rig
ht
s
re
se
rv
ed
.
Globalisation is dead and we need to invent a
new world order
A book excerpt and interview with Michael O’Sullivan, author of “The Levelling”
HAS THE world witnessed “peak democracy”? Is the future one in which open societies
with free markets vie for influence in global affairs with authoritarian countries under state
capitalism? The very questions evoke a nostalgia for a seemingly simpler past. For Michael
O’Sullivan, formerly an investment banker and economist at Princeton University, it is
more useful to consider the future.
Mr O’Sullivan’s book, “The Levelling: What’s Next After Globalisation” offers a roadmap.
He sees a multipolar world forming but international institutions unprepared for this. He
voices worries about a world of low growth and high debt—and calls for a “world treaty on
risk” so central banks only resort to measures like quantitative easing under agreed
conditions.
But his most intriguing framing of the issues is his comparison of today’s world and 17th
century England’s Putney Debates, when the practicalities of a rights-based democracy were
first enunciated by a faction called “ The Levellers” (which inspired the book’s title). The
world, he believes, will cleave into “Leveller” countries that hew to rights and freedoms,
and “Leviathan” ones that are content with state-managed growth and fewer liberties.
As part of The Economist’s Open Future initiative, we probed Mr O’Sullivan ideas in a short
interview. Below that is an excerpt from his book, on the end of globalisation.
* * *
The Economist: Describe what comes after globalisation—what does the world you foresee
look like?
Mr O’Sullivan: Globalisation is already behind us. We should say goodbye to it and set our
minds on the emerging multipolar world. This will be dominated by at least three large
regions: America, the European Union and a China-centric Asia. They will increasingly take
very different approaches to economic policy, liberty, warfare, technology and society. Mid-
sized countries like Russia, Britain, Australia and Japan will struggle to find their place in
the world, while new coalitions will emerge, such as a “Hanseatic League 2.0” of small,
advanced states like those of Scandinavia and the Baltics. Institutions of the 20th century—
the World Bank, the International Monetary Fund and the World Trade Organisation—will
appear increasingly defunct.
The Economist: What killed globalisation?
Michael O’Sullivan: At least two things have put paid to globalisation. First, global economic
growth has slowed, and as a result, the growth has become more “financialised”: debt has
increased and there has been more “monetary activism”—that is, central banks pumping
money into the economy by buying assets, such as bonds and in some cases even equities—
to sustain the international expansion. Second, the side effects, or rather the perceived
side-effects, of globalisation are more apparent: wealth inequality, the dominance of
multinationals and the dispersion of global supply chains, which have all become hot
political issues.
The Economist: Was the death of globalisation inevitable or could (and should) it have
been prevented?
Mr O’Sullivan: One problematic factor here is that there is no central body or authority to
shape globalisation, beyond perhaps the World Economic Forum or maybe the
Organisation for Economic Co-operation and Development. In many ways, the end of
globalisation is marked by the poor and inconclusive response to the global financial crisis.
In general, the response has been to cut the cost of capital and not to tackle the root causes
of the crisis. As such, the world economy will limp on, burdened by debt and in hock to
easy money from central banks.
The Economist: The book’s title comes from the “Levellers” during Britain’s Putney
Debates in the mid 1600s. Who were they and what can their story teach us today?
Mr O’Sullivan: The Levellers are a hidden gem from British history. They were a mid 17th
century group in England, who participated in debates about democracy that took place in a
part of London called Putney. Their achievement was crafting “An Agreement of the
People,” which were a series of manifestos that marked the first popular conceptions of
what a constitutional democracy might look like.
The Levellers are interesting for two reasons. First, in the context of the time, their approach
was constructive and practical. The “Agreement” spells out what people want from those
who govern them in a clear and tangible way. For example, they proposed term limits on
political office and that laws regarding indebtedness are applied equally to the rich and poor.
Second, they are interesting for the way the movement was countermanded and then
snuffed out by the military leader Oliver Cromwell and the Grandees (the elites of their
day). Like so many idealistic political start-ups, the Levellers failed. This should encourage
the growing number of new political parties, like Change UK and new candidates to be
worldy-wise in how they approach the process of political reform and change.
The Economist: You foresee new international institutions to replace archaic 20th-century
ones that are suited for a different time. How would they work? And can countries with
such different values (ie, democratic, market-based “Levellers” and state-managed societies
and economies, the “Leviathans”) really cooperate?
Mr O’Sullivan: A lot is made of the Cold War rivalry between communist Russia and
America, and now some want to see a clash of civilizations between America and China.
The “Levelling” characterises a future where there are at least two approaches to public life.
The most distinctive approach to nations doing things in their own way will be for what the
Levellers might call the “rights of freeborn men,” or the idea of the open society. The code
of the Levellers presents a very clear political formula that Europeans and Americans will
recognise for its values, though decreasingly in its practice.
The challenge to this code will come from the rising acceptance of less democratic ways of
ordering society, in both developed and developing countries. A related clash will be the
desire of a growing proportion of electorates to have a more open society as economies also
open up.
As the world evolves along the lines of Leveller-type and Leviathan-type societies, it is
possible that in some countries, such as Russia, a Leviathan-like approach—that is, order in
exchange for reduced democracy and rights—will be the accepted way of life. In other
countries, most interestingly China, as its economy loses momentum and evolves, there
may be a growing tension between groups holding the Leviathan view (supported inevitably
by Grandees) and opposing Leveller-like groups (who favor equality of opportunity and a
multiparty system). The role and views of women, especially in China, and of minority
groups like the gay community will be pivotal.
The emergence of a new world order, based on large regions and coloured by Leveller and
Leviathan modes of governance, echoes several periods in history. The challenge in the next
few years will be for Leviathan-oriented nations like China to maintain economic stability
so that rising unemployment, for instance, does not break the “Leviathan contract”.
Equally, the challenge in Leveller countries will be to maintain open, fraternal societies in
the face of political and potentially economic volatility.
* * *
Good-bye to Globalisation
Excerpted from “The Levelling: What’s Next After Globalization” by Michael
O’Sullivan (PublicAffairs, 2019).
It may well be better that those who have grown fond of globalization get over it, accept its
passing, and begin to adjust to a new reality. Many will resist and, like the thirty-five
foreign-policy experts who published an advertisement in the New York Times on July 26,
2018, under the banner “Why We Should Preserve International Institutions and Order,”
will feel that the existing world order and its institutions should be maintained. I disagree.
Globalization, at least in the form that people have come to enjoy it, is defunct. From here,
the passage away from globalization can take two new forms. One dangerous scenario is
that we witness the outright end of globalization in much the same manner as the first
period of globalization collapsed in 1913. This scenario is a favorite of commentators
because it allows them to write about bloody end-of-the-world calamities. This is,
thankfully, a low-probability outcome, and with apologies to the many armchair admirals in
the commentariat who, for instance, talk willfully of a conflict in the South China Sea, I
suggest that a full-scale sea battle between China and the United States is unlikely.
Instead, the evolution of a new world order—a fully multipolar world composed of three
(perhaps four, depending on how India develops) large regions that are distinct in the
workings of their economies, laws, cultures, and security networks—is manifestly
underway. My sense is that until 2018, multipolarity was a more theoretical concept—more
something to write about than to witness. This is changing quickly: trade tensions, advances
in technologies (such as quantum computing), and the regulation of technology are just
some of the fissures around which the world is splitting into distinct regions. Multipolarity is
gaining traction and will have two broad axes. First, the poles in the multipolar world have
to be large in terms of economic, financial, and geopolitical power. Second, the essence of
multipolarity is not simply that the poles are large and powerful but also that they develop
distinct, culturally consistent ways of doing things. Multipolarity, where regions do things
distinctly and differently, is also very different from multilateralism, where they do them
together.
China, in particular, is interesting in the context of the switch from globalization to
multipolarity, not least because at the 2017 World Economic Forum the Chinese president
claimed the mantle of globalization for China. China benefited greatly from globalization
and its accoutrements (e.g., WTO membership), and it played a vital role in the supply-
chain dynamic that drove globalization. However, trade flows into China increasingly
betray a move away from a globalized world and toward a more regionally focused one. For
instance, IMF data show that in 2018, compared with 2011, Cambodia, Vietnam, Laos, and
Malaysia traded more with China and relatively less with the United States. These
countries, together with Bangladesh and Pakistan, have allowed themselves to be enticed by
trade- and investment-based relationships with China and are now in its orbit.
However, China is itself not globalized: it is increasingly hard for Western companies to do
business there on equal terms with Chinese companies, and the flow of both money and
ideas—out of and into China, respectively—is heavily curtailed. Flow of people is another
indicator. Flows within China are dynamic and are perhaps more managed than before, but
flows of foreigners into China are miniscule by comparison to other countries, and China
has only recently established an agency (the State Immigration Administration created at
the 2018 Party Congress) to cultivate inward flows. So as China has become a major pole, it
has become less globalized and arguably is contributing to the trend toward deglobalization.
On a broader scale, without picking on individual countries, we can measure the extent to
which the world is becoming multipolar by examining aggregate trends in trade, GDP,
foreign direct investment, government budget size, and population. All of these are much
less concentrated, or more dispersed, than they used to be, and increasingly they are
collecting around several poles. For example, in the five years from 2012 to 2017, total
foreign direct investment into Australia from China increased at a rate of 21 percent per
annum, compared to 6 percent from the United States to Australia, suggesting that Asian
investment in Australia is picking up.
[…]
Even if multipolarity is based on the growing dispersion and regionalization of economic
power, it is also expressed in other ways, notably military power, political and
cyberfreedoms, technological sophistication, financial sector growth, and a greater sense of
cultural prerogative and confidence. These are not as easily measured as economic
multipolarity, but some clear strands are emerging. To try to synthesize what a pole entails,
we can point toward several initial factors: size of a country’s GDP, size of its population,
the existence of an imperial legacy, the extent of its regional economic role, its military size
and sophistication (e.g., absolute spending, number of fighter jets and ships), its place on the
UN Human Development Index relative to its region, and its participation (or not) in a
regional grouping (such as NATO or the European Union).
Under this schema the European Union, the United States, China, and potentially India are
poles, but Japan and Russia would not qualify as distinct poles. Russia, for instance, scores
well on certain aspects of multipolarity (e.g., militarily), but in its current state it may never
become a true pole in the sense employed here.
[…]
The path toward multipolarity will not be smooth. One tension is that since the Industrial
Revolution the world has had an anchor point in terms of the locus and spread of
globalization (Britain in the nineteenth century and the United States in the twentieth
century). The fact that there are now at least three points of reference introduces a new and
possibly uncertain dynamic to world affairs.
The potential is high for friction, misunderstanding, and conflict among the increasingly
different ways of doing things across the major poles. Essentially, multipolarity means that
instead of speaking a common language, the major poles speak different policy languages.
Trade-based tension is an obvious possibility here. Another form of tension is the crisis of
identity created for countries that are not wholly within one of the poles—again, Japan,
Australia, and the United Kingdom are the prime examples—and the crisis of ambition for
countries, such as Russia, that want to be poles but lack the wherewithal to do so
convincingly. At a more grassroots level, the implications of the end of globalization as we
know it and the path to multipolarity will become a greater part of the political debate. At
the margin, the flow of people, ideas, and capital may be less global and more regional and
in time could be reinforced by a growing sense of regionalization across the main poles. In a
negative way, a more multipolar world may be the watershed that signals the peak of
democracy and potentially the beginning of contests within regions for competing views of
democracy, institutional strength, statecraft, and control.
_______________
Excerpted from “The Levelling: What’s Next After Globalization.” Copyright © 2019 by
Michael O’Sullivan. Used with permission of PublicAffairs (Hachette Book Group). All
rights reserved.
Broadening participation Boosting productivity
and GDP
Changing the way
companies go global
857343
HIGHLIGHTS
MARCH 2016
DIGITAL GLOBALIZATION:
THE NEW ERA OF GLOBAL FLOWS
Copyright © McKinsey & Company 2016
In the 25 years since its founding, the McKinsey Global Institute (MGI) has sought to develop
a deeper understanding of the evolving global economy. As the business and economics
research arm of McKinsey & Company, MGI aims to provide leaders in the commercial,
public, and social sectors with the facts and insights on which to base management and
policy decisions. We are proud to be ranked the top private-sector think tank, according
to the authoritative 2015 Global Go To Think Tank Index, an annual report issued by the
University of Pennsylvania Think Tanks and Civil Societies Program at the Lauder Institute.
MGI research combines the disciplines of economics and management, employing the
analytical tools of economics with the insights of business leaders. Our “micro-to-macro”
methodology examines microeconomic industry trends to better understand the broad
macroeconomic forces affecting business strategy and public policy. MGI’s in-depth reports
have covered more than 20 countries and 30 industries. Current research focuses on six
themes: productivity and growth, natural resources, labor markets, the evolution of global
financial markets, the economic impact of technology and innovation, and urbanization.
Recent reports have assessed global flows; the economies of Brazil, Mexico, Nigeria, and
Japan; China’s digital transformation; India’s path from poverty to empowerment; affordable
housing; the effects of global debt; and the economics of tackling obesity.
MGI is led by three McKinsey & Company directors: Richard Dobbs, James Manyika,
and Jonathan Woetzel. Michael Chui, Susan Lund, Anu Madgavkar, and Jaana Remes
serve as MGI partners. Project teams are led by the MGI partners and a group of senior
fellows, and include consultants from McKinsey & Company’s offices around the world.
These teams draw on McKinsey & Company’s global network of partners and industry and
management experts. In addition, leading economists, including Nobel laureates, act as
research advisers.
The partners of McKinsey & Company fund MGI’s research; it is not commissioned by
any business, government, or other institution. For further information about MGI and to
download reports, please visit www.mckinsey.com/mgi.
http://www.mckinsey.com/mgi
James Manyika | San Francisco
Susan Lund | Washington, DC
Jacques Bughin | Brussels
Jonathan Woetzel | Shanghai
Kalin Stamenov | New York
Dhruv Dhingra | New York
MARCH 2016
DIGITAL GLOBALIZATION:
THE NEW ERA OF GLOBAL FLOWS
PREFACE
The web of global economic connections is growing deeper, broader, and
more intricate. Yet much of the public discussion surrounding globalization
is stuck on the narrow topic of trade surpluses and deficits. This lens fails to
take into account the new and more complex reality of a digitally connected
global economy. While the global goods trade and financial flows have
flattened since the Great Recession, cross-border flows of data are surging.
They now tie the world economy together just as surely as flows of traditional
manufactured goods.
Two years ago, the McKinsey Global Institute (MGI) set out to paint a
comprehensive picture of how globalization is evolving. The resulting report,
Global flows in a digital age: How trade, finance, people, and data connect the
world economy, assessed the network of cross-border inflows and outflows
of trade, services, finance, people, and data and its influence on economic
growth. Building on that earlier work, this report provides a more detailed
analysis of how global flows are continuing to evolve. It offers new insights
into how companies and countries are participating in the web of flows and
extends our econometric analysis, drawing on improved data and employing
more sophisticated methodology. We find even stronger evidence of the
economic value of participating in global flows—and we further find that
data flows account for a substantial portion of that impact. Both inflows and
outflows matter for growth as they circulate ideas, research, technologies,
talent, and best practices around the world.
Today’s more digital form of globalization is changing who is participating, how
business is done across borders, how rapidly competition moves, and where
the economic benefits are flowing. Even though advanced economies in
general continue to be the leaders in most flows, the door has opened to more
countries, to small companies and startups, and to billions of individuals. Our
previous research found the biggest benefits of trade flows go to countries at
the center of the global network. Interestingly, this report finds that countries at
the periphery of the network of data flows stand to gain even more than those
at the center. The convergence of globalization and digitization means that
business leaders and policy makers will need to reassess their strategies—and
given that we are only in the very early stages of this phenomenon, enormous
opportunities are still at stake.
This research was led by James Manyika, a director of the McKinsey Global
Institute based in San Francisco; Susan Lund, an MGI partner based in
Washington, DC; Jacques Bughin, a McKinsey director based in Brussels
who is a core leader of the Firm’s High Tech, Telecom, and Media Practice,
a current member of the MGI Council, and an incoming director of MGI;
and Jonathan Woetzel, an MGI director based in Shanghai. The project
team, led by Kalin Stamenov and Dhruv Dhingra, included Laura Cappellin,
Ritesh Jain, Ayush Mittal, Katie Ramish, Soyoko Umeno, and Amber Yang.
Esteban Arias, Joana Carreiro, Carlos Molina, Moira Pierce, and Vivien Singer
provided valuable research and analytics support. Lisa Renaud served as
senior editor. Sincere thanks go to our colleagues in operations, design,
production, and external relations, including Tim Beacom, Marisa Carder,
Matt Cooke, Deadra Henderson, Richard Johnson, Julie Philpot, Mary Reddy,
Rebeca Robboy, Margo Shimasaki, and Patrick White.
We thank McKinsey colleagues Jörg Bromberger, Michael Chui, Diaan-Yi Lin,
and Sree Ramaswamy for sharing their expertise and insights. Special thanks
go to the volunteers who helped us conduct our survey of global startups:
Ricardo Bernal, Mayank Bishnoi, Oleksandr Bondarenko, Tamas Csikai,
Karol Dolega, Rishika Garg, Thomas Grandin, Catherine Hart, Valeria Laszlo,
Mohato Lekena, Thandi Luzuka, Victoria Muwanga-Zake, Mohit Narotam, and
Siyi Amy Shi.
Our academic advisers provided valuable insights and challenged our
thinking. We are grateful to Matthew J. Slaughter, the Paul Danos Dean of the
Tuck School of Business at Dartmouth; Michael Spence, Nobel laureate and
William R. Berkley Professor in Economics and Business at NYU Stern School
of Business; and Laura Tyson, professor of business administration and
economics at the Haas School of Business, University of California, Berkeley.
Merit E. Janow, dean of the School of International and Public Affairs at
Columbia University, offered new perspectives and directions for our research.
Philip R. Lane, Governor of the Central Bank of Ireland and former Whatley
Professor of Political Economy at Trinity College, Dublin, generously shared
data and perspectives on the international investment positions of countries.
A number of individuals and organizations generously contributed their
time, data, and expertise. For their support in surveying startups, we thank:
Donna Harris, Patrick McAnaney, Morgan Gress, and Kaitlin Walls of 1776,
a global incubator and venture fund dedicated to accelerating innovation
in areas of essential human need. We are also grateful to Molly Jackman
of Facebook; Usman Ahmed of PayPal; Alan Elias of eBay; Robert Pepper
of Cisco; Jarrad Hubbard of TeleGeography; and Uwe Deichmann,
Deepak Mishra, and Daria Taglioni of the World Bank. Without them, this
report would not have been possible.
This report contributes to MGI’s mission to help business and policy leaders
understand the forces transforming the global economy, identify strategic
locations, and prepare for the next wave of growth. As with all MGI research,
this work is independent and has not been commissioned or sponsored in
any way by any business, government, or other institution. We welcome your
comments on the research at [email protected]
Richard Dobbs
Director, McKinsey Global Institute
London
James Manyika
Director, McKinsey Global Institute
San Francisco
Jonathan Woetzel
Director, McKinsey Global Institute
Shanghai
March 2016
© Shutterstock
CONTENTS
HIGHLIGHTS
Soaring cross-border
data flows
The MGI
Connectedness Index
How regions and
cities participate
30
57
63
In brief
Executive summary Page 1
1. A new era of digital globalization Page 23
2. Digital platforms open the door to new participants Page 43
3. How countries, cities, and regions are connecting Page 55
4. Global flows boost economic growth Page 73
5. Competing in a digital global landscape Page 85
6. The new world of policy challenges Page 97
Technical appendix Page 105
Bibliography Page 137
IN BRIEF
DIGITAL GLOBALIZATION:
THE NEW ERA OF GLOBAL FLOWS
The rapidly growing flows of international trade and finance that characterized the 20th century have flattened or
declined since 2008. Yet globalization is not moving into reverse. Instead digital flows are soaring—transmitting
information, ideas, and innovation around the world and broadening participation in the global economy.
� The world is more interconnected than ever. For the first time in history, emerging economies are counterparts
on more than half of global trade flows, and South-South trade is the fastest-growing type of connection.
� While flows of goods and finance have lost momentum, used cross-border bandwidth has grown 45 times
larger since 2005. It is projected to grow by another nine times in the next five years as digital flows of
commerce, information, searches, video, communication, and intracompany traffic continue to surge.
� Digital platforms change the economics of doing business across borders, bringing down the cost of
international interactions and transactions. They create markets and user communities with global scale,
providing businesses with a huge base of potential customers and effective ways to reach them.
� Small businesses worldwide are becoming “micro-multinationals” by using digital platforms such as eBay,
Amazon, Facebook, and Alibaba to connect with customers and suppliers in other countries. Even the
smallest enterprises can be born global: 86 percent of tech-based startups we surveyed report some
type of cross-border activity. The ability of small businesses to reach new markets supports economic
growth everywhere.
� Individuals are participating in globalization directly, using digital platforms to learn, find work, showcase their
talent, and build personal networks. Some 900 million people have international connections on social media,
and 360 million take part in cross-border e-commerce.
� Over a decade, global flows have raised world GDP by at least 10 percent; this value totaled $7.8 trillion
in 2014 alone. Data flows now account for a larger share of this impact than global trade in goods. Global
flows generate economic growth primarily by raising productivity, and countries benefit from both inflows
and outflows.
� The MGI Connectedness Index offers a comprehensive look at how countries participate in inflows and
outflows of goods, services, finance, people, and data. Singapore tops the latest rankings, followed by the
Netherlands, the United States, and Germany. China has surged from No. 25 to No. 7.
� Although more nations are participating, global flows remain concentrated among a small set of leading
countries. The gaps between the leaders and the rest of the world are closing very slowly, but catch-up
growth represents a major opportunity for lagging countries. Some economies could grow by 50 percent or
more over the long term by accelerating participation.
� Many companies grew more complex and inefficient as they expanded across borders. But digital
technologies can tame complexity and create leaner models for going global. This is a moment for companies
to rethink their organizational structures, products, assets, and competitors.
Countries cannot afford to shut themselves off from global flows, but narrow export strategies miss the real value
of globalization: the flow of ideas, talent, and inputs that spur innovation and productivity. Digital globalization
makes policy choices even more complex. Value chains are shifting, new hubs are emerging, and economic
activity is being transformed. This transition creates new openings for countries to carve out profitable roles in the
global economy. Those opportunities will favor locations that build the infrastructure, institutions, and business
environments that their companies and citizens need to participate fully.
Increase in world GDP,
worth $7.8T in 2014
GDP increase from data
flows, larger impact than
goods trade
Global flows increase economic growth
10% $2.8T
Potential GDP boost for some
countries by increasing
participation in
global flows
~50%
Digital technologies are changing how business is done
across borders and broadening participation
Large multinationals
Attain truly global scale with new
markets and suppliers
New strategies for products,
assets, organization
Individuals
New ways to work, learn, and
communicate across borders
>900M have international
connections on social media
SMEs
Use digital platforms to find
customers and suppliers abroad
50M on Facebook, 10M on Alibaba,
2M on Amazon
Startups
>80% of tech-based startups are
“born global”
Foreign customers, financing,
suppliers from day one
Global flows of trade and finance are flattening, while data flows are soaring
1980 2014
DATATRADE FINANCE
45X
growth in
data flows
2005–2014
The new era of digital globalization
© Getty Images
EXECUTIVE SUMMARY
Somewhere in Kenya, a girl logs on for a personalized math lesson from California-based
Khan Academy. Thousands of Syrian refugees rely on Facebook updates for the latest
information to guide their journey through Europe. A multinational energy giant launches
plans to use sensors on 4,000 oil wells around the world to monitor production remotely.
A manufacturer in Australia buys components from a Chinese supplier on Alibaba, and a
clinical trial in India transmits patient data to US pharmaceutical researchers.
The world has become more intricately connected than ever before. Back in 1990, the total
value of global flows of goods, services, and finance amounted to $5 trillion, or 24 percent
of world GDP. There were some 435 million international tourist arrivals, and the public
Internet was in its infancy. Fast forward to 2014: some $30 trillion worth of goods, services,
and finance, equivalent to 39 percent of GDP, was exchanged across the world’s borders.
International tourist arrivals soared above 1.1 billion. And the Internet is now a global network
instantly connecting billions of people and countless companies around the world.
Flows of physical goods and finance were the hallmarks of the 20th-century global
economy, but today those flows have flattened or declined. Twenty-first-century
globalization is increasingly defined by flows of data and information. This phenomenon now
underpins virtually all cross-border transactions within traditional flows while simultaneously
transmitting a valuable stream of ideas and innovation around the world.1
Digitization changes the economics of globalization in several ways. As digital platforms
become global in scope, they are driving down the cost of cross-border communications
and transactions, allowing businesses to connect with customers and suppliers in any
country. Globalization was once for large multinational corporations, but platforms reduce
the minimum scale needed to go global, enabling small business and entrepreneurs around
the world to participate. As a result, new types of competitors can emerge rapidly from any
corner of the world, increasing pressure on industry incumbents.
More than ever before, companies and countries cannot afford to ignore the opportunities
beyond their own borders. Our econometric research indicates that global flows of
goods, foreign direct investment, and data have increased current global GDP by roughly
10 percent compared to what would have occurred in a world without any flows. This
value was equivalent to $7.8 trillion in 2014 alone. Data flows account for $2.8 trillion of this
effect, exerting a larger impact on growth than traditional goods flows. This is a remarkable
development given that the world’s trade networks have developed over centuries but
cross-border data flows were nascent just 15 years ago.
1 This research builds on the 2014 McKinsey Global Institute report Global flows in a digital age: How trade,
finance, people, and data connect the world economy.
The shift to a more digital form of globalization
changes who is participating, how business is done
across borders, and where the economic benefits
are flowing.
2 McKinsey Global Institute Executive summary
Global flows support growth by raising productivity and creating more efficient markets with
truly global scale. But not all countries are making the most of this potential. Our updated
MGI Connectedness Index ranks countries on inflows and outflows of goods, services,
finance, people, and data. Advanced economies are still the most globally connected.
Although more developing countries are deepening their participation, they are narrowing
the gap with the leading advanced economies only very slowly over time.
Accelerating catch-up growth is a major opportunity for the developing world. Our 2014
report showed that countries in the center of trade networks derive more benefit from goods
flows than countries with few connections. But our new research shows that data flows offer
stronger economic benefits to countries on the periphery of the world’s digital networks.
The new age of digital globalization also poses challenges. Companies can enter new
markets, but they are exposed to pricing pressures, aggressive global competitors, and
disruptive digital business models. Data has to be protected against cybercrime. Students
can educate themselves online from anywhere on earth, but their view into other societies
can heighten their impatience with bleak job prospects at home. Social media creates
global communities but also allows networks of extremists to connect. It will take more
international coordination to deal with many of these issues. Today’s version of globalization
is vastly more complex and fast-paced, but connectedness can be a path to growth.
A NEW ERA OF DIGITAL GLOBALIZATION HAS BEGUN
The world has never been more deeply connected by commerce, communication,
and travel than it is today. But the pattern of globalization is shifting. Trade was once
dominated by tangible goods and was largely confined to advanced economies and
their large multinational companies. Today global data flows are surging, and digital
platforms allow more countries and smaller enterprises to participate. This shift has far-
reaching implications.
After a 20-year period of growing roughly twice as fast as the world economy, global flows of
goods, services, and finance hit roughly $30 trillion in 2007, peaking at 53 percent of global
GDP. But this rapid expansion has stopped in its tracks. Growth in global goods trade has
flattened, financial flows have fallen sharply, and trade in services has posted only modest
growth. These flows have finally regained their pre-recession levels in terms of dollar value,
but they are now just 39 percent of world GDP (Exhibit E1).
Many observers point to this trend as evidence that globalization has stopped.2 We have
a different view: globalization has instead entered a new era defined by data flows that
transmit information, ideas, and innovation. Digital platforms create more efficient and
transparent global markets in which far-flung buyers and sellers find each other with a few
clicks. The near-zero marginal costs of digital communications and transactions open new
possibilities for conducting business across borders on a massive scale.
2 See, for example, David Smick, “Could globalization crack up?” International Economy, fall 2012; Joshua
Cooper Ramo, “Globalism goes backward,” Fortune, November 20, 2012; and Jeffrey Rothfeder, “The great
unraveling of globalization,” Washington Post, April 24, 2015.
Soaring cross-border data flows now generate
more economic value than traditional flows of
traded goods.
3McKinsey Global Institute Digital globalization: The new era of global flows
Traditional flows of goods, services, and finance have flattened
For two decades, the world’s trade in goods (including commodities, finished goods, and
intermediate inputs) grew roughly twice as fast as global GDP as major multinationals
expanded their supply chains and established new bases of production in countries with
low-cost labor. Global trade in goods soared from 13.8 percent of world GDP in 1986 to
26.6 percent in 2008 on the eve of the Great Recession. After a sharp decline and short-
lived rebound, however, the goods trade has been growing more slowly than world GDP
in recent years, puzzling economists and business leaders alike. Some of this decline is
cyclical. Our analysis suggests that weak demand and plummeting prices for commodities
account for nearly three-quarters of the decline in trade.
But trade in both finished and intermediate manufactured goods has also declined, thanks
to several structural forces. The makers of many finished goods are beginning to place less
importance on labor costs and more on speed to market and non-labor costs. As a result,
some production is moving closer to end consumers. Trade is also declining for many
intermediate goods such as chemicals, paper, textile fabrics, and communications and
electrical equipment. This suggests that global value chains may be shortening, at least in
part because of the cost of managing complex, lengthy supply chains.
In the decade ahead, the global goods trade may continue to decline relative to world GDP.
At a minimum, it is unlikely to resume rapid growth. Not only are factor costs changing, but
3D printing and other technologies also have the potential to transform how—and where—
goods such as electronics, vehicle parts, other transportation equipment, machinery and
electrical equipment, medical instruments, and apparel are produced.
Cross-border financial flows—which include lending, foreign direct investment (FDI), and
purchases of equities and bonds—link together national financial markets, connecting
borrowers and savers from different countries. They grew from $0.5 trillion in 1980
(4.1 percent of global GDP) to $11.9 trillion in 2007 (20.7 percent of global GDP). But 2007
proved to be the height of a global credit bubble. Since then financial flows have fallen to less
Exhibit E1
39
3738
4141
53
37
32
27
24
2322
24
2221
2324
26
2014
18
28
3029
9
655
8
29
109
32
6
29
1990
24
5
10
643 433 333
1980
3
10
32
2000
12 11
49
28
3735
22
17
41
32
46
2007
30 31
24
21
13
26
SOURCE: UNCTAD; IMF Balance of Payments; W orld Bank; McKinsey Global Institute analysis
Finance
Goods
Services
After 20 years of rapid growth, traditional flows of goods, services, and finance have declined relative to GDP
Flows of goods, services, and finance, 1980–2014
$ trillion, nominal All flows as % of GDP
-14 p.p.
Global flows 2
ES
mc 0307
4 McKinsey Global Institute Executive summary
than half their previous value ($5.2 trillion in 2014); they are only one-third as high relative to
global GDP.3 A decline in cross-border lending accounts for the majority of the overall drop
in financial flows and may reflect a return to long-term trend. But other types of portfolio
investment and FDI have also fallen, raising concerns about financing for emerging markets.
Accelerating flows of data and information are changing the dynamics
of globalization
While global flows of trade and finance have lost momentum, the volume of data being
transmitted across borders has surged, creating an intricate web that connects countries,
companies, and individuals (Exhibits E2 and E3).4
Global flows of data primarily consist of information, searches, communications,
transactions, video, and intracompany traffic. They underpin and enable virtually every other
kind of cross-border flow. Container ships still move products to markets around the world,
but now customers order them online, track their movement using RFID codes, and pay
for them via digital transactions. Although videos use a majority of Internet bandwidth, the
Internet of Things and other business applications are gaining importance. Indeed, Cisco
estimates that machine-to-machine connections will account for more than 40 percent of
global devices and connections by 2019.5
3 Financial globalization: Retreat or reset? McKinsey Global Institute, March 2013.
4 To measure these flows, we track used cross-border bandwidth, which is highly correlated with Internet traffic.
5 Cisco Visual Networking Index: Forecast and methodology, 2014–2019, Cisco, May 2015.
Exhibit E2
Cross-border data flows are surging and connecting more countries
SOURCE: TeleGeography, Global Internet Geography; McKinsey Global Institute analysis
Used cross-border bandwidth
2005
100% = 4.7 Terabits per second (Tbps)
NA
United States and Canada
LA
Latin America
AF
Africa
EU
Europe
ME
Middle East
OC
Oceania
AS
Asia
Regions
Bandwidth
Gigabits per second (Gbps) <50 50–100 100–500 500–1,000 1,000–5,000 5,000–20,000 >20,000
2014
100% = 211.3 Tbps
NOTE: Lines represent interregional bandwidth (e.g., between Europe and North America) but exclude intraregional cross-border bandwidth (e.g., connecting
European nations with one another).
45x larger
NA
EU
LA
ME
AF
AS
OC
NA
EU
LA
ME
AF
AS
OC
5McKinsey Global Institute Digital globalization: The new era of global flows
Tangible flows
of physical goods
Flows mainly between
advanced economies
Capital- and labor-
intensive flows
Transportation
infrastructure is
critical for flows
Multinational
companies
drive flows
Flows mainly of
monetized
transactions
Ideas diffuse slowly
across borders
Intangible flows of
data and information
Greater participation by
emerging economies
More knowledge-
intensive flows
Digital infrastructure
becomes equally
important
Growing role of
small enterprises
and individuals
More exchanges of
free content and
services
Instant global access
to information
Innovation flows
from advanced to
emerging economies
Innovation flows in
both directions
Globalization: Then vs. now
Exhibit E3
6 McKinsey Global Institute Executive summary
Digital platforms are key to this new era of globalization. Over the past two decades, the
largest corporations built their own digital platforms to manage suppliers, connect to
customers, and enable internal communication and data sharing for employees around
the world. But a diverse set of public Internet platforms has emerged to connect anyone,
anywhere. These include operating systems, social networks, digital media platforms,
e-commerce websites, and all kinds of online marketplaces. Their use of automation and
algorithms drives the marginal costs of adding new interactions practically to zero, allowing
the biggest platforms to support hundreds of millions of global users (Exhibit E4). Now users
can more easily see details on products, services, prices, and alternative choices. This
removes some information asymmetries so that markets function more efficiently, although it
can disrupt some intermediaries in the process.
Exhibit E4
The biggest online platforms have user bases on par with the populations of the world’s
biggest countries
Active users of online platforms vs. country population
Million
205
256
300
300
320
321
400
407
650
1,000
1,000
1,314
1,372
1,590
China
YouTube
India
Alibaba
Skype
Brazil
Amazon
Indonesia
United States
Countries2
Online platforms1
1 4Q15 or latest available.
2 2015 population.
SOURCE: Facebook; Twitter; Alibaba; Fortune; Statista; Population Reference Bureau; McKinsey Global Institute analysis
7McKinsey Global Institute Digital globalization: The new era of global flows
Approximately 12 percent of the global goods trade is conducted via international
e-commerce, with much of it driven by platforms such as Alibaba, Amazon, eBay, Flipkart,
and Rakuten. Beyond e-commerce, digital platforms for both traditional employment
and freelance assignments are beginning to create a more global labor market.6 Some
50 percent of the world’s traded services are already digitized.7
Digitization also enables instantaneous exchanges of virtual goods. E-books, apps, online
games, MP3 music files and streaming services, software, and cloud computing services
can all be transmitted to customers anywhere in the world there is an Internet connection.
Many major media websites are shifting from building national audiences to global ones; a
range of publications, including The Guardian, Vogue, BBC, and BuzzFeed, attract more
than half of their online traffic from foreign countries. By expanding its business model
from mailing DVDs to selling subscriptions for online streaming, Netflix has dramatically
broadened its international reach to more than 190 countries. While media, music, books,
and games represent the first wave of digital trade, 3D printing could eventually expand
digital commerce to many more product categories.
Finally, “digital wrappers” are digital add-ons that enable and raise the value of other types
of flows. Logistics firms, for example, use sensors, data, and software to track physical
shipments, reducing losses in transit and enabling more valuable merchandise to be
shipped and insured. Online user-generated reviews and ratings give many individuals
the comfort level needed to make cross-border transactions, whether they are buying a
consumer product on Amazon or booking a hotel room halfway around the world on Airbnb,
Agoda, or TripAdvisor.
DIGITIZATION IS MAKING GLOBAL FLOWS MORE INCLUSIVE
Globalization was once driven almost exclusively by governments, large multinational
corporations, and major financial institutions. Today artisans, entrepreneurs, app
developers, freelancers, small businesses, and even individuals can participate directly on
digital platforms with global reach.
SMEs can be micro-multinationals, and digital startups are born global
Small and medium-sized enterprises (SMEs) worldwide are using the “plug-and-play”
infrastructure of Internet platforms to put themselves in front of an enormous global
customer base and become exporters. Amazon, for instance, now hosts some two million
third-party sellers. In countries around the world, the share of SMEs that export is sharply
higher on eBay than among offline businesses of comparable size. PayPal enables cross-
border transactions by acting as an intermediary for SMEs and their customers. Participants
from emerging economies are senders or receivers in 68 percent of cross-border PayPal
transactions. Microenterprises and projects in need of capital can turn to platforms such as
Kickstarter, where nearly 3.3 million people representing nearly all countries made pledges
in 2014.
Facebook estimates that 50 million SMEs are on its platform, up from 25 million in 2013; on
average 30 percent of their fans are from other countries. To put this number in perspective,
consider that the World Bank estimated there were 125 million SMEs worldwide in 2010.
For small businesses in the developing world, digital platforms are a way to overcome
constraints in their local markets. The ability of SMEs to reach global audiences supports
economic growth everywhere.
6 A labor market that works: Connecting talent with opportunity in the digital age, McKinsey Global Institute,
June 2015.
7 Daniel Castro and Alan McQuinn, Cross-border data flows enable growth in all industries, Information
Technology and Innovation Foundation, February 2015.
12%
of the global goods
trade is
e-commerce
8 McKinsey Global Institute Executive summary
The increasing globalization of small businesses is starting to show up in national statistics.
It is most clearly seen in the United States, where the share of exports by large multinational
corporations dropped from 84 percent in 1977 to 50 percent in 2013. Among SMEs that
export, the smallest (those with fewer than 50 employees) are gaining share the fastest. An
analysis of export data for 16 OECD countries shows mixed evidence, with the SME share of
total exports growing in ten of the countries.8
Even new startups can form global connections and market to international customers
from their inception. We surveyed 271 startups worldwide through a partnership with
1776, a global incubator and venture fund. By working with 1776 and its Startup Federation
program, we were able to expand the reach of the survey to 19 countries. While these
startups represent a more tech-savvy cross-section than the broader universe of
entrepreneurs, the results show that even the smallest and youngest enterprises can
execute a global vision if their business model is built on digital technologies. A surprising
86 percent of survey respondents pointed to at least one cross-border activity. Almost two-
thirds have customers or users in other countries, and almost half reported sourcing talent
from other countries.
Individuals can participate directly in globalization, with significant
economic impact
Thanks to social media and other Internet platforms, individuals are forming their own cross-
border connections. We estimate that 914 million people around the world have at least
one international connection on social media, and 361 million participate in cross-border
e-commerce (Exhibit E5). These figures are growing rapidly. On Facebook, 50 percent of
users now have at least one international friend. This share is even higher—and growing
faster—among users in emerging economies.
The business and economic implications of individual participation are significant. Digital
platforms provide a huge built-in base of potential customers and effective ways to market
to them directly. As social media exposes consumers from around the world to what is
available, products can go viral on a scale that has never been seen before. In 2015, Adele’s
song “Hello” racked up 50 million views on YouTube in its first 48 hours, and her album 25
sold a record 3.38 million copies in the United States in its first week alone, more than any
other album in history. In 2012, Michelle Obama wore a dress from British online fashion
retailer ASOS in a photo that was retweeted 816,000 times and shared more than four
million times on Facebook; it instantly sold out.
Digital platforms offer individuals new ways to learn, collaborate, and acquire new skills—
and then to showcase their talents to potential employers. Some 44 million people around
the world find freelance work on Freelancer.com, Upwork, and other digital platforms; nearly
400 million have posted their professional profiles on LinkedIn. Individuals with creativity and
drive can propel themselves onto a global stage in ways that would have been unimaginable
in the pre-digital world. A number of previously unknown singers have been discovered after
posting videos on YouTube. The Weeknd, spotted on YouTube by Drake, dominated the
Billboard charts in 2015 and recently earned an Oscar nomination for best original song.
8 Some countries where SME share of exports declined were those suffering from a post-crisis credit crunch,
such as Portugal.
86%
of surveyed
startups report at
least one
cross-border
activity
9McKinsey Global Institute Digital globalization: The new era of global flows
GLOBAL FLOWS DRIVE ECONOMIC GROWTH, BUT COUNTRY PARTICIPATION
IS UNEVEN
In this report, we set out to develop more robust estimates of whether global flows
contribute to economic growth, using an expanded and improved data set and more
sophisticated statistical methods than in our last report on this topic, in 2014.9 We find even
stronger evidence that global flows increase GDP in the long term by raising productivity
and that data flows have as much impact as goods trade. But we also find that country
participation varies widely, and every type of flow remains dominated by a small group of
leading countries. There is enormous value at stake for lagging countries in catching up.
9 We first test for cointegration in the data and then use an error-correction econometric model. Our data cover
1995–2013 and 97 countries. See the technical appendix for a comprehensive discussion of the econometric
model, different statistical tests, and the variables and data used.
Exhibit E5
Individuals are participating in globalization, and 914 million have cross-border social media connections
SOURCE: Facebook; AliResearch; US Department of Commerce; OECD; W orld Bank; McKinsey Global Institute analysis
Students
studying abroad
5 million
Cross-border
e-commerce shoppers
361 million
International
travelers
429 million
People living outside
home country
244 million
Cross-border
online students
13 million
Cross-border
online workers
44 million
Social networking users with
at least one foreign connection
914 million
NOTE: Numbers adjusted to account for overlap between platforms and for individuals making multiple international trips in the same year.
REPEATS
in report
10 McKinsey Global Institute Executive summary
Global flows raised world GDP growth by 10 percent, or $7.8 trillion,
in 2014 alone
Our econometric analysis finds robust evidence that global flows of goods, FDI, people, and
data contribute structurally to economic growth by increasing productivity.10 It breaks new
ground by testing the impact of all types of flows together, both inflows and outflows, and
considering how countries are positioned in each web of flows.
Our results indicate that over a decade, global flows have raised world GDP by roughly
10 percent over what would have resulted in a world in without any flows. In 2014 alone, they
generated roughly $7.8 trillion in value. Flows of goods and FDI account for about half of this
impact, while data flows, the hallmark of 21st-century globalization, account for $2.8 trillion.
All types of global flows boost productivity growth, and data flows additionally appear to
increase the amount of labor and capital used in the economy.
We also examine how a country’s position in the network of flows affects the benefits it
receives. Countries in the center of the global network of goods trade benefit more than
those at the periphery. The network of cross-border data flows, by contrast, is still rather
new and less dense. The United States and Europe are at the center of the world’s digital
networks, facilitating links to other countries. But we find that countries at the periphery of
this digital network stand to gain even more than those at the center. For economies that
have been relatively disconnected, the arrival of new digital platforms and cross-border data
flows can be transformational.
Overall, our analysis underscores the value of connectedness—and the benefits are much
broader and more nuanced than a simple accounting of net exports can capture. Countries
that participate in global flows gain exposure to ideas, research, technologies, talent,
and best practices from around the world. The most connected economies can draw on
these flows to enhance their own competitiveness, innovation, and efficiency, positioning
themselves to take advantage of growth opportunities in global markets. However, countries
also need to have supporting institutions and policies in place to realize this potential.
Although more countries are participating, global flows remain concentrated
among a relatively small group of leading countries
Today global connections link a larger and more diverse range of countries than ever. For
the first time in history, emerging economies are counterparts on more than half of global
trade flows, and South-South trade between these countries is the fastest-growing type
of connection. The value of traded goods and services plus financial flows exceeded
80 percent of GDP for only 72 countries (mainly developed ones) in 1990; by 2014, that was
true for 121 countries. But while more countries are participating in global flows, their level of
participation varies widely.
10 We include only the FDI component of total financial flows, since those have been shown by other research
to be correlated with GDP growth. The impact of other forms of financial flows on growth is mixed. We do
not include service flows in our econometric analysis because they are highly correlated with FDI and with
goods trade.
We find strong evidence that global flows increase
GDP over the long term by raising productivity. Both
inflows and outflows matter for growth.
11McKinsey Global Institute Digital globalization: The new era of global flows
The MGI Connectedness Index offers a comprehensive look at how countries participate in
inflows and outflows of goods, services, finance, people, and data (Exhibit E6).11 Our index
takes into account the size of each flow for a country relative to its own GDP or population
(flow intensity) as well as its share of each total global flow. Combining these measures
avoids making large and diversified economies appear closed simply due to the extent of
economic activity taking place within their own borders.
Singapore, a small country that punches far above its weight in all types of global flows, tops
this year’s rankings. It is followed by the Netherlands (one of Europe’s main digital hubs),
the United States, Germany, Ireland, and the United Kingdom. China’s surge is particularly
noteworthy; it has climbed from 25th in our previous index to the No. 7 spot.
However, the world is still far from fully globalized. Advanced economies in general
remain more connected than developing countries, and the top countries have far
higher connectedness scores than the rest of the world Exhibit E7). All types of flows are
concentrated among a small set of countries. The top 15 countries in traded goods account
for 63 percent of the global total; that share is 62 percent in services and 79 percent in FDI.
We use statistical tests of convergence to see if the gaps between country participation in
global flows are closing over time. Our results indicate that lagging countries are catching up
to leading countries—but extremely slowly, given that the global flows of leading countries
continue to rise. At current trends, cutting the gap in half would take eight years in the
goods trade and 13 years in FDI flows. For data flows, we do not see any sign that laggards
are catching up to leaders, perhaps reflecting that digitization has a long way to go in all
countries and it is a relatively young phenomenon.
Lagging countries could realize tremendous growth potential by accelerating their
participation in well-targeted ways. We find that countries in the top quartile increased their
flow of goods relative to GDP at an average of 3 percent annually, for example, while goods
flows grew at only 1 percent for the bottom quartile. The top-quartile countries increased
FDI flows by 5 percent of GDP annually during this period, while those flows shrank by
8 percent annually for countries in the bottom quartile. If countries in the bottom three
quartiles had increased participation in flows at the same rate as the top quartile over the
past decade, global GDP would be an additional $10 trillion, or 13 percent, higher today.
In other words, limited participation in global flows by many countries had a real cost to
the world economy. For some individual countries, GDP would be more than 50 percent
higher today.
Countries have taken different routes to become more globally connected. Top-ranked
Singapore emerged decades ago as Southeast Asia’s global shipping hub. It subsequently
mapped out an explicit strategy to become a regional hub for finance and services by
attracting skilled international talent and establishing incentives and promotional efforts
to attract FDI. The Netherlands is a major hub for Europe’s data traffic as well as a port
for traded goods. Like Ireland, it has created tax and regulatory regimes to attract many
subsidiaries, headquarters, and holding companies for multinational corporations. In
contrast, the United States and Germany both follow a generalist model with strength
across all five flows. The United Kingdom also has broad participation across flows, with a
spike in cross-border service and financial flows, a reflection of London’s role as a global
financial hub.
11 Several other indexes measure the degree to which countries are connected to global activity, although they
use different data and weighting. These include the DHL Global Connectedness Index produced by Pankaj
Ghemawat and Steven A. Altman and globalization indexes from Ernst & Young, A. T. Kearney, and the Swiss
Economic Institute. See, for example, Pankaj Ghemawat and Steven A. Altman, Depth Index of Globalization
2013: And the big shift to emerging economies, IESE Business School, University of Navarra, 2013.
12 McKinsey Global Institute Executive summary
Exhibit E6
MGI Connectedness Index
Country connectedness index and overall flows data, 2014
Rank of participation by flow as measured by flow intensity and share of world total
1–10 11–25 26–50 >50Connectedness index rank 100+ <7070–99Flow intensity Rank Country Score Connectedness Index rank Flow value1 $ billion Flow intensity2 % of GDPGoods Services Finance People Data 1 Singapore 64.2 1 2 2 12 6 1,392 452 2 Netherlands 54.3 3 3 6 21 1 1,834 211 3 United States 52.7 7 7 3 1 7 6,832 39 4 Germany 51.9 2 4 8 3 2 3,798 99 5 Ireland 45.9 32 1 1 28 9 559 227 6 United Kingdom 40.8 13 5 5 6 3 2,336 79 7 China 34.2 4 16 4 82 38 6,480 63 8 France 30.1 11 8 9 7 4 2,262 80 9 Belgium 28.0 5 6 33 33 8 1,313 246 10 Saudi Arabia 22.6 20 28 27 2 53 790 106 11 United Arab Emirates 22.2 6 23 17 4 46 789 196 12 Switzerland 18.0 12 11 10 17 13 848 115 13 Canada 17.3 16 22 11 11 18 1,403 79 14 Russia 16.1 21 25 18 5 25 1,059 57 15 Spain 14.4 25 13 19 14 16 1,105 79 16 Korea 14.0 8 12 28 50 44 1,510 107 17 Italy 13.4 17 18 24 16 19 1,587 74 18 Sweden 13.0 29 14 22 31 5 572 100 19 Austria 11.7 26 17 31 20 12 470 108 20 Malaysia 11.6 9 19 25 26 43 610 187 21 Mexico 10.7 14 63 34 18 41 1,022 80 22 Thailand 10.7 10 15 36 44 64 605 162 23 Kuwait 10.6 37 46 13 13 75 306 153 24 Japan 10.5 15 20 12 81 20 2,498 54 25 Kazakhstan 10.0 48 73 41 8 57 176 83 26 Ukraine 9.8 38 39 87 10 34 133 101 27 Australia 9.7 30 34 21 15 33 825 57 28 Denmark 8.9 35 9 32 41 11 369 108 29 Jordan 8.8 73 50 75 9 83 50 138 30 India 8.5 24 10 35 58 70 1,316 64 32 Czech Republic 7.5 18 33 57 59 15 397 193 34 Poland 7.0 23 31 47 34 22 585 107 35 Hungary 6.8 22 30 26 62 17 287 209 36 Norway 6.0 36 24 20 46 24 458 92 37 Vietnam 5.7 19 54 45 103 61 350 188 39 Finland 5.5 46 27 23 70 10 390 144 40 Portugal 5.5 47 36 30 23 31 255 111 41 Turkey 5.1 28 40 53 38 29 521 65 43 Israel 4.9 51 32 49 24 56 248 82 44 Brazil 4.5 41 38 14 125 30 869 37 45 Chile 4.1 45 58 16 102 27 239 92 47 Greece 4.1 60 29 54 35 42 160 67 48 New Zealand 3.9 67 48 61 25 51 130 63 51 Indonesia 3.4 31 49 38 106 76 504 57 53 South Africa 3.3 34 57 52 64 80 277 79 54 Philippines 3.2 54 41 44 52 67 230 81 64 Morocco 2.6 58 43 74 56 65 104 97 73 Egypt 2.2 68 42 69 73 71 158 55 83 Nigeria 1.9 55 76 48 128 98 268 47 86 Peru 1.8 62 88 51 104 49 122 60 118 Kenya 1.3 100 84 127 119 91 35 58 SOURCE: McKinsey Global Institute analysis 1 Flows value represents total goods, services, and financial inflows and outflows. 2 Flow intensity represents the total value of goods, services, and financial flows as a share of the country’s GDP. 13McKinsey Global Institute Digital globalization: The new era of global flows Exhibit E7 A small group of leading countries are much more connected than the rest of the world 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 0 805030 40 7060 1402010 Connectedness score, 2014 United Kingdom Singapore Sweden United Arab Emirates Mexico Russia Saudi Arabia Turkey Australia Belgium Switzerland Spain Indonesia Ireland India Norway Austria Japan South Korea France China Canada Italy United States Per capita GDP, 2014 $ thousand, purchasing power parity, current international dollar Netherlands Germany Portugal Kuwait Denmark Ukraine QatarFinland Czech RepublicGreece Emerging Developed Correlation coefficient (r) = 0.54 Size of circle represents $ value of flows in 2014 SOURCE: IMF; McKinsey Global Institute analysis Vietnam Brazil Thailand Malaysia 14 McKinsey Global Institute Executive summary Although our report mainly assesses the global connectedness of countries, nation-states are not the only lens through which to observe globalization. Cities, regions within countries, and broader blocs of countries are connecting with the global economy in myriad ways and to varying degrees. For instance, our previous report found that the world had only eight truly “global cities” with strong connections in at least four of the five major flows: New York, London, Tokyo, Los Angeles, San Francisco, Singapore, Hong Kong, and Dubai. This year Tokyo drops off the list due to a decline in goods trade, while Shanghai takes its place. Within countries there can be very different patterns of globalization. In the United Kingdom and Germany, for instance, the variation across regions is modest. China, by contrast, has a handful of highly connected coastal provinces and largely unconnected inland provinces. Some highly connected states and provinces rank as economic powerhouses in their own right: China’s booming province of Guangdong would rank sixth globally in terms of goods flows, while California would rank fourth in the world for people flows. We also look at the patterns of trade among neighbors and trading blocs. Europe is the most integrated region; more than 60 percent of its trade in goods is intraregional. But the corresponding shares are sharply lower in Africa, Latin America, and South Asia. This indicates a significant opportunity for developing countries to increase their participation in flows by trading with their neighbors (Exhibit E8). Exhibit E8 35 65 While much of the world’s trade in goods is long distance, roughly half or more of other global flows move within the same region Distribution of flows between intraregional (short haul) vs. interregional (long haul), 20141 % of world flow SOURCE: UNCTAD; UN W orld Tourism Organization; TeleGeography, Global Internet Geography; IMF; McKinsey Global Institute analysis 1 For goods, services, FDI, and travelers we have divided the world into 10 regions; for data flows we have used TeleGeography’s six regions. 2 Distribution of services flows for 2014 estimated based on 2011 data; 2013 bilateral traveler data used for people flows. NOTE: Numbers may not sum due to rounding. 36 64 Long haul (interregional) Short haul (intraregional) 53 47 54 46 67 33 Services2 FDI Data Goods People2 15McKinsey Global Institute Digital globalization: The new era of global flows COMPANIES MAY NEED TO REINVENT THEMSELVES TO WIN IN A DIGITAL GLOBAL MARKETPLACE The new era of digital globalization offers unprecedented opportunities for companies to achieve both global scale and efficiency, but it also calls for reevaluating existing strategies, business models, and operations. Business leaders in all industries should consider the following issues: � Do your footprint and organizational structure make sense in a more digital world? As companies expanded across borders, many encountered a “globalization penalty” due to the costs of rising complexity.12 But now digital technologies allow companies to globalize in a leaner way. Digital tools for remote collaboration and instant communication mean that it is possible to centralize some global functions, such as back-office operations or R&D; to create virtual global teams that span borders; or even to forgo having one global headquarters location. Digitization is also enabling business models that are less capital-intensive. Rather than establishing a large physical presence in many countries, some companies focus local offices on sales and marketing only. Those that deliver digital goods and services can enter new international markets without establishing a physical presence at all. � Should you offer one brand and one product line around the world, or customize for local markets? In some industries, product tailoring is driven by local regulatory requirements or language differences. In others, companies that sell into many global markets have expanded their product portfolios to appeal to local consumer preferences and price points. But others take a different approach: offering products that are the same everywhere in the world. Apple, for instance, offers just three models of its iPhone and iPad, all with consistent design and branding wherever they are sold. Facebook, Uber, and Airbnb have simply scaled up their digital platforms in country after country with limited customization. Many global automakers are attempting to strike a balance by whittling down the number of platforms used across their international manufacturing operations (that is, using fewer underlying designs that can be customized by swapping certain components to create differentiated models). The media and consumer technology industries are shifting to simultaneous global product launches, since consumers around the world can see instantaneously what is offered in other countries. � Do you have the right suppliers and customer channels? Digital tools can orchestrate a multitude of vendors around the globe with greater precision and efficiency. But even as technology enables more complex global value chains, the importance of different factor costs is shifting. Until relatively recently, many companies were willing to fully outsource manufacturing and other functions to locations with low-cost labor. Today many are reevaluating those decisions and giving greater weight to energy prices, distance to market, infrastructure, ease of doing business, and risk. According to a recent UPS survey, approximately one-third of high-tech companies are moving manufacturing or assembly closer to end-user markets; this number is up by 25 percentage points from 2010.13 As China’s labor costs rise and the country moves into higher-value-added industries, more of the world’s manufacturing business is up for grabs. Businesses will have to consider whether their suppliers and customer channels should change. 12 Martin Dewhurst, Jonathan Harris, and Suzanne Heywood, “Understanding your ‘globalization penalty,’” McKinsey Quarterly, July 2011. 13 Change in the (supply) chain, United Parcel Service, 2015. 16 McKinsey Global Institute Executive summary � Do you have the right assets to compete digitally and globally? Building digital platforms, online customer relationships, and data centers may be critical for a growing range of companies, far beyond the Internet giants. GE, for example, is transforming its core manufacturing capabilities to establish itself as a leader in Internet of Things technology. Businesses in all industries need to take a fresh look at their assets, including customer relationships and market data, and consider whether there are new ways to monetize them. Alibaba has a vast pool of transactional data on the vendors that operate on its platform, and it has built on it to move into new areas such as mobile payments and small business financing. The insurance industry could similarly harness its sophisticated data pools on different forms of risk to create new products and services. � Are you ready for a new era of digitally accelerated global competition? Competition is intensifying and product cycles are shortening due to the confluence of three trends. First, emerging-market giants are going global. Many of them are aggressive, deep-pocketed, and able to operate with different time horizons and financial targets. By 2025, MGI estimates that companies headquartered in emerging markets will make up 45 percent of the global Fortune 500, up from 26 percent today.14 Second, tech companies are expanding into new industries. Some of the truly disruptive players are siphoning value out of industries and giving it away for free to consumers as a way to build their positions. Finally, the largest Internet platforms allow millions of SMEs and startups to go head-to-head with incumbents. These new forms of competition have unleashed pricing pressures and industry disruptions. The Internet and international competition have cut into the window of exclusivity companies once enjoyed on new products and services; “copycat” versions can be launched in new markets even before the originator has time to scale up. It is more important than ever to stay alert to new competitive threats. � Are you prepared for new risks? As the world grows more dependent on information systems, the private sector is also becoming more vulnerable to cyberattacks. It is difficult to stay ahead of increasingly sophisticated hackers, but companies can prioritize their information assets, test continuously, and work with frontline employees to emphasize basic protective measures. If a breach does occur, a decisive and forthright response from marketing, public affairs, and customer service functions can be critical to restoring customer trust.15 Maintaining data security has to be a top priority for CEOs in every industry. POLICY MAKERS FACE A NEW WORLD OF CHALLENGES Countries cannot afford to shut themselves off from global flows, given the value at stake in raising productivity and long-term GDP growth. Pursuing this opportunity requires a new policy agenda that includes the issues outlined below. � Thinking strategically about the role your country can play. Policy makers should carefully consider how to build on their country’s comparative advantages. Many countries are trying to develop the next Silicon Valley, but innovation is notoriously difficult to orchestrate. Meanwhile, developing nations may face a shrinking opportunity to become low-cost manufacturers for the world as automation advances. But other opportunities exist. Some countries can build on their geographic proximity to major consumer markets, as Mexico and Eastern Europe have done. Others may develop a successful niche as global transit hubs, as Dubai has done in transportation and trade flows. Other countries have targeted a particular flow or industry to cultivate, 14 See Playing to win: The new global competition for corporate profits, McKinsey Global Institute, September 2015, and Urban world: The shifting global business landscape, McKinsey Global Institute, October 2013. 15 Risk and responsibility in a hyperconnected world: Implications for enterprises, McKinsey & Company and the World Economic Forum, January 2014. 17McKinsey Global Institute Digital globalization: The new era of global flows perhaps building on pools of talent within their borders (as India has done with business process outsourcing). � Addressing policy and administrative barriers that hinder global flows. Pursuing bilateral and multilateral trade partnerships is the cornerstone of a more open approach. Another important step is removing import tariffs, quotas, and subsidies for national industries, all of which can introduce distortions. Other types of legal and administrative barriers also have to be dismantled to make the most of global flows; these may include limitations on foreign business ownership and investment, import licensing, regulatory requirements that deviate from international norms, and limits on immigration. The Association of South East Asian Nations (ASEAN), for instance, has largely eliminated import tariffs among its ten member states, but its ongoing effort to build a seamless trading bloc involves harmonizing product standards, certification procedures, customs requirements, and cross-border regulations covering traded services and the movement of labor.16 � Addressing dislocations. Even though their net global effect is ultimately positive, global flows can cause job losses and displacement in the short run. Governments have to consider these trade-offs and open to global flows at a pace their economies and societies can absorb. Few countries have adequately supported the workers and communities affected by exposure to international competition and disruptive business models. But these workers will need a clearer path to new roles—and the societal cost of neglecting this issue grows over time. It will take a much more proactive response to ensure that labor markets and training systems can deal with rapid change. � Investing in human capital. The Internet can promote inclusiveness, but only if education and training systems provide language fluency, basic digital literacy, and other skills so that individuals can take advantage of the opportunities. Investment in human capital development will be a critical determinant of which nations come out on top. � Building the necessary infrastructure and closing the digital divide. Even in a more digital world, roads, ports, airports, and rail remain vital as the conduits of trade and mobility. But today any list of infrastructure priorities also has to include universal, affordable Internet access. At the end of 2015, 57 percent of the world’s population, or four billion people, remained offline, and only 15 percent had access to broadband.17 The value of connecting these people is significant. Our own econometric analysis shows that countries with higher Internet penetration reap up to 25 percent more benefit from cross-border data flows than those with limited Internet penetration. � Creating a strong business and institutional environment. A recent World Bank report finds that in many developing countries, the economic benefits of digital technologies have been limited by a lack of strong fundamentals such as education and good governance.18 To capture the full growth potential of digital globalization, countries need to cultivate a healthy business environment that nurtures startups, allows inefficient firms to exit, ensures a level playing field, and establishes a solid legal framework for intellectual property and property rights. � Protecting data privacy while maintaining an open Internet. Many countries have enacted or are considering limitations on what kind of data can be transmitted across borders; this may include requirements that companies use servers physically located 16 Southeast Asia at the crossroads: Three paths to prosperity, McKinsey Global Institute, November 2014. 17 The state of broadband 2015, International Telecommunication Union and UNESCO, September 2015. For more on policy approaches to addressing this issue, see Offline and falling behind: Barriers to Internet adoption, McKinsey Technology, Media and Telecom Practice, September 2014. 18 World development report 2016: Digital dividends, World Bank, January 2016. 18 McKinsey Global Institute Executive summary within their borders to process and store data generated there. As we went to press, for example, the future of the “safe harbor” agreement governing data transfers between the European Union and the United States remained uncertain. Legitimate privacy concerns need to be addressed through thoughtful frameworks, but data localization and fragmented regulation may have real economic costs.19 � Making cybersecurity a top priority. One study has estimated that cybercrime costs the global economy some $400 billion in annual losses through consumer data breaches, financial crimes, market manipulation, and theft of intellectual property.20 Hackers may also pose public safety and even national security risks. While companies are often at the forefront of ensuring cybersecurity, governments can invest in research, share information, model good security practices, and craft thoughtful rules. Governments will need to work closely with their global counterparts and with the business community to stay on top of new threats and share technology solutions. Regulators may need to mandate standards for securing consumer data, and public agencies need to safeguard their own assets. ••• Many of the challenges associated with digitizing economic activity are now playing out on a global scale. Even measuring digital globalization in statistics has become a more complex undertaking, since much of the value being generated winds up as consumer surplus. Our analysis provides strong evidence of the economic value of openness—and it shows that both inflows and outflows matter, as they expose an economy to ideas, research, technologies, talent, and best practices from around the world. For countries that have been slow to participate, the opportunities for catch-up growth are too substantial to ignore. 19 Matthias Bauer et al., The costs of data localization: Friendly fire on economic recovery, ECIPE occasional paper number 3/2014, May 2014, analyzes recently proposed or enacted data localization rules in seven economies. It found that these rules would lower GDP in all seven cases, with Vietnam (-1.7 percent), China (-1.1 percent), and Indonesia (-0.5 percent) poised for the largest losses. 20 Net losses: Estimating the global cost of cybercrime, Center for Strategic and International Studies and McAfee, June 2014. 19McKinsey Global Institute Digital globalization: The new era of global flows © Getty Images 20 McKinsey Global Institute Executive summary SOURCE: UNCTAD; McKinsey Global Institute analysis 1 Estimated from 2011 bilateral services flows data and 2014 services trade data from UNCTAD. NOTE: For cross-border data flows, see Exhibit E2. % of global GDP 0.01–0.05 0.05–0.10 0.10–0.25 0.25–0.50 >0.50
NA
United States.
Canada, and Mexico
LA
Latin
America
AF
Africa
EE
Eastern Europe
and Central Asia
ME
Middle
East
OA
Other
Asia
CH
China
region
WE
Western
Europe
AU
Austral-
asia
NE
Northeast
Asia
% of global GDP
0.02–0.05 0.05–0.10 0.10–0.25 0.25–0.50 0.50–1.00 >1.00
NA
United States
and Canada
LA
Latin
America
AF
Africa
EE
Eastern Europe
and Central Asia
ME
Middle
East
OA
Other
Asia
CH
China
region
WE
Western
Europe
AU
Australasia
NE
Northeast
Asia
GOODS FLOWS
SERVICES FLOWS
1980
100% = $1.8 trillion (18.6% of GDP)
NA WE
LA
ME
AF
CH
AU
EE
NE
OA
10.5x larger2014
100% = $19 trillion (24.6% of GDP)
NA WE
LA
ME
AF
CH
AU
EE
NE
OA
2002
100% = $1.6 trillion (4.9% of GDP)
NA WE
LA
ME
AF
CH
AU
EE
NE
OA
3.1x larger20141
100% = $4.9 trillion (6.4% of GDP)
NA WE
LA
ME
AF
CH
AU
EE
NE
OA
21McKinsey Global Institute Digital globalization: The new era of global flows
SOURCE: IMF CDIS; UN W orld Tourism Organization; McKinsey Global Institute analysis
1 Estimated from bilateral FDI stock data.
NOTE: For cross-border data flows, see Exhibit E2.
% of global GDP
0.02–0.05 0.05–0.10 0.10–0.25 0.25–0.50 0.50–1.00
FINANCIAL FLOWS (FDI)1
PEOPLE FLOWS
NA
United States,
Canada, and Mexico
LA
Latin
America
AF
Africa
EE
Eastern Europe
and Central Asia
ME
Middle
East
OA
Other
Asia
CH
China
region
WE
Western
Europe
AU
Austral-
asia
NE
Northeast
Asia
Million cross-border travelers
<1 1–5 5–10 10–50 >50
NA
United States
and Canada
LA
Latin
America
AF
Africa
EE
Eastern Europe
and Central Asia
ME
Middle
East
OA
Other
Asia
CH
China
region
WE
Western
Europe
AU
Australasia
NE
Northeast
Asia
2002
100% = $0.7 trillion (2.1% of GDP)
NA WE
LA
ME
AF
CH
AU
EE
NE
OA
2.3x larger2014
100% = $1.65 trillion (2.1% of GDP)
NA WE
LA
ME
AF
CH
AU
EE
NE
OA
2002
100% = 650 million
NA WE
LA
ME
AF
CH
AU
EE
NE
OA
1.6x larger2013
100% = 1.03 billion
NA WE
LA
ME
AF
CH
AU
EE
NE
OA
© Getty Images
For decades, the movement of traded goods, services, and finance defined our image
of globalization and deepened the connections between nations. Today, however, those
traditional flows have lost their momentum. At least part of this shift appears to be structural
rather than a temporary cyclical dip. But this does not mean that globalization has moved
into reverse. Enormous streams of data are transmitted across borders every minute, and
they are growing exponentially in both volume and variety. Today globalization is being
accelerated and redefined by flows of data that embody ideas, information, and innovation.
Increasingly, the World Wide Web provides the ties that bind the global economy together.
Consider that some 50 percent of the world’s traded services are already digitized.21
Approximately 12 percent of the global goods trade is conducted via international
e-commerce. Cross-border Skype calls equal 46 percent of the volume of traditional
international calls. Across 18 countries analyzed by eBay, anywhere from 88 to 100 percent
of the SMEs that use its platform are exporters.
By lowering the costs of communication and transactions, digitization opens new
possibilities for conducting business across borders. As digital platforms grow in scale and
sophistication, they are creating more efficient and transparent global markets in which
far-flung buyers and sellers find each other with a few clicks. They provide businesses
with enormous built-in customer bases and effective ways to connect with them—and
they enable even microenterprises to participate directly in global flows. Digital flows are
also shifting globalization into a faster gear as information ricochets around the world and
collaboration spans time zones. Taken together, these shifts create economic value by
increasing innovation, competition, and productivity.
Many of the challenges associated with digitizing national economies are now playing out on
a global scale. Industries are being disrupted by new entrants, value chains are re-forming,
and profit pools are shifting. Much of the value of digitization is going into unpriced benefits
for consumers that are not captured in official GDP statistics. While the digital world may
be more inclusive and closely connected, it is not truly “flat”—and digitization tends to
accentuate disparities.
Digitization changes the economics of globalization in several ways, as we discuss in this
chapter and the ones that follow. First, in contrast to the last era of globalization, we find
that countries on the periphery of the network of global data flows benefit even more than
the digital content producers at the center. In addition, the types of companies involved
are different: instead of waiting for the benefits of globalization to trickle down from large
corporations, SMEs can become micro-multinationals in their own right, and startups can
be “born global.” Finally, digitization is fueling competition as it enables innovative business
models and allows companies to scale up quickly. This chapter brings the new and more
digital version of globalization into focus.
21 Daniel Castro and Alan McQuinn, Cross-border data flows enable growth in all industries, Information
Technology and Innovation Foundation, February 2015.
45X
increase in used
cross-border
bandwidth since
2005
1. A NEW ERA OF
DIGITAL GLOBALIZATION
24 McKinsey Global Institute 1. A new era of digital globalization
GROWTH IN TRADITIONAL FLOWS OF GOODS, SERVICES, AND FINANCE
HAS FLATTENED
In the 20th-century version of globalization, the world built deeper and more intricate ties
as the goods trade and cross-border finance grew in volume and scope. But both types of
flows took sharp tumbles during the financial crisis and the Great Recession. Since then,
global trade bounced back but is now flattening—and capital flows remain at a fraction of
the heights reached during the bubble years. These traditional types of flows still form an
important part of the global economy, but flows of data and information are providing the
real momentum.
Global trade in goods has slowed dramatically since 2008, reflecting
structural shifts
Between 1985 and 2007, the world’s trade in goods grew roughly twice as fast as global
GDP. This reflects major multinationals expanding their supply chains and establishing
new bases of production to tap into the enormous pools of low-cost labor in emerging
economies. Global trade in goods rose from 13.8 percent of world GDP in 1985 ($2 trillion)
to 26.6 percent of GDP ($16 trillion) on the eve of the Great Recession. Since its post-crisis
rebound, however, growth in goods trade has flattened—and it has even receded when
measured relative to GDP (Exhibit 1).
Much of the growth in goods trade since 2000—and much of its subsequent deceleration—
is related to commodity prices. As emerging economies rapidly urbanized and
industrialized, their appetite for raw materials such as steel, copper, and agricultural goods
boosted trade volumes and sent commodity prices soaring to new heights. From 2000
to 2011, the price of many commodities doubled or even tripled. But today the picture is
Exhibit 1
24.6
26.1
21.7
26.6
20.1
13.8
12
14
16
18
20
22
24
26
28
05 20142000 10
-2.0
Global goods trade, 1980–2014
% of GDP
1985 90 95
After decades of steady growth relative to GDP, trade in goods has been declining since its post-recession rebound
SOURCE: UNCTAD; McKinsey Global Institute analysis
2.0
Global trade
$ trillion
GDP growth
rate
%
1995–2005 2005–141985–95
10.63.5 5.2 6.5 15.4 19.0
7.5 6.710.1
Global flows 2
Report
mc 0225
25McKinsey Global Institute Digital globalization: The new era of global flows
remarkably different. Prices have declined sharply over the past few years, and the volume
of commodities being traded has also flattened (Exhibit 2). From June to December 2014,
the price of Brent crude fell from $112 a barrel to $62, and the price of copper has fallen by
half since its peak in 2011.22 We calculate that this slowdown in commodities accounts for
nearly three-quarters of the decline in goods trade as a share of global GDP.
Yet there is more behind the slowdown in global goods trade than a commodities cycle.
Trade in manufactured goods has also been flat to declining for both finished goods and
intermediate inputs. Global container shipping volumes grew by 7.8 percent from 2000 to
2005, but from 2011 to 2014, growth was markedly slower, at only 2.8 percent.23
Multiple cyclical factors have sapped momentum in the trade of manufactured goods.
Many of the world’s major economies—notably China, Europe, and Japan—have been
experiencing slowdowns. China, for example, posted almost 18 percent annual growth in
both imports and exports from 2000 to 2011. But since then its export growth has slowed to
4.6 percent, and imports have actually shrunk.
However, there may be structural reasons in global manufacturing that explain decelerating
growth in traded goods. Our analysis find that global consumption growth is outpacing
trade growth for some types of finished goods, such as automobiles, pharmaceuticals,
fertilizers, and plastic and rubber goods. This indicates that more production is happening
in the countries where the good is consumed. This may reflect the “reshoring” of some
manufacturing to advanced economies as well as increasing consumption in emerging
markets where these goods are produced.
22 Oil prices from US Energy Information Administration data, January 2015. Our analysis extends through 2014,
but commodity prices have continued their sharp decline since then.
23 Data from IHS.
Exhibit 2
The commodities slump partly explains the loss of momentum in goods trade,
but finished and intermediate goods have declined as well
SOURCE: IHS; UNCTAD; McKinsey Global Institute analysis
Value of goods trade, 2002–14
% of world GDP
0
1
2
3
4
5
6
7
8
9
10
11
12
2002
-1.1
201408
0
1
2
3
4
5
6
7
8
9
10
11
12
-0.4
2002 08 2014
0
1
2
3
4
5
6
7
8
9
10
11
12
2014
-0.5
082002
Processed and raw materials Intermediate goods Finished goods
Trade value
$ trillion
4.6 6.1 8.5
26 McKinsey Global Institute 1. A new era of digital globalization
For intermediate goods, declines in trade are more widespread across product categories,
including chemicals, paper, textile fabrics, and communications and electrical equipment.
In fact, the value of trade declined in roughly half of the categories of intermediate
goods between 2011 and 2014 (Exhibit 3). This could indicate that global value chains
are shortening.
The current slowing of trade growth may or may not reverse in the years ahead. The
development of 3D printing has not yet had a clear effect on global trade, but if this
technology is widely adopted by global manufacturers, it could reduce global trade volumes
Exhibit 3
SOURCE: IHS; McKinsey Global Institute analysis
Chemicals
Telephones, microphones, etc.
-16
-72
Pharmaceutical inputs
Fertilizers -11
-10
Vehicle parts
Total increasing
-13Communications equipment
38
Other declining categories
38
Electrical equipment
Total declining
201
Textile fabrics
-67
16
Paper
Other increasing categories
-212
Office machines
75
33
Aircraft parts
-7
Steel products
-8
-8
Trade has declined in half of intermediate goods categories, reflecting shorter global value chains
47%
of categories
have posted
trade declines
since 2011
53%
of categories
have posted
trade increases
since 2011 but
grew more
slowly than
GDP
3.5
13.0
3.5
2.2
13.0
2.8
1.2
1.8
1.2
2.0
0.6
2.3
22.0
Change in trade in categories of intermediate goods products, 2011–14
$ billion
% of intermediate
goods trade, 2014
Declining
categories
Increasing
categories
31.0
NOTE: Numbers may not sum due to rounding.
27McKinsey Global Institute Digital globalization: The new era of global flows
as more products are “printed” where they are consumed. There are already examples
of this at work. Consider GE Aviation, which is beginning to use 3D printing to produce
fuel nozzles for its new Leap engine. A fuel nozzle made the traditional way consists of 20
different components, with a supply chain that spans countries. But 3D printing allows the
company to produce best-quality nozzles in one piece, at one location, eliminating the need
to ship intermediate parts across borders.
Examining a wide range of both R&D-intensive and labor-intensive products, we find
significant potential to transform how—and where—many categories of goods are
produced with 3D printing in the years ahead.24 The applications are particularly relevant
for electronics, vehicle parts, other transportation equipment, machinery and electrical
equipment, medical instruments, and apparel.
Cross-border financial flows have fallen sharply since 2008 and show no sign
of recovery
Cross-border capital flows—which include lending, foreign direct investment, and
purchases of equities and bonds—link national financial markets, connecting borrowers
and savers from different countries. For 25 years prior to the 2008 financial crisis, these
flows grew faster than global GDP, rising from $0.5 trillion in 1980 to $11.9 trillion in 2007
(Exhibit 4).
24 For more on 3D printing technology and its economic potential, see Disruptive technologies: Advances that
will transform life, business, and the global economy, McKinsey Global Institute, May 2013.
Exhibit 4
SOURCE: IMF Balance of Payments; Economist Intelligence Unit; Bank for International Settlements; Institute of International Finance; McKinsey Global
Institute analysis
1 Includes foreign direct investment, purchases of foreign bonds and equities, and cross-border loans and deposits.
2 Includes trade credits, loans, currency, and deposits.
NOTE: Numbers may not sum due to rounding.
Cross-border lending accounts for 70 percent of the drop in global financial flows,
reflecting new banking regulation
6.8
20.7
12.2
4.5
4.1
0
2
4
6
8
10
12
14
16
18
20
22
1980 200720001990 2014E
-14 p.p.
5.7
0.9
2.6
1.7
2.7
1.6
0.9
1.0
2007
11.9
FDI
Equity
5.2
-6.7
Loans2
Bonds
2014
2
-7
-6
-23
Compound
annual
growth rate,
2007–14
%
Global cross-border capital inflows-to-GDP ratio1
%
Global capital inflows by type
$ trillion
28 McKinsey Global Institute 1. A new era of digital globalization
But since their peak in 2007, financial flows have contracted sharply, dropping from
21 percent of global GDP in 2007 to just 7 percent in 2014. Much of the decline is in cross-
border lending. Facing new regulations on capital and liquidity, as well as pressures from
shareholders and regulators to reduce risk, many banks in advanced economies are
winnowing down the geographies and business lines in which they operate. From early
2007 through the end of 2012, commercial banks sold off more than $722 billion in assets
and operations, with foreign operations accounting for almost half of this total.25 There is no
sign of a reversal in this trend, and the sharp decline could indicate a reversion to a longer-
term trend prior to the credit bubble years. Overall, the decline in cross-border lending
explains 72 percent of the total drop in cross-border financial flows since 2007.
Beyond the retrenchment in cross-border lending, international investment flows in bonds,
equities, and FDI are also flat or down. Cross-border bond and FDI flows have declined
41 percent and 35 percent, respectively, in absolute terms between the end of 2007 and the
end of 2014. Cross-border equity flows are essentially flat in value but have also declined
relative to global GDP. Preliminary data for 2015 show that global financial flows declined
further across a broad range of developing countries.26
The only financial flows that have continued to grow since the Great Recession are
remittances sent from global migrants to their home countries. These have grown 7 percent
annually over the past five years and are now worth $583 billion annually. Although steady
in nature, remittances are significantly smaller than equity flows (which totaled $1 trillion in
2014) and bond flows ($1.6 trillion in 2014). Growth in remittances reflects the increasing
flows of migrants and other people flows (see Box 1, “People on the move”).
Global service trade continues to grow, albeit slowly
Global trade in services is a much smaller flow than trade in goods. It has grown slowly but
steadily over the years, rising from some $400 billion in 1985 to approximately $5 trillion
in 2014, or from 3.4 percent to 6.3 percent of global GDP. Compared with the $19 trillion
goods trade, global trade in services remains small. Its compound annual growth rate of
8.8 percent since 1985 has outpaced global GDP growth over that period.
In the past, trade in services often involved people traveling around the world to deliver
expertise, but today financial services, IT support, R&D, engineering and design, and
many other services can be delivered digitally. Emerging economies such as Costa Rica,
India, Morocco, the Philippines, and South Africa, for example, have relied on technology
to build flourishing business process outsourcing industries that offer call center and
technical support services to global clients. Trade in digitally deliverable services has more
than doubled over the past decade, reaching $2.4 trillion in 2014. This amounts to almost
50 percent of total services exports. Advanced economies accounted for 81 percent of total
digitally deliverable service exports in 2014. India and the Philippines were the only emerging
economies ranking in the top ten net exporters of such services.
In the years ahead, the continued expansion of digital technologies, cross-border Internet
connections, and global online marketplaces for freelance services could potentially
increase traded services. Still, compared to the value of global goods trade, the global
services trade is likely to remain a far smaller cousin.
25 Financial globalization: Retreat or reset? McKinsey Global Institute, March 2013.
26 Institute of International Finance data.
7%
financial flows as a
share of world GDP
in 2014, down from
21% in 2007
29McKinsey Global Institute Digital globalization: The new era of global flows
Box 1. People on the move
People are more mobile than ever—and digital
technologies may partly facilitate this trend. We find that
all types of people flows across borders are growing
faster than the global population (Exhibit 5). Roughly a
quarter of a billion people, or 3.4 percent of the world’s
population, lived outside the country of their birth in 2013,
compared with 120 million, or 2.7 percent of the global
population, in 1980. People can now use digital platforms
to find work abroad and then stay closely connected with
friends and family back home through voice over Internet
protocol (VoIP) or Skype calling, instant messaging,
and social media. New platforms can even help with
logistics such as managing foreign bank accounts and
remittances; TransferWise, for instance, offers users
instant international monetary transfers without hefty
currency conversion fees.
Other people have been forced from their homelands by
conflict. After a decade of slight decline, the number of
refugees worldwide jumped from 16.7 million in 2013 to
19.5 million in 2014—a spike that worsened in 2015 with
the escalation of the Syrian refugee crisis.1 Many of these
1 UNHCR global trends report: World at war, UN High Commissioner
on Refugees, June 2015. Note that this number does not include
the 38.2 million people who are internally displaced by war and
persecution, nor the 1.8 million people who are awaiting the
outcome of asylum claims. Both of these numbers are up sharply
from 2013.
recent refugees have been relying on real-time social
media updates to guide their journey (for more on this, see
Chapter 2).
The number of international tourist arrivals hit 1.1 billion in
2014, continuing a trend of steady growth.2 As incomes
rise in emerging economies, the citizens of these
countries are eager to experience in person the world they
have seen online. Recent years have brought a huge influx
of Chinese visitors to destinations ranging from Australia
to the United States to Europe. Digital platforms enable
these flows: online travel sites make it easier than ever for
users to compare and book airfares, while sites such as
Airbnb help them find the exact accommodations that
suit their needs. Having guidance, user reviews, and GPS
mapping at their fingertips has given travelers the ability to
navigate unfamiliar destinations with greater confidence.
Additionally, OECD statistics show that some 4.5 million
international students traveled abroad to study in 2012.
The robust growth rate in this number may be a hopeful
sign that a new generation is embracing the opportunity
to become global citizens in a more mobile world.
2 UN World Tourism Organization statistics. Note that the “people
flows” metric in the MGI Connectedness Index adjusts this number
down to account for individuals making multiple trips within a
given year.
Exhibit 5
50
100
150
200
250
300
350
400
450
10 2014059590851980 2000
All types of people flows are outpacing global population growth
Students1
Refugees
Travelers
World
population
Migrants1
Major change
1 Latest data available are from 2012 for students and 2013 for migrants; 2012–13 growth was used for linear extrapolation to 2014.
SOURCE: OECD; W orld Bank; UN W orld Tourism Organization; UN High Commissioner on Refugees; UN Population Division; McKinsey Global Institute
analysis
1980–90 1990–2000 2000–10 2010–14 1980–2014
Compound annual
growth rate (%)
People flows
across borders
Index:
100 = 1980
1.67.5 -3.5 -1.4Refugees 8.0
4.41.7 8.7 3.0Students 3.8
1.51.8 1.4 1.2World population 1.2
4.24.7 4.6 3.8Travelers 4.9
1.70.2 2.1 2.9Migrants 4.4
Box 1
30 McKinsey Global Institute 1. A new era of digital globalization
DIGITIZATION IS USHERING IN A NEW ERA OF GLOBALIZATION
As of the end of 2015, some 3.2 billion people around the world—accounting for
43.4 percent of the global population—were online.27 The expansion of the Internet,
combined with the introduction of digital platforms and other types of digital tools, has
opened a new chapter in the story of globalization.
Cross-border data flows are the hallmarks of 21st-century globalization. Not only do they
transmit valuable streams of information and ideas in their own right, but they also enable
other flows of goods, services, finance, and people. Virtually every type of cross-border
transaction now has a digital component. Container ships still move products to markets
around the world, but now customers order them on digital platforms, track their movement
using RFID codes, and pay for them via digital transactions. Massive online platforms such
as Alibaba, Amazon, eBay, and Facebook link businesses and customers anywhere in the
world. By reducing the cost of transactions and allowing digital goods, services, and capital
to change hands instantly, digitization is creating a more hyperconnected, hyperspeed era
of global flows.
Cross-border data flows are soaring and connecting more countries
As Internet usage continues to grow within individual economies, users are rapidly forming
and deepening international connections. In 2015, 50 percent of Facebook users had at
least one international friend, up from just 16 percent in 2012. Cross-border used bandwidth
has grown 45 times larger over the past decade. In absolute terms, it has grown from 4.7
terabits per second (Tbps) in 2005 to 211.3 Tbps in 2014, for an annual growth rate of
52 percent.28 Over the next five years, total Internet Protocol (IP) traffic is projected to triple,
while cross-border used bandwidth is projected to post a ninefold increase (Exhibit 6).29
Most international Internet traffic travels via an extensive cable network found on the world’s
ocean floors, running along coastlines and between continents. Cross-border capacity
expanded by 38 percent annually from 2007 to 2014 as new submarine cables were built
and old ones were upgraded. Emerging economies are becoming more integrated into this
network (Exhibit 7). In 2005, 75 countries used more than 1 gigabit per second of cross-
border bandwidth; by 2014, that number was up to 164. Emerging economies started from
a small base, but they have outpaced advanced economies in the growth of used cross-
border bandwidth over the past decade.
27 The state of broadband 2015: Broadband as a foundation for sustainable development, International
Telecommunication Union and UNESCO Broadband Commission for Digital Development, September 2015.
28 TeleGeography, Global Internet Geography.
29 Projections of total IP traffic from Cisco Visual Networking Index: Forecast and methodology, 2014–
2019, Cisco, May 2015; projection of cross-border bandwidth from TeleGeography, Global Bandwidth
Forecast Service.
Virtually every type of cross-border transaction now
has a digital component.
50%
share of Facebook
users with at least
one international
friend
31McKinsey Global Institute Digital globalization: The new era of global flows
Exhibit 6
397
1,020
18E
30
744
19E15E14 20E
1,397
17E
543
211
13
147
16E10
1,914
70
2021E
290
1211
46
0908
11
2005
5 19
0706
7
101
SOURCE: TeleGeography, Global Bandwidth Forecast Service; McKinsey Global Institute analysis
Used cross-border bandwidth, global
Terabits per second
Cross-border bandwidth has grown 45 times larger over the past decade—
and may grow another nine times larger by 2021
ForecastActual
45x
>9x
Exhibit 7
Cross-border data flows are surging and connecting more countries
SOURCE: TeleGeography, Global Internet Geography; McKinsey Global Institute analysis
Used cross-border bandwidth
2005
100% = 4.7 Terabits per second (Tbps)
NA
United States and Canada
LA
Latin America
AF
Africa
EU
Europe
ME
Middle East
OC
Oceania
AS
Asia
Regions
Bandwidth
Gigabits per second (Gbps) <50 50–100 100–500 500–1,000 1,000–5,000 5,000–20,000 >20,000
2014
100% = 211.3 Tbps
NOTE: Lines represent interregional bandwidth (e.g., between Europe and North America) but exclude intraregional cross-border bandwidth (e.g., connecting
European nations with one another).
45x larger
NA
EU
LA
ME
AF
AS
OC
NA
EU
LA
ME
AF
AS
OC
32 McKinsey Global Institute 1. A new era of digital globalization
However limitless the public Internet might seem, the portion of the Internet that has been
indexed and can be navigated by anyone using standard search engines is only the surface
of a much larger structure. The “Deep Web” cannot be accessed in the same way, and
it is estimated to be hundreds of times larger than the public Internet. It includes private
company networks, some enormous publicly accessible topic databases (such as climate
data from the US National Oceanic and Atmospheric Administration), libraries and archives,
private chat rooms, the underlying operations of social media sites and other platforms,
and much more. Private data networks have been growing faster than the public Internet as
technology giants expand their dedicated long-haul networks. The share of private networks
in international used bandwidth has increased from 20 percent in 2009 to 35 percent in
2014.30 Much of the Deep Web is legitimate and benign, but it does have some shadowy
corners, collectively known as the “Darknet,” where criminal trade flourishes.31 This report
measures data flows by analyzing used cross-border bandwidth. It therefore captures traffic
of all types—public and private, legitimate and illicit—since these cannot be disaggregated.
Data flows include a huge variety of business and personal communications, transactions,
information, videos, and other digital media content, gaming, and much more. We also
analyzed cross-border digital calls, which have more than doubled from 274 billion call
minutes in 2005 to 569 billion call minutes in 2014. This rising volume is primarily attributable
to the expanded use of voice over Internet protocol (VoIP) technology. Since 2005, VoIP call
minutes have grown by 19 percent per year, while traditional call minutes have grown by
4 percent. Additionally, cross-border computer-to-computer Skype communications have
soared, with call minutes increasing by some 500 percent over the past five years.32 In 2014,
computer-to-computer Skype call minutes were equal to 46 percent of traditional phone
call minutes.
Data flows—both within countries and between them—reflect the activities of individuals
and of businesses. Many people assume that the Internet is dominated by individuals
viewing YouTube and other streaming videos, trading e-mails, and posting on social media.
But a large share of Internet traffic is also driven by companies interacting with their foreign
operations, suppliers, and customers. The business aspect of data flows is likely to take
on a deeper dimension in the near future as more companies embed monitors, sensors,
and tracking devices into their physical assets. As the Internet of Things is more widely
adopted, Cisco estimates that machine-to-machine connections will account for more than
40 percent of global devices and connections by 2019. It could account for more than half
not long after that (Exhibit 8).33 These connections generate very small and intermittent data
bursts that account for only a small share of IP traffic. But those flows represent a great
deal of economic value for companies since they are directly related to making machines,
processes, and supply chains more efficient.
30 TeleGeography, Global Bandwidth Research Service.
31 For an in-depth discussion of this topic, see Daniel Sui, James Caverlee, and Dakota Rudesill, The Deep Web
and the Darknet: A look inside the Internet’s massive black box, Woodrow Wilson International Center for
Scholars, October 2015.
32 TeleGeography, Global Bandwidth Research Service.
33 Cisco Visual Networking Index: Forecast and methodology, 2014–2019, Cisco, May 2015.
Digital platforms are creating more efficient and
transparent global markets. They provide businesses
with enormous built-in customer bases and effective
ways to reach them.
33McKinsey Global Institute Digital globalization: The new era of global flows
Digitization transforms global flows in three ways
Globalization has a very different look today in part because digitization has introduced
three new phenomena into the equation. First and foremost, large-scale Internet platforms
have driven down the cost of cross-border interactions and transactions. Second, purely
digital goods and services are now traded virtually and instantly. And finally, the addition of
“digital wrappers” to traditional products is enhancing their value.
Digital platforms connect people around the world
Digital platforms include e-commerce marketplaces, operating systems (such as Google’s
Android and Apple’s iOS), social networks (such as Facebook, Instagram, Twitter, WeChat,
and QQ), and digital media platforms (such as YouTube, Uvideos, Spotify, Hulu, and Netflix).
Virtual global marketplaces now match job seekers with employers (LinkedIn), freelancers
with assignments (Upwork), borrowers with lenders (Kiva), creative projects with funders
(Kickstarter), travelers with accommodations (Airbnb), and students with education
providers (Khan Academy).
The biggest platforms are creating truly global markets and user communities on a scale
that has never been seen before (Exhibit 9). Facebook’s monthly active user base, for
example, has surpassed the size of China’s population. As of early 2015, creators filming in
YouTube Spaces have produced more than 10,000 videos that have generated more than
one billion views. Alibaba recorded more than $14 billion in sales on its platforms in just 24
hours during its 2015 “Singles Day” promotion, smashing its record set the previous year.
(See Chapter 2 for more on the scope of individual participation around the world.)
Exhibit 8
4
16
6
43
12
19
SOURCE: Cisco; McKinsey Global Institute analysis
By 2019, machine-to-machine connections are expected to account for more than 40 percent
of global devices and connections
Connections,
2019
PCs TabletsTVsMachine-to-machine
(M2M)
OtherSmartphones
22
0
19
33
3
23
Global devices and connections Global IP traffic by devices
100% =
25 billion
100% =
168 exabytes
per month
34 McKinsey Global Institute 1. A new era of digital globalization
The largest corporations can build their own e-commerce sites or open innovation
platforms. But the biggest and most widely recognized public platforms are open
ecosystems that host an entire universe of diverse participants. E-commerce marketplaces
such as Alibaba, Amazon, eBay, Flipkart, and Rakuten, for example, support millions
of vendors, creating enough product variety and price competition to attract enormous
global customer bases. These platforms give smaller enterprises exporting capabilities by
providing them with payment infrastructure, logistics support, and global visibility.
The influence of e-commerce marketplaces on international trade is significant—and still
growing. Today some 16 percent of B2C e-commerce transactions are cross-border, and
that share is projected to reach almost 30 percent by 2020, when international sales could
hit $1 trillion (Exhibit 10). Cross-border B2B e-commerce is even bigger. In 2014, it was
an estimated $1.8 trillion to $2 trillion market. Together, the roughly $2.2 trillion of cross-
border e-commerce in 2015 is equal to approximately 12 percent of global goods trade.
While growth in the overall goods trade has flattened, the portion enabled by e-commerce
is growing.
The size of these platforms, combined with their use of automated processes driven by
algorithms, lowers the marginal costs for platform operators practically to zero.34 Platforms
make it possible for users to research products, services, prices, and alternative choices.
This removes some information asymmetries so that markets function more efficiently,
although it may disrupt traditional intermediaries in the process.
34 Michael Chui and James Manyika, “Competition at the digital edge: ‘Hyperscale’ businesses,” McKinsey
Quarterly, March 2015.
Exhibit 9
SOURCE: Facebook; Twitter; Freelancer; Upwork; Mashable; Fortune; Statista; McKinsey Global Institute analysis
Active users on select platforms, 4Q15 or latest available
Million
Digital platforms are connecting billions of people around the world
YouTube
1,000
650
Alibaba
407
Twitter Skype Amazon PayPal eBay LinkedIn
1,000
1,590
400
320 300 300
179 162
100
35McKinsey Global Institute Digital globalization: The new era of global flows
For businesses, digital platforms provide a huge built-in base of potential customers and
effective ways to market to them directly and launch new products. As social media exposes
hundreds of millions of consumers from around the world to what is available, products can
launch globally and go viral in unprecedented ways. In the fashion industry, for example,
bloggers, vloggers, Instagram, and Twitter are accelerating trends by highlighting what
celebrities wear (from Beyonce’s “kale” sweatshirt to virtually any outfit Kate Middleton
appears in). In 2012, Michelle Obama wore an affordable red-and-white checked dress from
British online fashion retailer ASOS in a photo that was retweeted 816,000 times on Twitter
and shared more than four million times on Facebook; it instantly sold out.
Digital platforms enable small businesses, entrepreneurs, and individuals to connect across
borders, as we discuss in the next chapter.
Digital goods and services can be delivered instantaneously at very little cost
Today there is growing trade in digital goods and services that have near-zero transmission
costs. McKinsey research shows that global spending on digital media grew at 18 percent
annually from 2008 to 2013, compared with negative 5 percent growth in spending on
Exhibit 10
By 2020, some 940 million online shoppers are expected to spend
almost $1 trillion on cross-border e-commerce transactions
SOURCE: AliResearch; McKinsey Global Institute analysis
0.82.5
0.5
2.8
1.8
15
2.4
1.0
19E 2020E
3.1
2.2
18E
2.1
0.7
3.4
17E
1.9
16E
2.2
0.41.9
2014
0.31.6
1.6
0.2
1.4
1.2
16E
0.5
0.6
19E
1.2
2020E
1.2
0.9
1.2
0.7
17E 18E
1.2
0.8
15
1.11.0
0.3
0.4
2014
1.9
1.6
1.8
2.0
1.3
2.1
1.5
Global B2C e-commerce shoppers
Billion
Global B2C e-commerce transaction volume
$ trillion
Cross-
border
% of
total
+27% p.a.
+21% p.a.
Cross-border
Domestic
Forecast
NOTE: Numbers may not sum due to rounding.
15 16 18 21 24 27 29 23 25 28 33 37 42 45
36 McKinsey Global Institute 1. A new era of digital globalization
traditional media.35 E-books, apps, online games, MP3 music files and streaming services,
software, and cloud computing services can be transmitted instantaneously to customers
anywhere in the world there is an Internet connection.
Major media websites, for example, are building global audiences (Exhibit 11). The Wall
Street Journal’s international site traffic grew from 21 percent of its total online readership
in 2013 to 33 percent in 2015. The international share of readers rose from 29 percent to
50 percent for BuzzFeed, and from 27 percent to 61 percent for Yahoo Sports. Roughly
three-quarters of The Guardian’s online readership is outside the United Kingdom.36 Netflix
expanded its business model from mailing DVDs to selling subscriptions for online streaming
in 2007, and by the end of 2015, it had subscribers in more than 190 countries.
35 The McKinsey global media report defines this as spending on over-the-top (OTT) transactional digital
video, OTT subscription digital video, digital recorded music downloads, digital recorded music-streaming
subscriptions, consumer magazine digital circulation and advertising spend, daily newspaper digital
circulation and advertising spend, electronic consumer books, online video games, mobile video games, and
digital learning material. Traditional media consists of spending on physical home video sales and rentals,
physical recorded music, consumer magazine print circulation and advertising spend, daily newspaper print
circulation and advertising spend, print consumer books, and boxed-console and PC video games, and print
education material.
36 All international site traffic shares based on monthly site-traffic data from SimilarWeb, comparing shares in
October 2013 and October 2015.
Exhibit 11
Online traffic from outside country of origin as share of total traffic1
Digital media is attracting global audiences
SOURCE: Similar W eb; McKinsey Global Institute analysis
1 Based on monthly site traffic data from Similar W eb.
October 2015October 2013
91
77
73
27
40
27
29
24
22
24
21
22
78
56
New York Times 25
39
41
50
60
33
75
92
61
41
Business Week
Wall Street Journal
Vogue
Time
BBC News
Financial Times
The Economist
Netflix
Buzzfeed
The Guardian
Yahoo Sports
37McKinsey Global Institute Digital globalization: The new era of global flows
Companies can take advantage of digital platforms to create international buzz and
momentum for global product launches. In 2015, Adele’s song “Hello” racked up 50 million
views on YouTube in its first 48 hours, for instance. The following month, her smash album
25 was made available on CD and as a digital download (although not through streaming
services) featuring the song. It sold 900,000 downloads on the first day, and during its first
week of release, it was No. 1 on the download list of iTunes stores in 110 countries.37 In the
United States, 3.38 million copies of the album sold in the first week, the most since Nielsen
began tracking point-of-sale music purchases in 1991.
“Digital wrappers” enhance and enable other types of flows
Adding a digital component to traditional types of flows can enhance their value. Automotive
manufacturers are racing to develop “connected cars,” with features ranging from voice
recognition and smartphone functionality to preventive maintenance alerts, hazard reaction,
and self-driving capabilities. Logistics companies are using sensors, data, and software to
track physical shipments, reducing the volume of goods lost in transit. FedEx, for example,
allows customers to monitor the progress of packages almost continuously by placing small
tracking devices in them; it also uses sensors to monitor temperature, humidity, barometric
pressure, and light exposure for sensitive cargo.38
This type of continuous data availability is invaluable for companies that operate long and
complex supply chains. One study found that RFID technology can help to reduce inventory
costs by up to 70 percent while improving efficiency. Case studies in Germany, including the
logistics centers of Hewlett-Packard and BMW, found that the technology reduced losses
in transit by 11 to 14 percent.39 Rio Tinto, for example, transmits data continuously from
its mines, processing plants, and vehicle fleets around the world to “excellence centres”
located in Brisbane, Australia.
Online user-generated reviews and ratings are another type of digital wrapper. They give
many individuals the comfort level they need to conduct a cross-border transaction, whether
they are buying a consumer product on Amazon or booking a hotel room thousands of
miles away on Airbnb, Agoda, or TripAdvisor. TripAdvisor, for example, has more than
250 million reviews and opinions from travelers on more than 5.2 million businesses and
properties globally. A recent survey by the UN World Tourism Organization revealed that 70
to 92 percent of travelers in various advanced economies considered online guest reviews
important or very important for their hotel booking decisions—meaning that these reviews
influence billions of dollars in cumulative spending.40
37 Clarisse Loughrey, “Adele’s new album 25 is No. 1 on iTunes in almost every country in the world,” The
Independent, November 26, 2015.
38 Disruptive technologies: Advances that will transform life, business, and the global economy, McKinsey Global
Institute, May 2013.
39 Aysegul Sarac, Nabil Absi, and Stéphane Dauzère-Pérès, “A literature review on the impact of RFID
technologies on supply chain management,” International Journal of Production Economics, volume 128,
number 1, November 2010.
40 Online guest reviews and hotel classification systems: An integrated approach, UN World Tourism
Organization, 2014.
50M
YouTube views of
Adele’s “Hello” in
the first 48 hours
38 McKinsey Global Institute 1. A new era of digital globalization
DIGITAL GLOBALIZATION IS CREATING NEW CHALLENGES AS WELL AS NEW
CHANNELS FOR GROWTH
Just as digitization is transforming individual economies and the business models of
individual companies around the world, it is altering the broader global economy—and these
shifts are even bigger and more complex since they are taking place across nations with
different factor costs, levels of development, and regulatory regimes.
Opening to all types of flows, and particularly data flows and global platforms, has the
potential to disrupt traditional industries even as it creates new channels for growth. (For
more on the issue of jobs, see Chapter 3.) The ease of comparison shopping on digital
platforms, for instance, encourages companies to compete on price. This shift works in the
consumer’s favor but creates pressure on the bottom line for companies. It is becoming a
more competitive world in other ways as well. Digital platforms enable small enterprises and
foreign competitors to move into new markets, and technology-powered companies are
demonstrating the ability to add new business lines with ease.
Information can be transmitted halfway around the world in the blink of an eye, but so can
disruptions. Students can educate themselves online from anywhere on earth, but their view
into how other societies live can heighten their impatience with bleak job prospects at home.
Social media creates global communities in a positive sense, but it also allows networks of
extremists to form and strategize.
A world that runs on data flows is also more vulnerable to grid failures and cybercrime. One
study has estimated that cybercrime costs the global economy some $400 billion in annual
losses; these can include consumer data breaches, financial crimes, market manipulation,
and theft of intellectual property.41 This is not only a business risk; it can even pose
national security risks. The issue of combating global cybercrime is discussed more fully in
Chapter 5.
THE WORLD IS STILL FAR FROM FLAT
The Internet has enabled the creation of global markets, but it cannot fully erase the barriers
of geography. Distance still matters. A significant share of each of the major types of global
flows move within well-established regions rather than between them. This is particularly
true of people and data flows. But even in the goods trade, where transportation costs
have fallen dramatically over the past 25 years, 35 percent of global trade is still regional
(Exhibit 12).
41 Net losses: Estimating the global cost of cybercrime, Center for Strategic and International Studies and
McAfee, June 2014.
Information can be transmitted halfway around the
world in the blink of an eye, but so can disruptions.
39McKinsey Global Institute Digital globalization: The new era of global flows
A look at each country’s major trading partner (including exports and imports) reveals that
for many countries, nearby neighbors are their largest trade partners. Canada trades heavily
with the United States, Argentina and Brazil are major partners, and Germany trades with
the rest of Europe (Exhibit 13). There are exceptions to this pattern, however. Notably, China
has become the largest trade partner for large swaths of Latin America and Africa, while the
United States is China’s largest trade partner.
In e-commerce, the distance separating buyers and sellers reduces the volume of
transactions that are completed, although to only about half the degree seen in physical
trade. This effect may be due to the trust factor, language barriers, the presence of familiar
payment systems, and shipping costs. A recent study based on more than ten billion online
transactions found that a 0.5 percent increase in the distance between the two countries
lowers the volume of online trade between them by 1 percent. Sharing a land border also
significantly affects the volume of trade between two countries.42
The Internet itself is not a seamless global web, in part because the whole world is not
connected. Six billion people do not have high-speed broadband, almost four billion do not
have any Internet access at all, and nearly two billion do not have a mobile phone. Digital
divides persist across income, age, geography, and gender. In Africa, the richest 60 percent
are almost three times as likely to have Internet access as the bottom 40 percent, and the
young and urban have more than twice the access of older and rural citizens.43 These gaps
42 Bo Cowgill and Cosmina Dorobantu, Worldwide gravity in online commerce, August 2014.
43 World development report 2016: Digital dividends, World Bank, January 2016.
Exhibit 12
35
65
While much of the world’s trade in goods is long distance,
roughly half or more of other global flows move within the same region
Distribution of flows between intraregional (short haul)
vs. interregional (long haul), 20141
% of world flow
SOURCE: UNCTAD; UN W orld Tourism Organization; TeleGeography, Global Internet Geography; IMF; McKinsey Global
Institute analysis
1 For goods, services, FDI, and travelers we have divided the world into 10 regions; for data flows we have used
TeleGeography’s six regions.
2 Distribution of services flows for 2014 estimated based on 2011 data; 2013 bilateral traveler data used for people flows.
NOTE: Numbers may not sum due to rounding.
36
64
Long haul (interregional)
Short haul (intraregional)
53
47
54
46
67
33
Services2 FDI
Data
Goods
People2
40 McKinsey Global Institute 1. A new era of digital globalization
are hard to erase. Consider that electricity has been in use for more than a century—and yet
some 1.2 billion people, or 17 percent of the global population, still lacked electricity as of
2013.44
Even among those who are connected, cross-border Internet traffic tends to be regional in
nature. In 2014, two-thirds of used cross-border bandwidth was intraregional, reflecting to
some degree the fact that the world’s digital networks have a hub structure. Some academic
research has characterized the Internet not as a global web but as a series of small worlds.
Barnett and Park examine the network structure of the global Internet by looking at
international hyperlink connections, cross-border bandwidth, and shared website use (from
a website and national perspective). They find that the global Internet is very concentrated
among a few core countries that serve as hubs for broader regions drawn together by
shared language, cultural similarities, and historical ties. Their analysis of shared website
use (web perspective) resulted in a Gini coefficient of .930, indicating that the Web-based
network is very centralized among a small group of core countries: the United States, the
United Kingdom, China, Germany, Brazil, France, India, Italy, Japan, Spain, and Russia.45
44 World energy outlook 2015, International Energy Agency, November 2015.
45 George A. Barnett and Han Woo Park, “Examining the international Internet using multiple measures: New
methods for measuring the communication base of globalized cyberspace,” Quality and Quantity, volume 48,
issue 1, January 2014.
Exhibit 13
China, the United States, or Germany is the major trading partner for most countries
SOURCE: UNCTAD; McKinsey Global Institute analysis
Largest trading partner in goods (exports and imports combined), 2014
NOTE: Data omitted for some small nations as indicated in gray.
41McKinsey Global Institute Digital globalization: The new era of global flows
Another reason that the Internet is not yet fully global is that content production is
concentrated in a few advanced economies. Hollywood, for example, has ruled the global
movie box office and the world’s television screens for decades, creating entertainment
that attracts hundreds of millions of viewers worldwide. The United States continues to
play that role in a more digital world. It accounts for more than 50 percent of online content
consumed in all regions except Europe (Exhibit 14).46 The Internet is a democratizing force in
many regards, but some of the legacy structures and disparities that exist in the offline world
persist online.
Still, the rise of cross-border data flows and a truly global Internet infrastructure is still in its
early days. Just 15 years ago, cross-border data flows were negligible. As the underlying
infrastructure continues to expand and as more users around the world join, the barriers of
distance, languages, and cultural norms could erode, creating a more unified world.
•••
Immense flows of goods, services, finance, people, and data move across the world’s
borders, creating a more global world. The pervasiveness of Internet connectivity and
the spread of digital technologies enable data and information to travel around the world
instantaneously, and this capability is transforming all other types of flows. These forces
have unleashed an accelerating wave of change and intensifying global competition. As
digital platforms create new global marketplaces, they are making globalization a more
inclusive phenomenon. Chapter 2 will discuss how individuals and small businesses—in
advanced and emerging economies alike—are using these digital platforms to form their
own global connections.
46 TeleGeography, Global Internet Geography.
>50%
US share of digital
content consumed
in all regions
except Europe
Exhibit 14
88
66
56
51
36
10
16
32
17
61
18
12
32
3
United States
and Canada 2
Latin America
Europe
Africa
Asia Pacific
SOURCE: TeleGeography, Global Internet Geography; Pingdom; McKinsey Global Institute analysis
Content host
42
15
12
31
100% =
1 million
websites
Europe Asia Pacific OtherUnited States1
The United States is the largest producer of digital content for Internet users across the globe
Location of top 100 websites requested by users
% by user region, as of April 2015
Hosting location of
top 1 million websites, 20132
%
1 Includes United States and Canada for location of top 100 websites requested by users.
2 Based on Pingdom analysis of Alexa top 1 million websites.
© Getty Images
Globalization is no longer just the purview of the world’s largest multinational corporations.
Today digitization has erased many of the barriers that once prevented small and medium-
sized enterprises (SMEs), entrepreneurs, and ordinary citizens from making cross-
border connections.
Companies once had to grow to substantial size before they could afford the resources
needed to export, but digitization has dramatically reduced the minimum scale required
to do business across borders. Small businesses are joining the biggest e-commerce
marketplaces to connect with customers and suppliers anywhere in the world. Capital is
available for microenterprises on platforms such as Kickstarter, where close to 3.3 million
people representing nearly all countries made pledges in 2014.47 More than nine million
freelancers from 180 countries have connected with clients on Upwork for assignments
such as web development, graphic design, and marketing.48
The more inclusive nature of digital globalization has significant implications for businesses
and economies, particularly in developing countries. In these nations, companies and
individuals can use digital platforms as a way to overcome constraints in their local markets
and tap into global customers, suppliers, financing, and talent. Instead of waiting for the
benefits of globalization to trickle down from large corporations, SMEs can become micro-
multinationals in their own right, and startups can be “born global.” Individuals can discover
opportunities, information, and ideas from anywhere in the world.
SMALL AND MEDIUM-SIZED BUSINESSES ARE BECOMING
MICRO-MULTINATIONALS
The ability of SMEs to reach global audiences supports economic growth. Digitization
has empowered many to transform themselves into “micro-multinationals.” Digital
platforms provide small firms with “plug-and-play” infrastructure and the opportunity to put
themselves in front of an enormous built-in global customer base.
Consider all of the tools and platforms that a small Chinese manufacturer has at its
disposal when it becomes a Taobao merchant. The company can open and customize
a Taobao “storefront” for free using a mobile app and upload its merchandise for sale. It
can communicate with customers using an instant messaging service, handle payments
through Alipay, choose an Alibaba-affiliated logistics company for shipping, place targeted
digital ad buys through Alimama, and get a small loan instantly from an Alibaba microfinance
subsidiary that can evaluate the merchant’s credit based on data about its business
47 Kickstarter, 2014: Year in Review presentation.
48 Elance-oDesk annual impact report, 2014.
Instead of waiting for the benefits of globalization
to trickle down from large corporations, SMEs and
startups can go global in their own right.
2. DIGITAL PLATFORMS OPEN
THE DOOR TO NEW PARTICIPANTS
44 McKinsey Global Institute 2. Digital platforms open the door to new participants
performance on the platform. Finally, the company can use Alibaba itself to buy supplies and
professional services.49
Similarly, eBay has been helping merchants sell internationally by offering features such as
the ability to be featured on eBay sites in other countries, a global shipping program, and
the option to clear transactions with PayPal. One study found that more than 35 percent of
the top 1,000 eBay sellers have significant cross-border trade, with premium or featured
eBay stores in other countries.50 The company’s own analysis across select emerging and
advanced economies shows that the share of SMEs that export is sharply higher on eBay’s
platform than among offline businesses of comparable size (Exhibit 15). Small businesses
can use platforms to reach a greater number of markets: in China, South Korea, Indonesia,
Thailand, and South Africa, 90 percent or more of eBay sellers export to more than ten
international markets.51
PayPal enables cross-border transactions by acting as an intermediary for SMEs and their
customers. Participants from emerging economies are senders or receivers in 68 percent
of cross-border PayPal transactions. PayPal also helps facilitate small transactions: the
average point-of-sale transaction using a foreign credit card was $169 across four emerging
economies in 2013, while a sample of PayPal data from the same year suggests an average
49 “Alibaba defined,” Alibaba corporate website. See also China’s e-tail revolution: Online shopping as a catalyst
for growth, McKinsey Global Institute, March 2013.
50 Andy Geldman, “The world’s top eBay sellers,” Web Retailer, February 7, 2014.
51 Small online business growth report: Towards an inclusive global economy, eBay Public Policy Lab,
January 2016.
Exhibit 15
Share of eBay commercial sellers1 and offline enterprises that export, 2014
%
eBay enables SMEs to attain global reach that comparable offline businesses have not achieved
SOURCE: eBay; W orld Bank Enterprise Surveys (using latest data available); McKinsey Global Institute analysis
100100100100
95
100
18
14
8
3
10
22
MexicoIndonesiaIndiaChina South
Africa
Brazil
100
9697
20
16
4
South
Korea
GermanyUnited
States
Emerging economies Advanced economies
eBay sellers
All enterprises
1 eBay commercial sellers are defined as sellers with sales of over $10,000 and at least 10 transactions in previous year.
45McKinsey Global Institute Digital globalization: The new era of global flows
transaction sent to emerging economies of just $38. Alipay performs a similar function for
Taobao merchants, providing a critical element of trust needed to facilitate transactions.
SMEs worldwide are joining e-commerce marketplaces to access business resources and
reach new markets. Amazon now hosts two million third-party sellers, while some ten million
small businesses have become merchants on Alibaba platforms.52 Artisans and customers
from around the world find each other on Etsy, a marketplace for handcrafted and vintage
goods; nearly 30 percent of its gross merchandise sales are international.53 More than
20,000 independent designers and artists showcase their work on Pinkoi, a Taiwan-based
online marketplace. The company has connected with customers in more than 47 countries,
using Facebook to expand its reach throughout the Asia-Pacific region.
Cross-border B2B e-commerce sales were approximately a $1.8 trillion to $2 trillion market
in 2014. By 2020, cross-border e-commerce sales to consumers are projected to hit
$1 trillion, accounting for almost 30 percent of total B2C sales.
For businesses, the biggest social media platforms represent a huge base of potential
customers with built-in ways to reach them effectively and directly. Facebook estimates
that more than 50 million SMEs are on its platform, up from 25 million in 2013, and some
30 percent of their fans are cross-border (Exhibit 16). To put this number in perspective,
consider that the World Bank estimated there were 125 million micro, small, and medium-
sized enterprises in the 132 countries in its database in 2010.54 This points to the importance
of social media exposure as a crucial marketing tool, particularly for companies that hope to
raise their global profile.
52 Amazon.com company facts, corporate website; Jack Ma, “America’s online sales opportunity in China,” The
Wall Street Journal, June 8, 2015.
53 2015 third-quarter financial results, Etsy.
54 Khrystyna Kushnir, Melina Laura Mirmulstein, and Rita Ramalho, Micro, small, and medium enterprises around
the world: How many are there, and what affects the count? World Bank/IFC, 2010.
~10M
merchants operate
on Alibaba
Exhibit 16
50
30
25
201520142013
SOURCE: Facebook; McKinsey Global Institute analysis
50 million SMEs use Facebook to find customers, and 30 percent of their fans are from
other countries
Domestic
70
30
Cross-border
Estimated number of SME pages
on Facebook
Million
Share of SME fans that are cross-border
%
+41% p.a.
46 McKinsey Global Institute 2. Digital platforms open the door to new participants
The increasing ability of SMEs to participate directly in globalization is a relatively new
phenomenon, but it is starting to show up in national statistics. It is most clearly seen in the
United States, where the share of exports by large multinational corporations dropped from
84 percent in 1977 to 50 percent in 2013.55 Companies with fewer than 500 employees
accounted for 97.8 percent of all identified US exporters and 97.2 percent of all identified US
importers in 2011. The number of US exporting entities with fewer than 50 employees, in
particular, has grown more rapidly than firms with 50 to 500 employees.56
An analysis of export data for 16 OECD countries shows mixed evidence of increased SME
participation. Between 2005 and 2012, the SME share of total exports grew in ten of the
countries, including the United States and France. But SMEs lost ground in exports in the
remaining six countries (however, in Portugal and elsewhere, this was likely due to tightening
access to credit for small businesses during a prolonged financial crisis).
MANY TECHNOLOGY-BASED STARTUPS ARE NOW “BORN GLOBAL”
The ability to connect globally opens new entrepreneurial possibilities for individuals.
Anyone with a connection and a great idea can launch a business, drawing on the availability
of enterprise software and cheap computing power in the cloud. Academic literature has
highlighted the emergence of a new wave of global startups and SMEs that are making the
most of these types of Web 2.0 tools to innovate. These capabilities are highly relevant for
working with collaborators, customers, and partners in different countries.57
Today’s digitally powered startups can be born global—attracting users, hiring talent,
purchasing inputs, securing funding, and finding mentors across borders from day one.
Consider coModule, an Estonian startup that created technology that brings the Internet
of Things to electric bikes and scooters. Its prototype was unveiled in Barcelona, its seed
funding came from Germany, and its components are sourced from China. The company is
scaling up production and eyeing user markets across Europe and Asia.
MGI surveyed 271 startups worldwide through a partnership with 1776, a global incubator
and venture fund. By working with 1776’s Challenge Cup competition and its Startup
Federation program, we were able to expand the reach of the survey to 19 countries. The
businesses surveyed included members of the Startup Federation, the Global Accelerator
Network, and current and former competitors at 1776’s Challenge Cup events around
the world. (See the technical appendix for more detail on the survey.) While these startups
represent a more globally connected and tech-savvy selection than the typical small
business, the results suggest that even the smallest and youngest enterprises can execute
a global vision if their business model is built on digital technologies. This is a relatively new
development. When many of today’s Internet giants were started, they focused on the US
market alone for a significant period before expanding to other countries. Today, many
digital-based startups market to a global audience from their inception.
55 US trade in goods associated with US multinational corporations from US Bureau of Economic Analysis.
56 “A profile of US importing and exporting companies, 2010–2011,” US Census Bureau press release, April
5, 2013.
57 Jim Bell and Sharon Loane, “‘New-wave’ global firms: Web 2.0 and SME internationalization,” Journal of
Marketing Management, volume 26, issue 3–4, April 2010.
Digitally powered startups can be “born global,”
connecting with international customers, suppliers,
capital, and mentors from day one.
47McKinsey Global Institute Digital globalization: The new era of global flows
A surprising 86 percent of survey respondents pointed to at least one cross-border activity
(Exhibit 17). Almost two-thirds have customers or users in other countries, and almost half
reported sourcing talent from other countries. The rate of cross-border participation varies
widely by company stage. Companies in the growth and scaling phase report more than
twice as many cross-border activities as companies in the concept phase.
Moreover, the surveyed startups from emerging economies were more global across several
dimensions than their counterparts in advanced economies (Exhibit 18). Specifically, they
were more likely to report using foreign inputs, participating in an international accelerator or
incubator program, and having international customers. Those based in South America had
notably more global business activities than those in North America.
This disparity underscores the importance of global digital platforms for startups and small
businesses seeking to overcome limited domestic markets and credit constraints. But the
surveyed company founders reported a number of reasons for seeking out global ties.
MPOWER, a US-based student loan financing company for international students, set out
to solve a problem specific to users from other countries: the inability to get loans from
traditional banks in the United States. Others, such as South Africa–based Anaso Diabitiz,
a diabetes management platform, find that their product is discovered by international
consumers through social media. As the product took off in its home market of South
Africa, patients in Nigeria and Kenya learned about it through Facebook and Twitter—and
today two-thirds of the subscribers are from across the broader African continent and the
Middle East.
Exhibit 17
SOURCE: MGI Global Startup Survey 2015; McKinsey Global Institute analysis
14
36
36
39
47
62
Mentors or advisers
in other countries
Customers, clients, or
users in other countries
Funding or investors
from other countries
Talent (including freelancers)
hired from other countries
Accelerator or incubator
in other countries
Inputs sourced from other
countries (e.g., parts,
cloud services)
86 percent of the tech-enabled startups surveyed by MGI engage in at least one cross-border activity
Share of startup respondents
engaged in cross-border activity
%
Participation by activity type
%
100% = 271 respondents
>1
None
21
14
6
7
13
20
18
Average =
2.4
86%
1
2
3
4
5
6
7
Number of cross-
border activities
48 McKinsey Global Institute 2. Digital platforms open the door to new participants
Whether it is deliberate or serendipitous, international cross-pollination is likely to continue
and accelerate in a more digitally connected global world. But while digital platforms are
enablers for startups, they do not negate the need for local ties and personal connections
based on trust. Our surveyed startups continue to find talent in low-tech ways (with almost
three-quarters relying on personal referrals), and they rated local mentors as more effective
than foreign mentors.
INDIVIDUAL PARTICIPATION HAS GROWING ECONOMIC AND
SOCIAL IMPORTANCE
Thanks to social media and other Internet platforms, individuals are forming cross-border
connections as well. Some 29 percent of Skype call minutes are cross-border, for instance,
compared with less than 2 percent of standard telephone call minutes. Some 92 percent
of online project work on Freelancer.com is cross-border, while only 16 percent of the
workforce in the United States and 7 percent of the workforce in the European Union is
foreign-born. Some 37 percent of Khan Academy students are from foreign countries, while
only 8 percent of all tertiary students in the OECD are foreign-born.
We estimate that nearly one billion individuals around the world are direct participants in
some form of globalization (Exhibit 19). MGI analysis of international ties on Facebook,
Twitter, LinkedIn, and WeChat shows that some 914 million people have at least one
international connection on a social media platform (even adjusting for overlap between
users of these platforms). Each year, some 360 million people participate in cross-border
Exhibit 18
SOURCE: MGI Global Startup Survey 2015; McKinsey Global Institute analysis
1 The difference in participation rates between developed and emerging startups is statistically significant. P-values for inputs (p = 0.005), incubator/accelerator
(p = 0.05), and customers (p = 0.07) are significant at the 0.07 level and below.
2 Emerging economies represented in the survey are Argentina, Armenia, Brazil, China, Colombia, Cyprus, Egypt, Estonia, Hungary, India, Israel, Jordan,
Kenya, Mexico, Moldova, Pakistan, Poland, Russia, South Africa, and Ukraine. Advanced economies represented in the survey are Australia, Canada,
Germany, Ireland, the United Kingdom, and the United States.
Tech-based startups from emerging economies were more likely than other survey respondents
to report seeking out customers and inputs from abroad
Share of startups participating by activity type1
%, 100% = 271 respondents
Customers, clients, or
users in other countries
77
54
Inputs (e.g., cloud services)
sourced from other countries
45
32
Participation in foreign
accelerator or incubator
17 12
Emerging2
Advanced2
49McKinsey Global Institute Digital globalization: The new era of global flows
e-commerce.58 Forty-four million individuals provide services to clients in other countries
on the biggest digital marketplaces for freelance work.59 We estimate that some 12.6 million
people take part in online courses developed and offered in other countries.60
In addition to facilitating cross-border e-commerce, the biggest digital platforms have
built global communities that generate tremendous flows of personal communication,
information, news, and content. Facebook averaged more than a billion daily active users
in December 2015, and Google processes some 3.5 billion searches a day.61 Tencent’s
WeChat instant messaging platform now has 549 million monthly active users—a number
58 Cross-border B2C e-commerce market trends, AliResearch and Accenture, June 2015.
59 Rate observed from site traffic on Freelancer.com applied to total participation in the major platforms
described in Siou Chew Kuek et al., The global opportunity in online outsourcing, World Bank and Dalberg
Global Development Advisors, June 2015.
60 We sum the total membership of leading educational platforms (Coursera, Khan Academy, edX, and Udacity)
and assume the Khan Academy cross-border rate of 37.2 percent for all.
61 Corporate websites and Internet Live Stats.
Exhibit 19
Individuals are participating in globalization, and 914 million have cross-border social media connections
SOURCE: Facebook; AliResearch; US Department of Commerce; OECD; W orld Bank; McKinsey Global Institute analysis
Students
studying abroad
5 million
Cross-border
e-commerce shoppers
361 million
International
travelers
429 million
People living outside
home country
244 million
Cross-border
online students
13 million
Cross-border
online workers
44 million
Social networking users with
at least one foreign connection
914 million
NOTE: Numbers adjusted to account for overlap between platforms and for individuals making multiple international trips in the same year.
DUPLICATE
from ES
50 McKinsey Global Institute 2. Digital platforms open the door to new participants
that is beginning to approach the population of the entire ASEAN region.62 YouTube has
more than one billion users, and 300 hours of video are uploaded every minute.63
Furthermore, social media plays an increasingly important role in connecting people in
developing nations to the rest of the world, opening new avenues for work, learning, and
communication. The share of Facebook users with cross-border friendships is higher in
emerging countries than in developed countries (54 percent vs. 44 percent). It is growing
rapidly, having increased by 3.7 times since 2012 (Exhibit 20). Almost half of those followed
by Twitter users in emerging economies are from other countries, compared with just under
40 percent in advanced economies. This could be because social media users in emerging
economies would already tend to be among that country’s most tech-savvy and globally
minded segment of the population. But it also speaks to the point that people in developing
countries are embracing social media platforms as their link to the rest of the world.
Virtual connections are changing the way people interact with friends, family, and even
strangers across borders—and they sometimes spill over into the physical world. In the
24 hours after the 2015 terrorist attacks in Paris, 4.1 million people in and around the
city marked themselves as safe through Facebook’s Safety Check feature, triggering
notifications to 360 million users around the world that their friends and family were
unharmed.64 Earlier that year, after a major earthquake struck in Nepal, Facebook
implemented a special “donate” button that brought in more $15 million in contributions to
relief efforts from 770,000 people in more than 175 countries.65
62 Tencent corporate financial statement, first quarter 2015.
63 YouTube corporate website.
64 Robinson Meyer, “One small worry about Facebook Safety Check,” The Atlantic, November 18, 2015.
65 Ken Yeung, “Over 770K Facebook users donated $15M to support Nepal earthquake relief,” VentureBeat,
September 28, 2015.
Exhibit 20
The share of Facebook users with at least one international friend tripled in just three years,
with the fastest growth in emerging economies
SOURCE: Facebook; W orld Bank; UNCTAD; McKinsey Global Institute analysis
% of monthly active Facebook users with at least one international friend1
0
10
20
30
40
50
60
+35 p.p.
201514132012
0
10
20
30
40
50
60 +40 p.p.
2012 201513 14
0
10
20
30
40
50
60
+26 p.p.
2012 13 14 2015
World Emerging economies Advanced economies
1 W eighted by number of Internet users to arrive at world average and averages for emerging and advanced economies.
51McKinsey Global Institute Digital globalization: The new era of global flows
Of course, there are limits to the depth and endurance of online connections. Oxford
anthropologist and psychologist Robin Dunbar famously concluded that the maximum
number of casual friendships the human brain can sustain is around 150, while the average
Facebook user has 388. His findings have been borne out by subsequent researchers
applying his theory to the world of social media.66 A study of one million Coursera users
taking massive open online courses offered by the University of Pennsylvania concluded
that only around half of those who register for a course view any lectures, and the average
completion rate was just 4 percent.67 And while social media can be an effective political
organizing tool, it can also be a vehicle for what has been termed “clicktivism,” in which large
numbers of people join a cause by liking, tweeting, or signing an online petition but are not
motivated to take real action.
Despite these caveats, there are in fact ways in which online connections can translate into
tangible economic impact. Digital platforms such as Expedia, TripAdvisor, Yelp, Agoda,
and many more facilitate leisure travel and tourism. Founded in 2008, Airbnb has quickly
built a network of hosts in 34,000 cities in more than 190 countries. Within 40 days of the
United States and Cuba reopening ties, the platform was listing some 2,000 properties on
the island.68 Couchsurfing.com has ten million members around the globe; its network of
free homestays gives travelers on a tight budget the means to see the world and take part in
face-to-face cultural exchanges.69
Digital networks may also facilitate migration. Online talent platforms aggregate data on
candidates and job openings across broader regions and even globally, enabling users to
find international career opportunities.70 As social media use becomes more ingrained into
daily life around the world, expatriates have a way to stay in closer touch with family and
friends in their homeland, reducing the social and emotional hardship of moving to a new
country. Social media can also notify migrants and refugees of changes to government
policies. When a government agency in Nuremburg tweeted in German that it would loosen
immigration protocols for Syrian citizens, the news went viral and led to a rush of 20,000
refugees over the German border within a week.71
Digital platforms offer individuals new ways to learn, collaborate, and develop new skills—
and to showcase their talents to potential employers. Some 44 million people worldwide
find freelance work on Freelancer.com, Upwork, and other digital platforms, while nearly
400 million have posted their professional profiles on LinkedIn. Individuals with creativity and
drive can use digital platforms to propel themselves onto a global stage in ways that would
have been unimaginable in the pre-digital world (see Box 2, “The YouTube economy”).
66 R. I. M. Dunbar, “Coevolution of neocortical size, group size, and language in humans,” Behavioral and Brain
Sciences, volume 16, issue 4, December 1993. See also Nicole Ellison, “Connection strategies: Social capital
implications of Facebook-enabled communication practices,” New Media & Society, volume 13, number
6, September 2011; Bruno Goncalves, Nicola Perra, and Alessandro Vespignani, “Modeling users’ activity
on Twitter networks: Validation of Dunbar’s number,” PLOS One, volume 6, number 8, 2011; and Maria
Konnikova, “The limits of friendship,” The New Yorker, October 7, 2014.
67 Laura Perna et al., “The life cycle of a million MOOC users,” presented at the MOOC Research Initiative
Conference in Austin, Texas, December 5, 2013.
68 Airbnb corporate website. See also Mark Scott, “What Uber can learn from Airbnb’s global expansion,” The
New York Times, July 7, 2015.
69 Couchsurfing.com, About Us.
70 A labor market that works: Connecting talent and opportunity in the digital age, McKinsey Global Institute,
June 2015.
71 Andrea Thomas, Matt Bradley, and Friedrich Geiger, “Obscure German Tweet helped spur migrant march from
Hungary,” The Wall Street Journal, September 10, 2015.
190+
countries with
Airbnb listings
52 McKinsey Global Institute 2. Digital platforms open the door to new participants
Box 2. The YouTube economy
YouTube is proving to be a powerful marketing tool for global brands and
established stars, but it can also launch individuals from obscurity to
opportunity. Consider Havard Rugland, a Norwegian whose dreams of playing
American football once seemed far-fetched. But his mind-boggling YouTube
video of his kicking skills received three million hits, earning him tryouts and an
NFL signing (if not yet stardom).
A number of self-made YouTube stars have created channels that attract
millions of subscribers—and millions in ad revenue. While many adults may
not recognize their names, teenagers certainly do. A recent survey by Variety
asked teens to rank celebrities on various measures of influence, and eight
of the top ten slots went to YouTube creators rather than to more mainstream
media stars.1
Felix Kjellberg, the Swedish Internet sensation known as “PewDiePie,”
became a sensation after posting videos of himself playing video games. He
eventually racked up some 40 million subscribers to his YouTube channel and
millions in annual earnings. Now he and some of his fellow YouTube stars will
be showcased on Revelmode, a digital network backed by Disney-owned
Maker Studios.2 DisneyCollectorBR, one of the most popular channels on
YouTube, is devoted simply to unwrapping and critiquing toys. More than eight
million subscribers follow beauty vlogger Michelle Phan’s channel for makeup
tutorials. She parlayed this influence into deals with Lancôme and L’Oréal—
and eventually into her own subscription cosmetics company, recently valued
at $500 million.3
A number of singers have been discovered after posting videos of themselves
performing on YouTube. They include Cody Simpson, Austin Mahone (who
became an opening act for Taylor Swift), and 5 Seconds of Summer (who
wound up touring with One Direction). Arnel Pineda, who spent part of
his youth homeless in the Philippines, famously posted videos of his band
performing Journey covers—and went on to be hired by Journey as its lead
singer. The Weeknd, spotted on YouTube by Drake, dominated the Billboard
charts in 2015 and recently earned an Oscar nomination for best original song.
1 Susanne Ault, “Digital star popularity grows versus mainstream celebrities,” Variety, July
22, 2015.
2 David Pierson, “YouTube sensation PewDiePie launches his own network,” Los Angeles
Times, January 13, 2016.
3 Natalie Robehmed, “How Michelle Phan built a $500 million company,” Forbes, October 5,
2015; Andrea Chang, “YouTube’s biggest stars are cashing in offline,” Los Angeles Times,
August 7, 2014.
53McKinsey Global Institute Digital globalization: The new era of global flows
Platforms such as Facebook, Instagram, YouTube, and Twitter, along with countless blogs
and comment boards, are changing the nature of media content in fundamental ways. For
decades, movies, music, and news were created by large companies and consumed by
passive audiences. But now individual users anywhere in the world can comment, debate,
share, parody, co-create, and upload their own original content. The Internet has turned a
one-way monologue into a two-way digital conversation in which anyone can make their
voice heard.
•••
Globalization was once driven almost exclusively by the world’s governments, large
multinationals, and major financial institutions. Today artisans, app developers, freelancers,
and all manner of startups can participate. SMEs can scale up rapidly and even go head-
to-head with more established businesses. And individuals from Canada to Cameroon can
forge their own global connections, whether for business, personal ties, entertainment,
creativity, or simple curiosity about the world beyond their own borders. While this capability
is a tremendous social good, it does not mean that all countries share in the benefits of
globalization equally, as Chapter 3 will discuss.
© Getty Images
Now that commerce, data, and people move more fluidly across the world’s borders, most
countries have been drawn into networks of global flows to at least some degree. But their
connectedness varies enormously, and there is room for nations to play different types of
roles within those networks.
Our previous report on global flows introduced the MGI Connectedness Index to provide a
multidimensional picture of how extensively countries around the world are engaging with
the broader global economy.72 But globalization does not stand still. Our latest index offers
an updated snapshot—and by using improved data sources, it also brings our picture of the
world’s connections into sharper focus. This year’s index finds Singapore, a small country
that punches far above its weight in all types of global flows, at the top of the rankings. It is
followed by the Netherlands (one of Europe’s main digital hubs), the United States, Germany,
Ireland, and the United Kingdom.
However, nation-states are not the only unit of analysis for understanding globalization.
Within countries, individual cities and states are pursuing their own opportunities to take
part in global flows. We find that there are only eight truly “global” cities, most of which are
in advanced economies. Some states and provinces, such as California and Guangdong,
have developed such extensive international ties that they should be considered major
players in the global economy in their own right. At a broader level, larger trading blocs and
various regions of the world have different flow dynamics and levels of connectedness. This
view reveals that regions populated by countries that rank lower on our Connectedness
Index, such as those in Africa and Latin America, could start the process of deepening their
global ties by focusing on interactions with their neighbors. Measuring globalization through
each of these lenses provides additional insights into how the world’s web of connections is
constantly evolving.
SINGAPORE TOPS THE LATEST MGI CONNECTEDNESS INDEX
The MGI Connectedness Index offers a comprehensive ranking for 139 countries based
on inflows and outflows of goods, services, finance, people, and data. For each country, it
takes into account the size of each flow relative to GDP or population (flow intensity); it also
considers that country’s share of the global total within each type of flow. The combination
of these two factors results in a “connectedness score” for each type of flow. To gauge the
overall connectedness of a given country, we calculate an average of its score on each of the
five flows. (See the technical appendix for a full discussion of methodology.)
72 Global flows in a digital age: How trade, finance, people, and data connect the world economy, McKinsey
Global Institute, April 2014.
Singapore, the Netherlands, the United States,
Germany, Ireland, the United Kingdom, and China
top the latest MGI Connectedness Index.
3. HOW COUNTRIES, CITIES, AND
REGIONS ARE CONNECTING
56 McKinsey Global Institute 3. How countries, cities, and regions are connecting
Our index is not the only ranking that examines how countries participate in globalization.73
Other studies differ in the metrics included and how each is weighted. Some look at a
combination of flows and other indicators such as trade barriers. Others also include
measures of cultural globalization. The technical appendix of this report provides an
overview of several of these indexes and their key differences.
We believe that our Connectedness Index is the simplest, most transparent measure of how
countries are actually participating in globalization, or their openness to all types of flows.
In addition, our index methodology controls for the size effect of a country when measuring
openness. Some other indexes focus solely on flow intensity, examining each country’s
level of trade or capital flows relative to GDP. However, in dollar terms, this leaves the world’s
largest economies—notably the United States and China—looking comparatively closed;
their domestic markets are so large that only a relatively small share of their economic
activity is cross-border. Conversely, Luxembourg, Singapore, and other smaller countries
inevitably have larger flows of goods, services, and finance relative to the size of their
economies. We therefore combine flow intensity with each country’s share of the global total
to offer a more accurate perspective on its significance in world flows.
Exhibit 21 shows the index rankings for the top 25 countries along with a selection of other
major advanced and emerging economies. The technical appendix contains the complete
results for all 139 countries we analyzed.
Singapore claims the top spot in the rankings this year. Its globalization journey began
decades ago when it emerged as Southeast Asia’s global shipping hub, particularly for oil
and fuels. The nation also adopted four official languages (English, Mandarin, Tamil, and
Malay); this not only reflected its multicultural population but also positioned the country
to do business with the world. In recent decades, Singapore has implemented an explicit
strategy to become a regional hub for services and finance; attracting skilled international
talent and establishing tax policies and incentives to draw FDI were key priorities to make
this a reality. Singapore ranks second overall in the World Economic Forum’s Global
Competitiveness rankings, with particularly high scores for infrastructure, higher education
and training systems, and transparent and efficient institutions. It is also one of the most
digitally wired countries in the world; government statistics show that 87 percent of
households had broadband access in 2014.
Following Singapore are the Netherlands, the United States, Germany, Ireland, and the
United Kingdom. All of these countries have slightly different profiles. The Netherlands is a
major hub for Europe’s data traffic as well as a port and distribution point for traded goods.
It has created tax and regulatory regimes to become a major financial center and to attract
the holding companies and subsidiaries of many major corporations, boosting both financial
flows and services trade. Ireland has taken a similar approach, establishing corporate
tax and regulatory policies to attract high levels of FDI and significant trade in services. A
long list of major US companies, including some of the biggest tech and pharmaceutical
firms, have incorporated or established European operations in Ireland because of its tax
advantages. In fact, foreign-owned enterprises contribute 55 to 60 percent of the gross
value added of all companies located there. As a result, Ireland rose from No. 14 in MGI’s
previous index to No. 5 in this year’s ranking.
73 See, for example, Pankaj Ghemawat and Steven A. Altman, DHL Global Connectedness Index 2014:
Analyzing global flows and their power to increase prosperity, DHL, 2014; and The ICT globalisation index, the
Economist Intelligence Unit, 2014.
57McKinsey Global Institute Digital globalization: The new era of global flows
Exhibit 21
MGI Connectedness Index
Country connectedness index and overall flows data, 2014
Rank of participation by flow as measured by flow intensity and share of world total
1–10 11–25 26–50 >50Connectedness index rank 100+ <7070–99Flow intensity Rank Country Score Connectedness Index rank Flow value1 $ billion Flow intensity2 % of GDPGoods Services Finance People Data 1 Singapore 64.2 1 2 2 12 6 1,392 452 2 Netherlands 54.3 3 3 6 21 1 1,834 211 3 United States 52.7 7 7 3 1 7 6,832 39 4 Germany 51.9 2 4 8 3 2 3,798 99 5 Ireland 45.9 32 1 1 28 9 559 227 6 United Kingdom 40.8 13 5 5 6 3 2,336 79 7 China 34.2 4 16 4 82 38 6,480 63 8 France 30.1 11 8 9 7 4 2,262 80 9 Belgium 28.0 5 6 33 33 8 1,313 246 10 Saudi Arabia 22.6 20 28 27 2 53 790 106 11 United Arab Emirates 22.2 6 23 17 4 46 789 196 12 Switzerland 18.0 12 11 10 17 13 848 115 13 Canada 17.3 16 22 11 11 18 1,403 79 14 Russia 16.1 21 25 18 5 25 1,059 57 15 Spain 14.4 25 13 19 14 16 1,105 79 16 Korea 14.0 8 12 28 50 44 1,510 107 17 Italy 13.4 17 18 24 16 19 1,587 74 18 Sweden 13.0 29 14 22 31 5 572 100 19 Austria 11.7 26 17 31 20 12 470 108 20 Malaysia 11.6 9 19 25 26 43 610 187 21 Mexico 10.7 14 63 34 18 41 1,022 80 22 Thailand 10.7 10 15 36 44 64 605 162 23 Kuwait 10.6 37 46 13 13 75 306 153 24 Japan 10.5 15 20 12 81 20 2,498 54 25 Kazakhstan 10.0 48 73 41 8 57 176 83 26 Ukraine 9.8 38 39 87 10 34 133 101 27 Australia 9.7 30 34 21 15 33 825 57 28 Denmark 8.9 35 9 32 41 11 369 108 29 Jordan 8.8 73 50 75 9 83 50 138 30 India 8.5 24 10 35 58 70 1,316 64 32 Czech Republic 7.5 18 33 57 59 15 397 193 34 Poland 7.0 23 31 47 34 22 585 107 35 Hungary 6.8 22 30 26 62 17 287 209 36 Norway 6.0 36 24 20 46 24 458 92 37 Vietnam 5.7 19 54 45 103 61 350 188 39 Finland 5.5 46 27 23 70 10 390 144 40 Portugal 5.5 47 36 30 23 31 255 111 41 Turkey 5.1 28 40 53 38 29 521 65 43 Israel 4.9 51 32 49 24 56 248 82 44 Brazil 4.5 41 38 14 125 30 869 37 45 Chile 4.1 45 58 16 102 27 239 92 47 Greece 4.1 60 29 54 35 42 160 67 48 New Zealand 3.9 67 48 61 25 51 130 63 51 Indonesia 3.4 31 49 38 106 76 504 57 53 South Africa 3.3 34 57 52 64 80 277 79 54 Philippines 3.2 54 41 44 52 67 230 81 64 Morocco 2.6 58 43 74 56 65 104 97 73 Egypt 2.2 68 42 69 73 71 158 55 83 Nigeria 1.9 55 76 48 128 98 268 47 86 Peru 1.8 62 88 51 104 49 122 60 118 Kenya 1.3 100 84 127 119 91 35 58 SOURCE: McKinsey Global Institute analysis 1 Flows value represents total goods, services, and financial inflows and outflows. 2 Flow intensity represents the total value of goods, services, and financial flows as a share of the country’s GDP. 58 McKinsey Global Institute 3. How countries, cities, and regions are connecting In contrast, the United States and Germany both follow a generalist model with strength across all five flows. The United Kingdom also has broad participation across flows, with a spike in cross-border service and financial flows, a reflection of London’s role as a global financial hub. China’s surge in our rankings is particularly noteworthy. It has climbed from 25th in our previous index to the No. 7 spot. Now the second-largest economy in the world, it accounts for a larger share of all the total world’s flows. Although its participation has risen across all types of flows, it is strongest in trade and FDI and notably lower in data flows and people flows. China ranks only 38th for data flows, reflecting government restrictions on access to foreign Internet sites. China’s short-term people flows are rising: it welcomed 56 million tourists in 2014, and 109 million Chinese traveled abroad, a tenfold increase since 2000.74 Despite those trends, it ranks only 82nd in people flows, since this metric is based heavily on long-term migrants. Japan and South Korea have similar patterns of strong participation in traditional flows of goods, services, and finance but much weaker participation in data and people flows. Japan ranks only 24th in overall connectedness, below Malaysia, Mexico, and Thailand. Like China, it posts a particularly low ranking on people flows, at only 81st in the world. Japan also ranks only 20th for both data flows and service flows. Language and cultural barriers may explain part of this result, as may the residual effects of the country’s history as a closed society. Japan also attracts relatively little foreign investment, and despite its manufacturing prowess, its goods trade has been hindered over the years by an adherence to its own proprietary technologies and standards rather than an embrace of open platforms.75 South Korea ranks higher in our index, although still only No. 16 in the world, due to lower rankings on finance, data, and people flows. Saudi Arabia places 10th in the rankings, followed by the United Arab Emirates at No. 11. Both are major oil exporters that meet most of their demand for manufactured consumer goods through imports. In addition, both have very high rankings in people flows—second and fourth in the world, respectively—reflecting large numbers of both low-skill guest workers and highly skilled professionals. In fact, more than half of Saudi Arabia’s labor force is made up of foreign workers on temporary contracts.76 The UAE has significant service and financial flows in addition to oil exports. Dubai has actively built a role as one of the world’s major ports, transit hubs, and financial centers. 74 United Nations World Tourism Organization statistics. 75 The future of Japan: Reigniting productivity and growth, McKinsey Global Institute, March 2015. 76 Saudi Arabia beyond oil: The investment and productivity transformation, McKinsey Global Institute, December 2015. Singapore’s top ranking reflects its successful strategy to become a regional hub for services and finance. Attracting skilled talent and FDI were among its key priorities. 59McKinsey Global Institute Digital globalization: The new era of global flows ALTHOUGH MORE COUNTRIES ARE BUILDING GLOBAL CONNECTIONS, THERE ARE LARGE GAPS IN PARTICIPATION Today global connections link a larger and more diverse range of countries. For the first time in history, emerging economies are counterparts on more than half of global trade flows, and South-South trade among these countries is the fastest-growing type of connection (Exhibit 22). The value of traded goods and services plus financial flows exceeded 80 percent of GDP for 72 countries in 1990. By 2014, 121 nations were above this threshold. Today even the smallest countries are participating to some degree in globalization. Flows are concentrated in a few leading countries Despite these trends, the world is still far from fully globalized. The MGI Connectedness Index shows that advanced economies in general remain more connected than developing countries—and the leading countries are far ahead of everyone else. The aggregate connectedness score (shown in the third column of Exhibit 21) reveals a very large distance between Singapore, the overall leader with a score of 64, and every other country. Ireland is No. 5 in the ranking, but its overall score is only 46, indicating that it is roughly two-thirds as connected as Singapore.77 Saudi Arabia ranks No. 10 overall, but its score is only 23, meaning it is only one-third as connected as Singapore. Beyond the top 25, all 114 77 See the technical appendix for details on how the scores were calculated and how the index was constructed. Exhibit 22 60 20 80 100 40 0 Other South-North 2014 North-North China-South China-North Other South-South 2000 111002 06 12090403 0801 07 1305 7x 20x 8x 3x 2x Change in value of trade, 2000–14 SOURCE: UNCTAD; McKinsey Global Institute analysis Emerging economies are now involved in more than half of the world’s goods trade Bilateral goods trade by development status, 2000–14 % of total goods trade 60 McKinsey Global Institute 3. How countries, cities, and regions are connecting remaining countries have overall connectedness scores in the single digits. This indicates that all countries have significant room to expand their participation in global flows. This pattern is apparent not only in the aggregate connectedness score but also within each type of flow. All flows are disproportionately concentrated among a small set of countries, with huge gaps between leaders and laggards (Exhibit 23). The top 15 countries in traded goods, for instance, account for 63 percent of the global total, down only slightly since 2005. That share is 62 percent in services, and 79 percent in FDI. At this simple level, only cross- border data flows have a clear reduction in the share of the top 15 countries, falling from 86 percent in 2005 to 77 percent in 2014. Exhibit 23 13 21 66 63 23 14 22 65 13 14 23 62 4 17 79 79 5 16 11 86 2 77 7 16 SOURCE: UNCTAD; IMF; TeleGeography, Global Internet Geography; McKinsey Global Institute analysis Flows remain concentrated among a few leading countries, with little change since 2005 % of world total Global goods flow distribution 2005 2014 Global services flow distribution 2005 2014 Global FDI flow distribution 2005 2014 Global data flow distribution 2005 2014 1 Tbps = terabits per second. NOTE: Numbers may not sum due to rounding. Next 20 countriesTop 15 countries All others $10.6 trillion $19.0 trillion $2.5 trillion $4.9 trillion $1.39 trillion $1.63 trillion 4.8 Tbps1 211 Tbps1 61McKinsey Global Institute Digital globalization: The new era of global flows The individual country scores for each type of flow present an even starker picture, with large drop-offs between the most connected country and the second, third, and following countries. For goods flows, the fifth-ranked country, Belgium, has a score that is only 78 percent of Singapore’s leading score. Global service flows are even more skewed, with the fifth-ranked country (the United Kingdom) at just 35 percent of the leader (Ireland). In data flows, the fifth-ranked country (Sweden) is only 25 percent as connected as the leader (the Netherlands).78 Lagging countries are catching up to leaders only very slowly We use statistical tests of convergence to see if the gaps between country participation in global flows are closing over time. Beta convergence, for instance, measures the extent to which lagging countries are catching up to leaders, while sigma convergence measures whether the variation between countries overall is narrowing. The results show that goods and migration flows have positive beta convergence, indicating that lagging countries are catching up—but only very slowly. If current trends hold, the convergence tests indicate that it will take at least eight years for the gap in goods flows between leaders and laggards to be reduced by half, and 11 years for the same to happen in migration flows. The gap between the leading countries and the rest of the pack is particularly stubborn in foreign investment. The top 15 countries in FDI flows have consistently accounted for approximately 80 percent of the world total over the past 20 years—and we estimate it would take 13 years for the current trajectory to cut the gap between laggards and leaders by half. Data flows do not show beta convergence, indicating that lagging countries are not catching up to the leaders. This may reflect the fact that data flows are still a relatively new phenomenon and continuing to grow quickly for even leading countries. These gaps in participation matter given the economic value of global flows—a topic that will be explored more fully in Chapter 4. Exhibit 24 graphically illustrates the large gaps between leading countries and the rest of the world in the context of their relative prosperity. It shows each country’s overall connectedness score on the vertical axis and its per capita income on the horizontal axis. In general, country connectedness rises with income, but there are interesting exceptions. China stands out for its high level of connectedness relative to its income level. This reflects not only its outsized role in global manufacturing value chains, but also its significant financial flows and its growing flows of services. At the other extreme are several Middle Eastern economies that have high per capita income but much lower global connectedness. 78 See the technical appendix for rankings by type of flow. The MGI Connectedness Index shows that advanced economies are generally more connected than developing countries—and the leaders are far ahead of everyone else. 62 McKinsey Global Institute 3. How countries, cities, and regions are connecting Exhibit 24 A small group of leading countries are much more connected than the rest of the world 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 10 7040 600 20 8050 14030 United States Belgium Australia India Italy China South Korea Russia Canada Japan Germany Mexico Spain United Kingdom Switzerland Connectedness score, 2014 Per capita GDP, 2014 $ thousand, purchasing power parity, current international dollar Ukraine Greece Czech Republic Norway Austria Turkey QatarFinland Denmark Ireland Singapore Indonesia Sweden United Arab EmiratesSaudi Arabia Portugal Kuwait France Netherlands Emerging Developed Correlation coefficient (r) = 0.54 Size of circle represents $ value of flows in 2014 SOURCE: IMF; McKinsey Global Institute analysis Vietnam Brazil Thailand Malaysia 63McKinsey Global Institute Digital globalization: The new era of global flows GLOBAL CONNECTEDNESS CAN ALSO BE VIEWED THROUGH THE LENS OF CITIES, PROVINCES, AND REGIONAL BLOCS The MGI Connectedness Index assesses how countries are participating in global flows— but they are not the only actors in the global economy. Cities, regions within countries, and broader blocs of countries are connecting with the global economy in myriad ways and to varying degrees. Looking through these various lenses provides unique insights into the new patterns of globalization. There are eight truly global cities, but only two are in emerging countries Cities are the real engines of the world economy and serve as major waypoints for global flows. Previous MGI research suggested that just 600 cities will generate some two-thirds of world GDP by 2025.79 Cities with major ports can become important hubs in the global flow of goods. Urban centers with significant Internet infrastructure can play the same kind of role for data and communication flows. Large cities in countries with low barriers to immigration can attract people from around the world. Acting as a waypoint generates significant economic output and high-quality jobs, and it helps a city accumulate knowledge, skills, and talent, with positive spillover effects on its broader economy. Once a city has established itself as a waypoint for a particular flow, other economic activity tends to coalesce or co-locate along with it. A city that establishes itself as a financial hub, for instance, is likely to attract insurance companies and other professional service firms. One that acts as a major waypoint for people and goods traffic through a major airport will become a magnet for logistics and distribution companies. Amsterdam invested in the Amsterdam Internet Exchange (AMS-IX) in the early days of the Internet and continues to be one of the largest data hubs in the world. Since it offers faster connection speeds than any other European city, it has attracted other types of high-tech businesses. Unfortunately, data on global flows are not available at the city level. However, we have obtained data that serve as proxies for each of our five global flows. Container port volumes approximate goods flows; airport passenger volumes serve as a proxy for goods, service, and people flows; the ranking of cities in the Global Financial Centers Index by the Z/Yen Group provides an indication of financial flows; the number of foreign-born residents in a city measures people flows; and Internet bandwidth approximates data flows. Our last report found that the world had only eight truly “global cities,” defined as ranking in the top 25 in at least four of the five major flows: New York, London, Tokyo, Los Angeles, San Francisco, Singapore, Hong Kong, and Dubai. This year, using updated data, we find that Tokyo drops off the list due to a decline in goods trade, while Shanghai takes its place (Exhibit 25). In fact, Chinese cities generally rose in the rankings, while a number of European cities (such as Bremen, Hamburg, Frankfurt, Brussels, Geneva, and Madrid) fell. 79 Urban world: Cities and the rise of the consuming class, McKinsey Global Institute, June 2012. New York, Los Angeles, San Francisco, London, Singapore, Shanghai, Hong Kong, and Dubai are the world’s truly global cities. 64 McKinsey Global Institute 3. How countries, cities, and regions are connecting Exhibit 25 Only eight of the world’s major cities are hubs for at least four of the five major flows SOURCE: Lloyd’s List; Containerisation International; Airports Council International; Global Financial Centers Index; Migration Policy Institute; TeleGeography, Global Internet Geography; McKinsey Global Institute analysis City participation in major flows by rank and change over previous year in each flow1 1 Metropolitan areas with at least 1 million foreign-born residents. Exact foreign-born population of Jeddah not known, so it is included at the bottom of the list. 2 Rankings come from different years: ports (2014), airports (2014), financial centers (2014), migration (2011), and online traffic (2015). Rank2 Goods Goods, services, people Financial People Data and communication 1 Shanghai Atlanta London New York Frankfurt 2 Singapore Beijing New York Los Angeles London 3 Shenzhen London Hong Kong London Amsterdam 4 Hong Kong Tokyo Singapore Hong Kong Paris 5 Ningbo Los Angeles Tokyo Toronto New York 6 Busan Dubai Seoul Paris Los Angeles 7 Guangzhou Chicago Zurich Miami Miami 8 Qingdao Paris Toronto Sydney Stockholm 9 Dubai Dallas/Fort Worth San Francisco Chicago San Francisco 10 Tianjin Hong Kong Washington, DC Singapore Singapore 11 Rotterdam Frankfurt Chicago San Francisco Hong Kong 12 Port Klang Jakarta Boston Melbourne Tokyo 13 Kaohsiung Istanbul Geneva Moscow Moscow 14 Dalian Amsterdam Frankfurt Houston Milan 15 Hamburg Guangzhou Sydney Dubai Vienna 16 Antwerp Singapore Dubai Riyadh Washington, DC 17 Xiamen Denver Montreal Washington, DC Hamburg 18 Tanjung Pelepas New York Vancouver Dallas Beijing 19 Los Angeles Shanghai Luxembourg Jeddah Marseille 20 Long Beach Kuala Lumpur Osaka Copenhagen 21 Laem Chabang San Francisco Shanghai Brussels 22 Tanjung Priok Bangkok Qatar Warsaw 23 Ho Chi Minh City Incheon Shenzhen Shanghai 24 Bremen Charlotte Busan São Paulo 25 New York Las Vegas Tel Aviv Madrid 65McKinsey Global Institute Digital globalization: The new era of global flows It is notable how few cities in the developing world appear in the top 25 across multiple flows, although some do play major roles in one individual flow. Only two of the eight global cities are in emerging economies (Dubai and Shanghai). Five cities in advanced economies (Tokyo, Paris, Chicago, Frankfurt, and Washington, DC), but none in emerging economies, appear in the top 25 for three of the five dimensions. Four cities in emerging countries (Beijing, Guangzhou, Shenzhen, and Moscow) appear in the top 25 for two types of flows. The urban giants of the developing world have room to expand the ways in which they participate in global glows, especially as the world’s economic center of gravity shifts in their direction. Examining the global connectedness of states and provinces reveals large variations within countries Within countries, the global connectedness of different states and provinces may vary dramatically. For several large countries (China, Germany, the United Kingdom, and the United States), we have obtained data on state-level participation in global flows of goods and migration. China has the most striking variation in global goods flows across its provinces (Exhibit 26). Only six of its 34 administrative divisions account for some three-quarters of its foreign trade in goods. 80 These coastal provinces, plus the areas around Beijing and Shanghai, are hubs of cross-border activity. But the interior of China remains largely disconnected from the world. The United States also has significant variation in how states participate in the global goods trade: California and Texas rank highest, reflecting ports in California and the oil sector in Texas. By contrast, industry is spread widely across multiple provinces in Germany and the United Kingdom, and so are their global flows. Goods flows relative to GDP for Germany’s most connected state (North Rhine–Westphalia) were 8.7 times the median, while China’s most connected state (Guangdong) stood at 26 times the median. Evaluating how states and provinces would stack up in the global rankings against nation- states underscores their range of performance and the global role that some play. China’s booming coastal province of Guangdong, for instance, is a major manufacturing and trade hub that would rank sixth globally in terms of goods flows—above the United States. Shanghai and Beijing would top Japan and Italy in goods flows (Exhibit 27). California would rank fourth in the world for people flows, above the United Kingdom, France, and the UAE. At the other end of the spectrum, South Dakota would rank 261st globally in goods flows, below Grenada, Samoa, and Cape Verde. These types of extremes are not apparent in Germany, however. Germany’s most connected states for flows of goods and people rank only 44th and 33rd, respectively, in global terms—but Germany as a whole ranks second in goods flows and third in people flows, reflecting the fact that all of its regions actively participate in global flows. Building global connections does not have to be left to national policy makers. Local business and government leaders everywhere have the ability to forge their own global ties directly—and digital globalization gives them a multitude of new ways to pursue these opportunities. 80 China’s administrative divisions include provinces, municipalities, autonomous regions, and special administrative regions. 66 McKinsey Global Institute 3. How countries, cities, and regions are connecting Exhibit 26 Within countries, regional connectedness varies greatly SOURCE: McKinsey Global Institute analysis Regional goods trade connectivity score, 2014 Higher number indicates greater connectedness NOTE: Countries not to scale. 0–10.0 10.1–20.0 20.1–40.0 40.1–100.0 United States China California Texas United Kingdom West Midlands East of England South East Germany Hamburg Niedersachsen North Rhine- Westphalia Baden- Württemberg Bavaria Guangdong Shanghai Beijing Jiangsu Zhejiang 67McKinsey Global Institute Digital globalization: The new era of global flows Exhibit 27 South Korea United Kingdom Switzerland 44.1 Guangdong (China) Shanghai (China)1 Mexico 25.0 France 28.6 23.7 Beijing (China)1 Canada 24.5 25.8 Thailand 24.4 Japan 24.5 29.6 28.4 Malaysia Italy 30.0 United States 45.9 United Arab Emirates 46.9 90.9 Belgium 100.0Singapore China Netherlands 87.8 51.3 78.4 Germany 92.6 23.6 16.7 Austria Baden–Württemberg (Germany) 16.5 16.3 15.4 Poland 16.7 16.7 Hungary Russia Jiangsu (China) Spain 28.4 14.7 Saudi Arabia 16.2 India 19.7 Vietnam 20.8 18.4 North Rhine–Westphalia (Germany) Texas (United States) 11.2 Bavaria (Germany) 13.8 14.6Slovak Republic Czech Republic If states and provinces are ranked among countries, Guangdong would be sixth globally in goods flows and California would rank fourth in people flows Connectedness index score, goods and people flows, 2014 Goods People SOURCE: McKinsey Global Institute analysis Qatar 21.5 Mexico 22.3 17.0Netherlands 14.8 New York State (United States) 18.3 Austria 21.4 30.6 Italy Australia 24.4 Spain Kuwait Switzerland 31.4 25.6 26.6 Ukraine 36.7 Jordan 33.7 Canada 35.1 Singapore 51.2 100.0 49.9 45.7 France United Kingdom 43.3 Russia Kazakhstan 44.1 Saudi Arabia 81.7 39.9 United Arab Emirates Germany United States California (United States) 40.8 37.0 Malaysia Lebanon 11.5 11.5Croatia 12.0 New Zealand London (United Kingdom)1 Israel Portugal Texas (United States) 10.9 Florida (United States) 12.9 13.6 11.4 12.3 12.6 12.3 12.7 Belarus Ireland North Rhine–Westphalia (Germany) State/region Country 1 Metropolitan areas considered as states/provinces by national authorities. 68 McKinsey Global Institute 3. How countries, cities, and regions are connecting A great deal of global trade circulates among neighbors Connecting with the world starts close to home. As described in Chapter 1, we find a significant share of each flow circulates within geographic regions rather than across long distances. However, this pattern varies around the world. In goods trade, the highest share of intraregional trade is in Europe (61 percent). Countries including Croatia, the Czech Republic, Hungary, Latvia, Norway, Portugal, and Romania conduct most of their trade with their European neighbors (Exhibit 28). Some 45 percent of East Asia’s trade is intraregional, followed by North America (which includes Mexico) at 42 percent. The corresponding share is far lower in Africa, Latin America, and South Asia, where most countries score far lower on the MGI Connectedness Index. This indicates a significant opportunity for developing countries to increase their participation in flows by trading with their neighbors. One way to facilitate greater intraregional trade is to form an economic community or trade bloc. Academic research shows that being part of a formal trade bloc—that is, a region defined by an intergovernmental agreement to reduce or eliminate regional barriers to trade—increases both the level and the impact of global goods trade.81 However, our data show that the intensity of intraregional trade within the world’s biggest trading blocs varies widely. Exhibit 29 shows the share of goods flows that occurs within these established regions. After the North American Free Trade Agreement (NAFTA) was established, the share of goods trade between the United States, Canada, and Mexico peaked at 45 percent of their total in 2000. However, that share had fallen back to 42 percent in 2014 as these countries all began to trade more actively with China. China now accounts for 13 percent of trade with NAFTA countries. Trade between the Association of Southeast Asian Nations group of countries remains far lower than trade within the European Union and NAFTA, at just 24 percent of their value of total trade in 2014, somewhat higher than the 16 percent share in 1990. Recognizing the opportunities at stake, ASEAN’s leaders are working to implement an ambitious ASEAN Economic Community (AEC) integration plan to harmonize many administrative procedures and regulatory standards among member states, creating an open market of 600 million consumers and a more seamless production base.82 The East African Community (EAC) enacted a free trade zone in recent years to eliminate all tariffs among its five member countries (Burundi, Kenya, Rwanda, Tanzania, and Uganda). However, its intraregional trade stands at only 10 percent. Interestingly, EAC trade with China is higher than even ASEAN’s trade with China, at 17 percent compared with 15 percent, despite ASEAN’s proximity. 81 The World Trade Organization has investigated preferential trade agreements and their effects in detail. These bilateral agreements have grown in importance as multilateral agreements have become more difficult to establish and negotiate. See World trade report 2011: The WTO and preferential trade agreements: From co- existence to coherence, World Trade Organization, 2011. 82 Southeast Asia at the crossroads: Three paths to prosperity, McKinsey Global Institute, November 2014. Developing countries have a significant opportunity to increase their participation in global flows by trading with their neighbors. 69McKinsey Global Institute Digital globalization: The new era of global flows Increasing intraregional trade among developing country trade blocs is a clear priority—one that could be a stepping-stone to increasing their overall connectedness and capturing the benefits of scale, specialization, and competition. Although the impact of distance on all forms of global flows is falling as transportation and communication costs decline, economic ties among countries in the same region are still the norm. Bolstering regional ties, particularly in Africa and Latin America, is an opportunity to boost flows—and to raise GDP, as we will discuss in the next chapter. Exhibit 28 Some European countries trade predominantly with their neighbors Intraregional goods trade in Europe (includes imports and exports), 20141 % share of total trade 1 Includes EU-28 and select W estern European countries including Andorra, Iceland, Norway, and Switzerland. SOURCE: UNCTAD; McKinsey Global Institute analysis 40–50 50–60 60–70 70+ Portugal Iceland Lithuania Latvia Estonia Finland Sweden Norway Denmark United Kingdom NetherlandsIreland Germany Poland Belgium Luxembourg Italy France SwitzerlandSpain Greece Malta Cyprus Czech Republic Austria Hungary Bulgaria Romania Croatia Slovak Republic 70 McKinsey Global Institute 3. How countries, cities, and regions are connecting ••• Measuring the connectedness of different countries, cities, and regions around the world illustrates how far globalization has advanced—and how much further it could still progress. Much of the momentum behind globalization is still being driven by a handful of highly connected countries, and much of the world has only just begun to tap into global flows. The next chapter will provide detailed econometric analysis of the value of connectedness, underscoring the importance for emerging economies to pursue catch-up growth. Exhibit 29 24 61 15 42 45 13 SOURCE: UNCTAD; IMF; McKinsey Global Institute analysis The EU is the most integrated of the world’s major trading blocs Trade within and outside of trading blocs, 2014 % share of total goods trade Extra-regional (excluding China) China Intraregional 34 61 6 NAFTAEU-28 73 17 10 ASEAN EAC (East African Community)1 100% = $5.6 trillion 100% = $11.8 trillion 100% = $2.5 trillion 100% = $0.05 trillion 1 Comprises Burundi, Kenya, Rwanda, Tanzania, and Uganda. NOTE: Numbers may not sum due to rounding. 71McKinsey Global Institute Digital globalization: The new era of global flows © Getty Images © Getty Images The economic impact of globalization is growing as the web of connections between countries and companies grows broader and deeper. As vast populations in emerging economies gain a new foothold in the middle class, more of the world’s business involves cross-border flows. Digital flows of data, information, and communication allow ideas and innovation to ricochet around the world; they enable companies to bring together the best inputs and talents to create higher-quality goods and services. Perhaps most important, global flows expose companies to more competition and best practices, spurring them to improve performance and productivity. To measure the economic impact of global flows, we have undertaken extensive econometric analysis, building on our previous report and applying improved data and more sophisticated methodology. We use data on the cross-border inflows and outflows of goods, services, finance, people, and data for 97 countries from 1995 to 2013 to assess the impact of flows on GDP and productivity.83 Our analysis breaks new ground by testing the impact of all types of flows together and by analyzing the impact of gross inflows and outflows, rather than looking more narrowly at trade balances. We find strong evidence that country participation in global flows has both short-term (cyclical) and long-term (structural) impact on GDP. Our results indicate that over the past decade, global flows have raised world GDP by roughly 10 percent over what would have resulted in a world in without any flows. In 2014 alone, this impact was equal to $7.8 trillion in global GDP. Flows of goods and FDI, the two categories most heavily associated with the last era of globalization, account for about half of this. But data flows, worth some $2.8 trillion in 2014, now exert a larger impact on global GDP than the flow of goods. Their role in the global economy has expanded with astonishing speed. After all, the world’s trade networks have developed over the course of centuries while global data flows were negligible as recently as 15 years ago. Chapter 3 showed that each type of flow remains concentrated among a short list of leading countries, and lagging countries are catching up only very slowly. Here, we find that some lagging countries could have boosted their current GDP by 50 percent or even more by accelerating their participation in global flows over the past ten years. They can still realize tremendous opportunities by targeting ways to build deeper ties to the global economy. 83 Our data end in 2013 because that is the most recent year for which a large set of countries report inflows and outflows of migrants. Both inflows and outflows support growth by circulating ideas, research, technologies, talent, and best practices around the world. $7.8T contribution to world GDP in 2014 by all types of global flows combined 4. GLOBAL FLOWS BOOST ECONOMIC GROWTH 74 McKinsey Global Institute 4. Global flows boost economic growth Overall, our analysis provides strong evidence of the economic value of openness—and the benefits are much broader and more nuanced than a simple accounting of net exports. Consumers, for instance, gain purchasing power from trade in goods, which delivers a wider variety of products at lower prices. One study found that middle-class Americans gain more than a quarter of their purchasing power from trade.84 Both inflows and outflows of goods, services, finance, people, and data are important as they expose economies to ideas, research, technologies, talent, and best practices from around the world. Countries that participate in global flows can specialize in what they produce, realize economies of scale, and invite competition, which spurs both efficiency and innovation in local industries. These benefits have been recognized since they were outlined by economist David Ricardo in the early 19th century—but today there is a difference. As they transform traditional flows and enable broader participation, digital technologies are amplifying these effects (Exhibit 30). 84 The economic benefits of US trade, Executive Office of the President, May 2015. Comparative advantage and specialization Broader participation Accelerated information flows Truly global market scale Economies of scale Increased competition boosting efficiency Knowledge diffusion Capital deepening and widening Human capital development HIGHER PRODUCTIVITY MORE INNOVATION INCREASED GDP D IG IT IZ A TI O N How globalization increases GDP Exhibit 30 75McKinsey Global Institute Digital globalization: The new era of global flows GLOBAL FLOWS HAVE INCREASED CURRENT WORLD GDP BY AT LEAST 10 PERCENT, ADDING $7.8 TRILLION IN 2014 Our ongoing research into global flows aims to create a clearer view of a diffuse and evolving phenomenon. A key part of that is providing a better understanding of how flows affect economic growth. Our last report established that global flows contribute to GDP growth and that countries that are more central in the networks of global trade in goods benefit disproportionately, irrespective of the direction of the flows.85 It also found that data flows are an increasingly important driver of economic growth. For this report, we set out to sharpen our picture of global flows by refining our econometric methodology and using updated and improved data. For example, we previously used international telephone calls as a proxy for data flows. But this arguably measured only the communication aspect of data flows rather than the way they transmit information and enable transactions. Now we are able to track used cross-border bandwidth, which is a much closer measure of cross-border data flows. To measure people flows, we use the stock of migrants, both inbound and outbound. Our analysis also separates the impact of foreign direct investment and other types of financial flows. We make this distinction since a large body of academic literature finds that while FDI has a clear positive impact on growth, the evidence for the impact of other types of financial flows (cross-border lending and portfolio investments of equity and debt) is mixed.86 The technical appendix of this report contains a detailed description of our methodology. In short, we use a co-integrated, two-step error-correction model that enables us to examine without bias both the short- and long-term impacts of global flows on growth. To measure the one-way effect from flows on GDP growth, we use lagged covariates as instrumental variables to correct for possibly endogeneity. Our methodology allows us to control for unobserved country-specific effect and for correlation among the variables. We also test the impact of flows on productivity compared with the utilization of capital and labor. Finally, using two different measures of network centrality, we test how a country’s position in the network of global flows contributes to its growth. Our model considers how all types of global flows act in concert to generate growth Our results provide robust evidence that goods, data, and FDI flows have both a positive short-term impact on GDP and a positive long-term structural impact (Exhibit 31). Our results for migration flows are surprising: we find a negligible or even negative impact on GDP growth for developing economies, possibly reflecting the loss of skilled labor (commonly referred to as “brain drain”) or the difficulty that some have in absorbing a large influx of refugees or migrants. We find, however, that migration flows have a positive impact on productivity for advanced economies, which is consistent with other literature.87 85 Global flows in a digital age: How trade, finance, people, and data connect the world economy, McKinsey Global Institute, April 2014. 86 See, for example, Joshua Aizenman, Yothin Jinjarak, and Donghyun Park, “Capital flows and economic growth in the era of financial integration and crisis, 1990–2010,” Open Economies Review, volume 24, issue 3, July 2013. 87 For instance, see Ekrame Boubtane, Jean-Christophe Dumont, and Christophe Rault, Immigration and economic growth in the OECD countries, 1986–2006, IZA discussion paper number 8681, November 2014. Cross-border data flows added some $2.8 trillion to world GDP in 2014, surpassing the impact of the global goods trade. 76 McKinsey Global Institute 4. Global flows boost economic growth The magnitude of the increases in both GDP level and growth generated by global flows is quite striking. In one specification of the model, using the value of flows for each country normalized by their size, we find that over a decade, all types of flows acting together have raised world GDP by 10.1 percent over what would have resulted in a world without any cross-border flows. This amounted to some $7.8 trillion in 2014 alone. Flows of goods trade and FDI account for roughly half of this growth effect (raising world GDP by 3.5 percent and 1.6 percent, respectively; Exhibit 32). Data flows alone directly raised world GDP by 3.0 percent. In addition, however, they also enable trade flows, FDI, and even people flows. If we account for both the direct and indirect impact of data flows, we find they exert an even larger impact on world GDP than physical trade flows (see Box 3, “Valuing cross-border data flows”). People flows have raised the level of world GDP by 2.0 percent.88 Viewed another way, the impact of flows on global GDP may even be larger than 10.1 percent. In a second econometric model specification, we used data on each country’s score in the MGI Connectedness Index for each flow, rather than the value of inflows and outflows. This corrects for the fact that cross-border flows may be lower relative to GDP for a very large economy (such as the United States) simply due to the huge size of its domestic markets. Using connectedness scores instead of actual flows, we find that the contribution 88 As noted above, our econometric results show that migration flows have a negative elasticity with respect to GDP at a global level (despite the positive impact of immigration on productivity in advanced economies). We find an overall positive effect on lifting global GDP in the period we analyzed (2003–13) because global migration flows actually declined slightly over the period relative to world population, thus producing a net positive effect on global growth. Although people flows include growing business travel, tourism, and study abroad in reality, our model considers only long-term migration due to a lack of historical data on other types of people flows for many countries around the world. Exhibit 31 Short-/long-term impact Name of variable Granger causality with real GDP Expected sign of coefficient Estimated sign of coefficient FDI Two-way Positive/positive Positive/positive Goods trade flow Two-way Positive/positive Positive/positive Immigration Two-way Positive/positive Insignificant/negative1 Data flows Two-way Positive/positive Positive/positive Services trade flow Two-way Positive/positive Extended due to correlation with FDI Fixed capital stock n/a Positive/positive Positive/positive Employment n/a Positive/positive Positive/positive Average years of education n/a Positive/positive Insignificant/negative SOURCE: McKinsey Global Institute analysis The coefficients from our econometric model have the expected sign Flow variables Dependent variable (Log) Real GDP Independent variables (Log) 1 Migration flows are negligible or slightly negative at the global level, possibly due to the loss of skilled labor in developing countries or the difficulties of absorbing a large influx of refugees or migrants. However, migration flows have a positive impact on productivity in advanced economies. 77McKinsey Global Institute Digital globalization: The new era of global flows of flows to world GDP in 2014 could be as high as 18.7 percent, or $14.4 trillion—meaning that one in every six dollars of value added in the world comes from cross-border connections.89 89 See technical appendix for the model results. Exhibit 32 Our econometric model shows that global flows account for approximately 10 percent of global GDP output 3.5 3.0 2.0 1.6 10.1All flows FDI Migration2 Data flows Goods trade SOURCE: McKinsey Global Institute analysis 1 Includes inflows and outflows data for 139 countries in MGI Global Flows model; see technical appendix for more details. 2 Global migration flows declined slightly from 2003 to 2013, resulting in a positive impact despite a negative coefficient. Migration flows are negligible or slightly negative at the global level, possibly due to the loss of skilled labor in developing countries or the difficulties of absorbing a large influx of refugees or migrants. However, migration flows have a positive impact on productivity in advanced economies. NOTE: Numbers may not sum due to rounding. Impact on GDP, 20141 $ trillion 2.7 2.3 1.5 1.3 7.8 Long-term impact on level of GDP1 % Box 3. Valuing cross-border data flows Flows of goods, services, and finance are measured in dollar terms that reflects the monetary value of the transactions. Our data on cross-border data flows, in contrast, are measured in volume, or terabits per second. To accurately compare the scale of these flows, we need a common unit of measurement. We have therefore attempted to calculate cross-border data flows in value-added terms. Our methodology is admittedly imprecise by its nature and requires some assumptions. Still, the results are striking even if only directionally correct. We begin with our econometric analysis, which finds that the direct impact of data flows raises world GDP by 3.0 percent annually. This equates to $2.2 trillion in 2014. However, we also know that cross-border data flows enable other types of flows. Consider that cross-border e-commerce now accounts for 12 percent of global trade. In addition, digital communication and platforms likely enable nearly every trade transaction on the planet today. In addition, data flows allow service exports to be delivered digitally, and digital transactions and communication enable foreign direct investment. Even people flows may benefit from digital platforms and the ability to book travel online and research foreign destinations. We make the conservative assumption that data flows account for 12 percent of the value created by other types of flows. This contributes an additional $0.6 trillion to their 2014 impact. Adding these indirect effects to the direct effects found in our model, we find that cross-border data flows may have raised world GDP by roughly $2.8 trillion in 2014. This surpasses the $2.7 trillion impact of the global goods trade. In just a decade, global data flows have generated as much economic value as trade networks formed over the course of centuries. 78 McKinsey Global Institute 4. Global flows boost economic growth Our research is unique in that it considers the impact of all types of flows acting together. Yet our estimates of the impact of flows on GDP are broadly consistent with other academic findings (see the technical appendix for a complete literature review), even if other studies typically focus on one type of flow for a set of countries. This is an important distinction, because we know that flows are correlated with one other. We performed a principal component analysis on how flows move together. The results indicate that one component captures more than 60 percent of all flows. This means that examining the impact of only one type of flow may overestimate its impact on GDP by picking up observed effects from other flows. Nonetheless, our findings for the impact of each individual flow are in line with academic findings. In general, our estimates are smaller since we correct for the effect that one flow may have on another. For example, Wacziarg posits that every increase in trade of ten percentage points relative to GDP adds 0.7 percent to GDP, while our model shows an increase of 0.5 percent.90 Har et al. find that every increase in FDI of ten percentage points relative to GDP raises GDP by as much as 0.5 percent, vs. 0.4 percent in our model.91 Boubtane et al. observe that every ten percentage point increase in immigration raises GDP by one percentage point in OECD countries.92 Our model shows a negligible or even negative impact at the global level, but it similarly finds that migration has a positive impact on productivity in advanced economies. We have not found a study that measures the effect of cross-border data flows on GDP. Due to the difficulties of collecting data on this subject, most of the literature has assessed the GDP impact of increasing Internet penetration. For example, Meijers finds that a ten percentage point increase in Internet penetration leads indirectly to a 0.17 percentage point increase in economic growth.93 Our model shows that a ten percentage point increase in data flows relative to global population raises GDP by 0.2 percent. The economic impact of flows comes mainly from raising productivity, and data flows are central to this effect Global flows can raise GDP growth either by increasing productivity or by increasing the amount of capital and labor used in production. We find both effects at work, although different flows act through different channels, and the productivity effect has by far the largest impact on GDP. Traditional flows of goods and FDI raise productivity in an economy but do not necessarily increase the capital and labor inputs used. This suggests that in the long term, participation in flows promotes more efficiency and innovation, perhaps stemming from increased competition or the faster spread of best practices. We find that data flows support both productivity improvement and increased capital and labor inputs. This indicates that so far, data flows and digitization have raised net employment within countries rather than reducing it, contrary to conventional wisdom. The benefits of data flows are not due to replacing workers (see Box 4, “The impact of global flows on employment”). 90 Romain Wacziarg, “Measuring the dynamics gains from trade,” World Bank Economic Review, volume 15, number 3, October 2001. 91 Wai Mun Har, Kai-Lin Teo, and Yee Kar Man, “FDI and economic growth relationship: An empirical study on Malaysia,” International Business Research, volume 1, number 2, April 2008. 92 Ekrame Boubtane, Jean-Christophe Dumont, and Christophe Rault, Immigration and economic growth in the OECD countries, 1986-2006, IZA discussion paper number 8681, November 2014. 93 Huub Meijers, “Does the Internet generate economic growth, international trade, or both?” International Economics and Economic Policy, volume 11, issue 1, February 2014. 79McKinsey Global Institute Digital globalization: The new era of global flows Box 4. The impact of global flows on employment Our own analysis, like that of academic researchers, shows that trade, foreign direct investment, people flows, and data flows have a net positive effect on growth and prosperity in the long run. As countries specialize in what they do best, global GDP rises and employment growth should follow.1 But globalization also delivers an element of creative destruction as it exposes local industries to international competition and disruptive business models. This is ultimately a healthy dynamic that spurs efficiency and innovation, but it may displace some local industries and workers in the process. In the long run, countries that participate in global flows will find new channels for growth, and workers who lose their jobs in one industry should find opportunities elsewhere. Yet this process does not always play out neatly and quickly. Workers in a shrinking industry must gain new skills to be employed in other sectors—and those opportunities may not be in the same geographies, requiring them to relocate. This can have profound effects on entire communities that lose traditional industries to global trade and competition. Recent research has found, for instance, that increased US goods trade with China since 2000 resulted in the loss of thousands of manufacturing jobs in the United States and that the effects have persisted for more than a decade.2 Providing support to workers who are affected—and to broader local communities that suffer when industries are lost—has too rarely been treated as an urgent policy priority. But the societal cost of neglecting this issue grows over time and spurs negative sentiments toward globalization. Traditional labor market policies and training systems in most countries are not prepared to deal with large-scale, rapid changes.3 Workers displaced by trade (and similarly by automation) will need clearer paths to new roles. This means creating widespread access to accelerated training programs that will help them acquire skills that are in demand. A range of policy options are possible, but all governments must address the issue. One possible policy response is wage insurance.4 Germany may offer a useful model for other countries as well. It is one of the world’s most connected countries, ranking fourth in our global index. But it has avoided widespread unemployment, even at the height of the global recession, by providing income support, paying companies to retain workers, and taking a proactive approach to labor market reforms and reemployment services.5 Interestingly, our econometric analysis suggests that global data flows increase employment in the long run. While the goods trade changes the location of some production activities, data flows enable innovation, remote work, and new types of economic activity that did not exist before. It is too early to say definitively how cross-border data flows will affect employment in the future, particularly given the unpredictable nature of technology innovation, but digital platforms could be a welcome source of opportunity to find work for some individuals. They could also help to deliver some of the educational and training programs that individuals will need to reinvent themselves with new skills. 1 For a discussion of when this may not be the case, see Paul Samuelson, “Where Ricardo and Mill rebut and confirm arguments of mainstream economists supporting globalization,” Journal of Economic Perspectives, volume 18, number 3, summer 2004. 2 David H. Autor, David Dorn, and Gordon H. Hanson, “The China syndrome: Labor market effects of import competition in the United States,” American Economic Review, volume 103, number 6, 2013, and David H. Autor, David Dorn, and Gordon H. Hanson, The China shock: Learning from labor market adjustment to large changes in trade, NBER working paper number 21906, January 2016. 3 For a deeper discussion of labor market inflexibility and possible responses to it, see A labor market that works: Connecting talent with opportunity in the digital age, McKinsey Global Institute, June 2015. 4 See, for example, Lori G. Kletzer, “Why the US needs wage insurance,” Harvard Business Review, January 25, 2016. 5 Marco Caliendo, Income support systems, labor market policies and labor supply: The German experience, IZA discussion paper number 4665, December 2009. 80 McKinsey Global Institute 4. Global flows boost economic growth COUNTRIES ON THE PERIPHERY HAVE THE MOST TO GAIN FROM CROSS-BORDER DATA FLOWS Finally, we assess how a country’s position in the global network of flows changes the economic impact. We do this using two different measures of network centrality: one that measures the number of a country’s bilateral connections compared with the total number possible, and one that measures how connected a country’s trading partners are (a concept known as eigenvector centrality). Our last report tested centrality only within trade flows, given data limitations within other types of flows. We found that the benefits to GDP growth were up to 40 percent higher for countries that were more central in the global network of goods trade—with more bilateral partners and partners that were themselves more connected—than for countries with only a few trading partners.94 This finding showed that it is better to have a broader network of connections and to connect with countries that are more central to the network than to trade solely with a few neighbors. This report extends our centrality analysis to cross-border data flows—and the findings do not show the same effects as in trade flows. Data flows are still in a nascent stage, with links between countries that are less dense and have less reciprocity. In the case of data flows, we find that the benefits to GDP growth for countries at the periphery of the global network are actually higher than for countries at the center of the network. Periphery countries may be using their exposure to data flows to enable broader participation. Moreover, data flows offer access to all the world’s knowledge, information, and innovations. For economies that have been relatively disconnected, the arrival of new digital platforms can have a bigger ripple effect on trade in goods and services than it might have in advanced economies that already have more extensive trade links in place. Our analysis shows that 15 to 30 percent of the GDP impact of data flows comes from consumers, while the remainder comes from B2B linkages within value chains. As policy makers in many countries consider how their countries should participate in a more digital global economy, many are seeking to create the “next Silicon Valley.” Others are erecting barriers to global digital platforms to allow domestic platform providers to grow. But we find that countries benefit from receiving cross-border digital flows as well as producing them. In other words, countries do not need to transform themselves into digital content or platform producers to benefit from data flows. 94 Global flows in a digital age: How trade, finance, people, and data connect the world economy, McKinsey Global Institute, April 2014. Countries at the periphery of the world’s digital networks stand to gain even more than those at the center. 81McKinsey Global Institute Digital globalization: The new era of global flows SOME LAGGING COUNTRIES COULD INCREASE GDP BY 50 PERCENT IN THE LONG TERM BY RAISING PARTICIPATION IN FLOWS With clearer evidence in hand that global flows drive GDP growth, it becomes apparent that there is an opportunity cost associated with limited participation—and closing the gap between leaders in global flows and laggards would create substantial economic value. To determine the size of this unrealized opportunity, we calculate the value that countries realized by increasing participation in each type of flow relative to the size of their economies from 2003 to 2013 (the latest year for which there are global data for all flows). We find that countries in the top quartile increased their flow of goods relative to GDP at an average of 3 percent annually, while goods flows grew at only 1 percent for the bottom quartile. The top-quartile countries increased FDI flows relative to GDP by 5 percent of GDP annually during this period, while those flows shrank by 8 percent annually for countries in the bottom quartile. A similar pattern holds for data flows and people flows. We calculate that if the bottom three quartiles of countries had increased participation in each of the flows at the same rate as the top quartile over the past ten years, global GDP would be some $10 trillion, or 13 percent, higher today. In other words, limited participation in global flows by many countries had a real cost to the world economy. Because few countries are consistently strong across all five flows, there is substantial potential value for many to capture. As discussed in Chapter 3, we find that most countries are relatively connected within one or two flows: for instance, Luxembourg in financial flows and services, or Belgium in goods and service trade. Only a few countries, such as Germany or the United States, rank highly across all flows. In addition, a large set of countries, mostly in the developing world, post low or barely average scores across all flows, and they are catching up to the most connected countries at a very slow pace. Some countries have made explicit moves to open their economies and participate more fully in global flows. For lagging countries, the potential payoff from a well-targeted strategy of opening may be substantial (Exhibit 33). If India, for instance, had accelerated its participation in all types of global flows to match the top-quartile countries over the past ten years, its GDP would have been $1.2 trillion higher (or 58 percent larger) by 2014. Despite its thriving business process offshoring sector, India ranks only 70th in the world for data flows—down from 64th in the previous edition of the index, indicating that other countries have increased data flows faster than India. According to the Internet and Mobile Association of India, the country passed the benchmark of 350 million Internet users in 2015, and penetration is continuing to grow. It will not be simple to create the digital infrastructure and skills development necessary to bring the rest of India online, but the government’s Digital India initiative aims to accelerate the process of connecting the country’s vast rural population to the world economy. Limited participation in global flows by many countries has had a real economic cost. 82 McKinsey Global Institute 4. Global flows boost economic growth Similarly, Brazil could have added some $1.4 trillion to its GDP over the past ten years, making its economy 60 percent bigger by 2014, by accelerating its participation in all types of global flows. Previous MGI research has explored Brazil’s relative lack of global connectedness. Brazil has not capitalized on its proximity to the lucrative US consumer market, for instance. In 2012, even before the current commodity price decline, exports were equivalent to 13 percent of GDP, far below India (24 percent) and Mexico (33 percent). Brazil’s imports are also lower, at 10 percent of GDP, compared with 22 percent for India and 32 percent for Mexico. A number of barriers have constrained trade growth, including poor road and rail infrastructure, cumbersome procedures and inadequate capacity Exhibit 33 Lagging countries could realize enormous growth potential by increasing their participation in global flows SOURCE: McKinsey Global Institute analysis Output gap % of GDP >75
51–75
26–50
1–25
<1
No data
83McKinsey Global Institute Digital globalization: The new era of global flows
at its ports, and high import tariffs and taxes. Brazil is particularly low on people flows,
ranking only 125th in the world for migrants and travelers. Since 1999, the country has lost
38 percent of its share of South America’s inbound tourism and 30 percent of its share of
world inbound tourism.95
These missed opportunities are not lost forever, however—and digitization is creating new
ways for countries to participate. As growth in conventional flows of goods and finance
flatten, the next wave of growth from globalization will come from data and information flows.
Countries can increase their exposure to these flows by expanding Internet penetration and
creating thoughtful frameworks that allow data to move both securely and freely across their
borders. Conversely, academic studies point out that restrictions on data flows may lower
GDP growth by one to two percentage points.96 Chapter 6 contains further discussion of
these and other public policy implications.
•••
Participating in global flows is not a panacea for the other factors that may dampen a
country’s economic growth, such as an uncompetitive business environment, weak rule of
law, or corruption. Still, countries that seal themselves off from global flows—and particularly
data flows—are forgoing important sources of growth. Continuing to expand the network
of global data flows is possible because this new version of globalization is not merely a
zero-sum game in which countries compete for low-cost manufacturing: one country’s
participation in data flows does not necessarily come at another’s expense; it can actually
increase economic growth across the board.
95 Connecting Brazil to the world: A path to inclusive growth, McKinsey Global Institute, May 2014.
96 Matthias Bauer et al., The costs of data localization: Friendly fire on economic recovery, ECIPE occasional
paper number 3/2014, May 2014.
© Getty Images
After expanding across borders in pursuit of new international markets and the advantages
of scale, many major corporations now derive more than half of their revenue internationally.
But along the way, they may have incurred a “globalization penalty.” Managing across
multiple geographies with different cultures, languages, regulations, and tax regimes is no
small challenge. It often involves going up against local competitors that may have deeper
market insights and better ability to execute on their own turf. The costs of coping with
complexity can take a toll on the bottom line as well as organizational health, making it
harder to create a cohesive culture and strategy.97
Digitization can tame complexity, however, allowing large companies to manage their global
operations in a leaner and more efficient way. Using digital platforms and tools effectively
can enable companies to sell in far-flung but fast-growing markets while keeping virtual
teams connected in real time. Companies have new ways to identify the best suppliers,
inputs, and talent from around the world.
The convergence of globalization and digitization means that the world is changing
rapidly—and business leaders will need to reassess their organization, strategy, assets,
and operations accordingly. The approaches that worked for going global even ten years
ago may no longer be relevant. As digitization changes how companies think about what
should be global and what should be local, many are reevaluating past decisions about their
footprint and international market strategies. The successful global companies of the future
will have a very different look than those of the past.
A SUCCESSFUL DIGITAL GLOBALIZATION STRATEGY CONSIDERS SEVEN
KEY DIMENSIONS
Digital innovation has greatly expanded the toolbox companies can use to globalize
their market reach and operations (Exhibit 34). Because of these new capabilities,
business leaders have an opportunity to rethink organization and strategy—starting with
the fundamental question of what kind of global footprint is optimal. For many years,
globalization centered on building production facilities around the world to take advantage
of low-cost labor, but digitization is changing that equation. Additionally, while in the last era
of globalization products were significantly tailored to local markets, some companies are
moving toward global products and global product launches. As global value chains shift,
decisions about where to base production are becoming more nuanced than simply looking
for the lowest labor costs. To compete effectively, companies will have to build new types
of assets, including the right arsenal of digital capabilities—and meeting that requirement
will require a more global approach to finding and deploying the best talent. Finally, digital
globalization calls for building more resilience into organizations.
97 Martin Dewhurst, Jonathan Harris, and Suzanne Heywood, “Understanding your ‘globalization penalty,’”
McKinsey Quarterly, July 2011.
The successful global companies of the future will
have a very different look than those of the past.
5. COMPETING IN A DIGITAL
GLOBAL LANDSCAPE
86 McKinsey Global Institute 5. Competing in a digital global landscape
Do your footprint and organizational structure make sense in a more
digital world?
Companies once expanded globally by building a replica of their entire firm in each country
where they operated, complete with human resources, finance, sales and marketing, and
product development departments. Each country operation mirrored the vertical structure
of the parent entity. But now digital technology is expanding the options for how to organize
a global company. Companies can establish global hubs for some functions and combine
those structures with smaller dispersed sites and sales offices around the world. This option
captures the advantages of efficiency and scale for some functions but maintains sales and
marketing teams in proximity to local customers. The choices companies make about their
footprint and how to manage it have never been wider—and the payoff, in terms of both
reach and agility, has never been bigger.
Many multinationals are centralizing global functions and back-office operations. In human
resources, for example, self-service digital platforms can draw on data flows between
countries to handle many overarching issues, while regional managers can retain discretion
over decisions such as hiring. Singapore-based Flextronics migrated its fragmented
human resources systems for 200,000 workers in 25 countries into one global platform that
Exhibit 34
SOURCE: McKinsey Global Institute analysis
Cross-border implications of digitization
Flow type
Data Goods Services Finance FDI
Remote
monitoring
Remote tracking
Remote maintenance
Supply-chain
management
Remote inventory
management
Supplier management
Access to
global
markets
Cross-border access to
customers
Cross-border access to
labor
Cross-border access to
finance
Business
operations
and strategy
Centralized back-office
operations
Cross-border digital
payments
Real-time communications
and collaboration
Data sharing and analytics-
driven decision making
Digitization is transforming business models in ways that enable more cross-border activity
87McKinsey Global Institute Digital globalization: The new era of global flows
automatically supports 14 languages.98 The Canada-based Four Seasons chain, which
has 42,000 employees in hotels around the world, similarly moved to a globally scaled
cloud-based HR system that offers organizational consistency where possible and local
customization where necessary.99
Companies can also create virtual teams that span borders, using digital tools for remote
collaboration (such as Box and Slack) and customer relationship management (such as
Intercom and Zendesk). Unilever, for example, used technology solutions to streamline
some 40 global service lines, including financial reporting, internal communications, market
research, requisitions and payments, and HR. It created global, virtual delivery organizations
with team members who meet via video conference. In the past, Unilever had more than
400 intranets spanning different countries, product groups, and functions—a structure that
led to unnecessary expense and misaligned communications. These were replaced with
one global intranet that is accessible in more than 20 languages.100
There is no one-size-fits-all solution that will work for every company, however. In R&D and
product development, for instance, Apple concentrates its engineering and design talent in
Cupertino, California, and brings foreign hires to work there. Google has gone another route,
with engineering and design teams in major tech hubs around the world.
Sometimes proximity to specialized talent pools can shape these decisions. Boeing
maintains a design center in Moscow to take advantage of Russia’s abundance of
aerospace scientists and engineers. Cisco established a global development center in
Bangalore to tap into the local concentration of engineering talent. This center, Cisco’s
largest facility outside the United States, houses R&D operations focused on developing
new disruptive technologies and strategies for emerging markets. More than 1,000 of the
company’s patents have been filed from India as a result.101
Footprint decisions can also be driven by market demand and expansion opportunities.
As emerging economies build their health-care systems, virtually all major Western
pharmaceutical companies have established R&D operations in Asia. AstraZeneca, for
instance, is creating a new global hub for pharmaceutical development in Shanghai to
complement its R&D centers in the United Kingdom and Sweden; it already has 11,000
employees in China.102 The main R&D hub for Novartis is co-located with its global
headquarters in Switzerland, but the company also has research facilities in China, India,
Japan, Singapore, and the United States. Pharmaceutical companies in emerging markets
are similarly expanding their international R&D capabilities. India’s Sun Pharmaceuticals, for
example, has R&D centers in Israel and the United States.
In a more digitally connected age, some companies are breaking from the traditional model
of having one global headquarters location. Lenovo is incorporated in Hong Kong and
has headquarters in both Beijing and North Carolina; the company also bases some top
executives and research centers in other major hubs across the globe. GM has a China
office in Shanghai to serve what is now the world’s largest auto market, plus an international
headquarters in Singapore to oversee operations in the rest of Asia, Africa, Australia, and
the Middle East. Honda has established multiple independent manufacturing subsidiaries,
98 “Flextronics completes world’s largest deployment of a core HR system in the cloud,” Workday corporate
press release, December 9, 2011.
99 Matthew Finnegan, “Four Seasons chooses Workday HCM cloud over ‘costly’ on-premises ERP systems to
manage global workforce,” ComputerworldUK, October 20, 2015.
100 Pascal Visée, “The globally effective enterprise,” McKinsey Quarterly, April 2015.
101 Cisco India Overview, corporate fact sheet, 2016.
102 “AstraZeneca continues strategic investment in China to accelerate delivery of innovative biologics and
targeted medicines,” corporate press release, December 16, 2015. See also “Innovating in China’s pharma
market: An interview with AstraZeneca’s head of R&D in Asia and emerging markets,” McKinsey.com Insights
& Publications, February 2012.
88 McKinsey Global Institute 5. Competing in a digital global landscape
including Honda China, Honda of North America, and Honda Europe. These divisions are
run locally, and each market determines which models to sell.
Digitization also enables less capital-intensive business models. Companies that deliver
digital goods and services can enter new international markets without establishing a
physical presence at all. When Netflix created a subscription model for online streaming
of video content in 2007, it gained the ability to add global customers without setting up
full-fledged physical operations around the world (after receiving local regulatory approval).
By the end of 2015, the company had broadened its international reach to more than 190
countries. As it did so, it closed all of its data centers and moved all of its streaming services
onto the public cloud.103
Should you offer one product line around the world or customize for
local markets?
In many industries, companies that sell into a range of global markets have expanded their
product portfolios with tailored offerings that appeal to local consumer preferences and
are sensitive to their price points. South Korea’s LG, for example, markets original products
in India, including appliances with programming menus in local languages, large washing
machines for big families, and microwaves with one-touch “Indian menu” functions.104
Mondelez found that the Oreos beloved by Americans did not have the same appeal
elsewhere, so it tweaked the cookie’s formulation, size, shape, packaging, and price points
in China and India to appeal to local palates and preferences.105
In some industries, product tailoring is driven by local regulatory requirements or language
differences. Food companies, for example, must meet one set of requirements for their
products to be certified as halal in Malaysia and a different set of standards in Indonesia (and
neither country’s requirements are in line with Saudi Arabia’s standards).106
But some companies eschew this strategy and offer truly global products that are the same
everywhere in the world. Apple, for instance, offers just three models of its iPhone and iPad,
all with consistent design and branding no matter where they are sold; settings can be
reconfigured to change the language. Even its retail stores have the same design aesthetics
the world over. Luxury brands typically take the same strategy. Brands such as Gucci,
Burberry, and Prada position themselves as aspirational for newly affluent consumers in
emerging markets and provide the same products and customer experience everywhere.
Creating a single global product is not just for companies that target upscale customers.
Now that social media allows everyone to see the latest celebrity trends, many consumers
around the world want the same styles. Some mass market retailers, including H&M and
Ikea, offer a consistent brand, store format, and core product selection worldwide but tweak
a small share of the merchandise mix to reflect local differences.
Digital products often follow the one-world model. Google, for example, has search,
mapping, and e-mail products that are not designed with any particular set of regional
customers in mind (although different languages are available). Facebook, Uber, Airbnb,
and various e-commerce marketplaces have simply scaled up their digital platforms in
country after country with limited customization. The success of this model has encouraged
digital entrepreneurs to think about designing globally scalable products and expanding
103 Caroline Donnelly, “Netflix shuts down final data centre to go all-in on public cloud,” Computer Weekly, August
17, 2015.
104 Winning the $30 trillion decathlon: Going for gold in emerging markets, McKinsey & Company, August 2012.
105 See Jeff Beer, “Marketing to China: Oreo’s Chinese twist,” Canadian Business, November 22, 2012; and
“Smart cookie,” case study from the Aditya Birla India Centre of London Business School, Business Today
(India), March 31, 2013.
106 Southeast Asia at the crossroads: Three paths to prosperity, McKinsey Global Institute, November 2014.
89McKinsey Global Institute Digital globalization: The new era of global flows
internationally much earlier in their life cycle than earlier generations of companies (a topic
discussed in Chapter 2).
Digitization has also created a shift toward launching products globally as opposed to
staggering releases country by country. The entertainment industry is a case in point. For
many years, Hollywood studios waited until after a movie’s US run to release it overseas,
where the foreign box office could still compensate for domestic disappointments. The
highest-grossing movie of 1995, Die Hard: With a Vengeance, was screening in only
three countries within ten days of its US release date; they represented only 10 percent
of the total markets where it would eventually be released. In 2015, Star Wars: The Force
Awakens launched in every major market (80 countries) except China in the same week,
and more than half of its record-breaking debut weekend box office came from international
ticket sales (Exhibit 35). The information transparency offered by the Internet means
that consumers around the world can see reviews immediately. There is no longer an
opportunity to tweak and remarket; if a movie bombs in one place, it will be a global bomb.
80
countries where
the latest Star Wars
was released in its
first week
Exhibit 35
Hollywood releases illustrate the growing trend toward simultaneous global launches
SOURCE: IMDB.com; boxofficemojo.com; McKinsey Global Institute analysis
Simultaneous release of highest-grossing Hollywood movies in multiple countries1
1 W ithin 10 days of release data; excludes movie premieres. Highest-grossing movies in their respective years.
2 Rounded.
1990
1
1995
3
2000
5
2005
43
2012 2015
80
Total number
of countries
in which
movie was
released2
30 30 60 65 80 80
70
90 McKinsey Global Institute 5. Competing in a digital global landscape
Music, games, and books are following the same trend toward global releases. As social
media exposes consumers from around the world to what is available, products have a
chance to go viral. In 2015, Adele’s song “Hello” racked up 50 million views on YouTube in
its first 48 hours, and during its first week of release, her smash album 25 was No. 1 on the
download list of iTunes stores in 110 countries.107 An additional incentive for global product
releases is to protect intellectual property and capitalize on the initial wave of demand before
piracy can strike.
This is not to say that tailoring for local markets is dead. The need to create offerings at
a lower price point for emerging economies is still valid. Microsoft, for instance, recently
announced a new Nokia smartphone with extended battery life that will retail for around
$20, with an eye toward selling it in emerging markets.108 GE Healthcare has produced
low-cost products such as a handheld ultrasound device and a CT scanner with emerging
markets specifically in mind; it recently announced the formation of a Sustainable Healthcare
Solutions business line to continue this focus on “frugal” product development.109 In fact,
some of the low-cost offerings developed for emerging markets can now boomerang back
to advanced economies, where there is also demand for value-engineered products.
Many global automakers are attempting to balance the need for localization against the
need for global scale in a complex manufacturing process by using a platform approach.
They rely on a set of common underlying designs that can be customized by swapping
certain components to create differentiated models. Today most major carmakers
are whittling down the number of platforms across their international manufacturing
operations—a streamlining effort that could produce billions in savings.
Some companies are finding artful ways of combining the global and the local in their
marketing initiatives. The right approach can put a more intimate or relatable face on a
global brand or take a fundamentally local experience and imbue it with a broader, more
aspirational message. Starbucks, for example, launched a “Meet Me at Starbucks”
global campaign through a mini-documentary shot in 59 different stores in 29 countries.
The project, which shows a day in the life of a Starbucks store to emphasize its role as a
community gathering place around the world, spans New York, Rio, Bogotá, Singapore,
Beijing, Mumbai, Toronto, Paris, and Berlin.110 The long-running “keep walking” campaign
for Diageo’s Johnnie Walker scotch was adapted for the Chinese market by associating
it with the traditional Confucian saying that “the journey of a thousand miles begins with a
single step.”
As global value chains shift, do your suppliers and customers channels still
make sense?
Multinationals and their long, intricate supply chains are the driving force behind the world’s
flows of goods. Digital tools can orchestrate a multitude of vendors stretching around the
globe with greater precision and efficiency, opening up new possibilities for procurement.
Companies such as Cisco and P&G have built “control towers” that offer up-to-the-minute
visibility across complex global supply chains. These hubs synthesize information from
sensors, actuators, RFID tags, GPS tracking, and more into dynamic models that can help
managers evaluate alternatives instantly if risks or bottlenecks arise.111
107 Clarisse Loughrey, “Adele’s new album 25 is No. 1 on iTunes in almost every country in the world,” The
Independent, November 26, 2015.
108 “The new Nokia 105 helps give people a voice,” Microsoft corporate blog, June 3, 2015.
109 “GE Healthcare announces $300 million commitment to support emerging market health,” corporate press
release, September 23, 2015.
110 Maureen Morrison, “Starbucks launches first brand campaign, ‘Meet me at Starbucks,’” Ad Age, September
29, 2014.
111 See, for example, the interviews with supply chain executives in Bob Trebilcock, “What does it take to remain
a supply chain leader?” Supply Chain Management Review, January 2, 2015; and Steve Banker, “Procter &
Gamble’s futuristic control tower environment,” Forbes.com, July 1, 2015.
91McKinsey Global Institute Digital globalization: The new era of global flows
But even as technology enables more complex procurement and collaboration, the
importance of different factor costs is shifting. After years of choosing production locations
largely on the basis of where low-cost labor is available, many manufacturers are beginning
to reassess those decisions. Factors such as logistics costs, lead time, productivity,
consumer preferences, and proximity to other company operations are being given
greater weight.
For some products, low-cost labor will continue to be the decisive factor. As China’s wages
rise and the country makes a push to move up the value chain into more innovative, higher-
value-added industries, more of its manufacturing business is up for grabs. Japan, for
instance, has been shifting FDI from China to Southeast Asia. Now even some Chinese
companies are shifting textile and apparel production to Africa.
However, more than two-thirds of global manufacturing activity takes place in industries
that tend to locate close to demand—and companies have to consider emerging markets
as sources of that demand, not just supply. As incomes rise in these regions, demand
is fragmenting as customers expect greater variation and more after-sales service.112
According to a recent UPS survey, approximately one-third of high-tech companies are
moving manufacturing or assembly closer to end-user markets—and this number is up by
25 percentage points from 2010.113
In the future, 3D printing could change the very definition of what constitutes an intermediate
good, disrupting logistics companies. UPS, for example, recently launched a pilot program
to experiment with offering industrial-grade 3D printing services.114
Do you have the right assets to compete digitally and globally?
Companies will need new types of assets to succeed in this new landscape. Building digital
platforms and data centers may be critical for a growing range of companies, not just the
Internet giants. Advanced digital capabilities are a major source of competitive advantage,
and even traditional industries that lagged behind in the first wave of digitization are
beginning to transform rapidly.
Consider how big data analytics are transforming manufacturing. The manufacturers at
the leading edge are digitizing and connecting their equipment, facilities, fleets, and other
physical assets with the Internet of Things.115 They are developing expertise in big data
analytics to generate insights from the flood of data being collected. GE, for example, is
boosting investment to position itself as a leader in the industrial Internet. The company
hopes to improve productivity, innovation, and customer retention in its own manufacturing
operations and to become a provider of related services, applications, and platforms to
other industrial firms.
Other companies are taking a similar path. Rio Tinto, for example, transmits data
continuously from its mines, processing plants, and vehicle fleet to “excellence centres”
located in Brisbane, Australia. Analysts interact with this data on some of the largest touch
screens in the world, creating models to head off potential production delays before they
occur and making decisions about operational efficiency almost in real time. The company
plans to open a third excellence centre for analytics in India in early 2016.116
112 Katy George, Sree Ramaswamy, and Lou Rassey, “Next-shoring: A CEO’s guide,” McKinsey Quarterly,
January 2014. See also Manufacturing the future: The next era of global growth and innovation, McKinsey
Global Institute, November 2012.
113 Change in the (supply) chain, United Parcel Service, 2015.
114 Lindsay Ellis and Laura Stevens, “UPS tests a 3D printing service,” The Wall Street Journal, September
18, 2015.
115 Digital America: A tale of the haves and have-mores, McKinsey Global Institute, December 2015.
116 Carly Leonida, “Rio Tinto to open Mining Excellence Centre,” Mining Magazine, February 25, 2015.
1 OF 3
high-tech
companies report
plans to move
production closer
to end-user
markets
92 McKinsey Global Institute 5. Competing in a digital global landscape
Businesses in all industries need to take a fresh look at the assets they hold, including
customer relationships and market data, and consider whether there are new ways to
monetize them given the emergence of new markets and new technologies. Alibaba,
for instance, is at the center of China’s e-commerce ecosystem and has a vast pool
of transactional data on the vendors that operate on its platform. Building on these
advantages, the companies has moved into new areas such as mobile payments and small
business financing. The insurance industry could similarly harness its sophisticated data
pools on different forms of risk to create new products and services.
Another key asset that many companies undervalue is an effective online presence. A
passive corporate website is no longer sufficient. Companies also need a responsive
social media voice and perhaps even their own proprietary platform. In consumer-
facing industries, customer reviews on social media are increasingly important in driving
business—and conversely, poor reviews can cause harm. Airline complaints on Twitter can
go viral, and bad reviews on TripAdvisor or Yelp can cost hotels and restaurants dearly as
growing numbers of international travelers rely on these sites to shape spending decisions.
Many e-commerce sites now invite customers to post product reviews, and integrating
their feedback quickly into the next product development cycle can pay real dividends.
Creating a 24/7 team to monitor social media, handle customer complaints, and maintain a
reputation is critical.
Are you ready for a new world of digitally accelerated global competition?
Corporate competition has intensified dramatically as emerging-market giants and digital
disruptors go global. Both sets of competitors are lean, agile, and aggressive; both have
cost advantages that enable them to take on established industry leaders. They are also
demonstrating the ability to operate fluidly across geographic and sector boundaries.117
As digital technologies reduce the time, capital, and minimum scale needed for startups to
compete globally, these dynamics are accelerating. From 1965 to 2012, the “topple rate” at
which companies lose market-leading positions increased by almost 40 percent—and the
world is only speeding up from here.118
Emerging-market companies are breaking into the top ranks of their industries globally.
Between 1980 and 2000, the share of Fortune Global 500 companies based outside
developed regions stayed relatively flat, at 5 percent. By 2010, this share was up to
17 percent of the total, and it climbed further still to reach 26 percent in 2013. Based on
projected growth by region, MGI has forecast that emerging economies will account for
more than 45 percent of the Fortune Global 500 by 2025 (Exhibit 36).119
Unlike publicly listed companies in the United States and Europe, many of the new
emerging-market competitors are state- or family-owned, which can give them the flexibility
to pursue longer-term strategies. Having grown up in difficult operating environments, they
have a natural advantage in other fast-growing emerging markets. The Chinese telecom
firm Huawei, for example, has become the third-largest smartphone vendor in the world,
with a strong presence in markets from Africa and India to Myanmar. Indonesia’s Indofood
has successfully introduced its Indomie noodles across Africa, becoming the most popular
brand in the huge Nigerian market.
As the global playing field becomes more crowded with international companies, the war for
talent is taking on another dimension. A recent survey of US executives found that almost
40 percent of companies had missed business opportunities in the past five years due to
lack of international competencies. More than a quarter of companies indicate that it is
117 Playing to win: The new global competition for corporate profits, McKinsey Global Institute, September 2015.
118 John Hagel III et al., 2013 Shift Index metrics: The burdens of the past, Deloitte, 2013.
119 Urban world: The shifting global business landscape, McKinsey Global Institute, October 2013.
93McKinsey Global Institute Digital globalization: The new era of global flows
difficult to find US talent with the international knowledge, expertise, and language skills
needed to manage global operations.120 Online talent platforms are creating a more global
labor market—and this development gives workers more mobility and gives companies new
ways to poach their competitors’ top performers. Attracting and retaining valued employees
is a growing issue for Western multinationals operating in emerging economies. These
companies were once considered the most prestigious employers in these countries, but
now local firms are becoming global players themselves, and they can offer competitive
compensation and career paths. One survey found that 34 foreign firms were listed among
the 50 most attractive employers in China in 2004, but only 15 made the list in 2014.121
120 Shirley J. Daniel, Fujiao Xie, and Ben L. Kedia, 2014 US business needs for employees with international
expertise, presented at the Internationalization of US Education in the 21st Century research conference in
Williamsburg, Virginia, April 11–13, 2014.
121 Universum global survey of most attractive employers, 2014.
Exhibit 36
Number of Fortune Global 500 companies1
By 2025, emerging regions are expected to be home to almost 230 companies in the Fortune Global 500,
up from 85 in 2010
SOURCE: MGI CompanyScope; McKinsey Global Institute analysis
1 The Fortune Global 500 is an annual ranking of the top 500 companies worldwide by gross revenue in US dollars.
2 All emerging regions with the exceptions of China and Latin America combined until 2000.
3 Fortune Global 500 share in 2025 projected from revenue share of countries in 2025.
NOTE: Numbers may not sum due to rounding.
12
54
120
10
34
13 15
8
26
78
11
8
12
26
South Asia
2
Southeast Asia
Africa
Latin America
Other emerging2
2025E3
271
476
2000
6
1990
477
6
1
1980
477
2010
415
41
Developed regions
China
Total in
emerging
regions
Emerging
regions
23 23 24 85 229
94 McKinsey Global Institute 5. Competing in a digital global landscape
Technology firms represent another huge source of competition. Some of the truly disruptive
players are siphoning value out of industries and giving it away for free to consumers as a
way to build their positions. Skype, for instance, shifted some $37 billion to consumers in
2013 alone by offering free international calls.122 Many technology-enabled firms are blurring
traditional industry boundaries as they add new business lines. Alphabet has expanded into
areas well beyond Google’s original search and advertising businesses, including longevity
and biotech research, smart home products, venture capital investing, and high-speed
Internet fiber services.
The largest Internet platform operators are giving rise to yet another competitive threat: huge
pools of SMEs that can now reach customers around the world. Thousands of Chinese
manufacturers operating on Alibaba now have global reach, as do thousands of SMEs using
eBay. The smaller firms operating within global e-commerce marketplaces now have the
resources and reach to cherry-pick customers from industry incumbents.
The Internet is also creating global pricing pressures. Consumers can comparison shop
across multiple channels and markets, making it more difficult for companies to implement
tiered pricing strategies. Apparel brands that could maintain a luxury image and command
higher prices in certain markets now have difficult decisions to make. To get around this
issue, some supermarkets and other retailers have introduced private-label brands. This
trend is even spreading to major B2B distributors, with companies such as Grainger and
Sysco increasingly emphasizing their own private labels.
The Internet has also cut into the window of exclusivity companies once enjoyed on
new products and services. Similar versions can be launched in new markets before the
originator has time to scale up. Just months after the launch of Uber, there were similar
companies operating in China and India. Rocket Internet, a German startup incubator,
specializes in financing online businesses that bring successful business models from one
country to new international markets.123
How do you manage new types of risk in a more digital and
interconnected world?
The impact of external shocks is magnified in a more interconnected world—and ripple
effects spread even faster in a more digital world. The 2008 financial crisis showed how
rapidly the linkages between the world’s capital markets can allow contagion to spread.
Just a few years later, a series of natural disasters underscored the vulnerability of long
global supply chains. Toyota’s production, for example, took a major hit in the aftermath of
the 2011 Japanese earthquake and tsunami—because of damage not to its own factories
but to the operations of its suppliers. In fact, the ramifications were felt around the world,
with component shortages causing the temporary shutdown of GM and Ford plants in the
United States.124 Heavy monsoons in Thailand that same year produced flooding in a region
that produced nearly half of the world’s supply of hard drive disks, sending global prices
soaring.125 Global executives polled in McKinsey’s latest survey on economic conditions
cited geopolitical instability as the greatest risk to growth.126
122 Playing to win: The new global competition for corporate profits, McKinsey Global Institute, September 2015.
123 Sarah Gordon and Dan McCrum, “Rocket Internet: Waiting for lift-off,” Financial Times, October 19, 2015.
124 Bill Canis, The motor vehicle supply chain: Effects of the Japanese earthquake and tsunami, Congressional
Research Service report for Congress, May 2011.
125 Rich Miller, “How floods in Thailand made AWS rethink its supply chain,” Data Center Frontier, October
8, 2015.
126 McKinsey survey of more than 2,000 executives from companies representing a range of geographies,
industries, and sizes. Economic conditions snapshot, December 2015: McKinsey Global Survey results,
McKinsey & Company, December 2015.
95McKinsey Global Institute Digital globalization: The new era of global flows
Risk management has to be near the top of every corporation’s agenda. Some companies
have created multidisciplinary risk teams and implemented more flexible procurement
contracts and manufacturing systems. Regionalizing production near large end markets
can both reduce complexity in supply chains and minimize exposure to disruptions in
transit. Manufacturers need to periodically reevaluate the right balance between the use of
global suppliers and the resilience of operations. One of the most important precautions
is to diversify the supply chain, avoiding overreliance on any single supplier. In the
pharmaceuticals manufacturing industry, up to 30 percent of company revenue can be
traced to a single production site; up to three-quarters of revenue for some blockbuster
drugs is at risk due to single-sourcing somewhere along the supply chain.127
As the world grows more dependent on information systems, new types of risk—such as the
failure of power grids or damaging information leaks—enter the equation. The private sector
is also becoming more vulnerable to cyberattacks by disgruntled employees, criminals,
political activists, and even other nations. High-profile hacks and breaches have hit many of
the world’s largest companies. One study has estimated that cybercrime costs the global
economy some $400 billion in annual losses; these can include consumer data breaches,
financial crimes, market manipulation, and theft of intellectual property. This is line with an
estimate from Lloyd’s of London.128
A recent joint study by McKinsey and the World Economic Forum found that nearly
80 percent of technology executives said that they cannot keep up with attackers’
increasing sophistication and that protective measures (such as avoiding public cloud
services or limiting the degree to which employees share information) are already having a
negative business impact. Companies can prioritize information assets based on business
risks, test continuously to improve incident response, and work with frontline employees
to emphasize basic protective measures. If a breach does occur, a quick, decisive, and
forthright response from marketing, public affairs, and customer service functions can be
critical to restoring customer trust.129
•••
As digital technologies and globalization continue to reshape industries, the challenges for
companies are mounting—but so are the opportunities. This new world similarly poses more
complex questions for policy makers and regulators as economies around the world race to
carve out new roles in global value chains. Chapter 6 will explore some of the implications for
governments seeking to capture benefits of participating in global flows.
127 Manufacturing the future: The next era of global growth and innovation, McKinsey Global Institute,
November 2012.
128 Net losses: Estimating the global cost of cybercrime, Center for Strategic and International Studies and
McAfee, June 2014. See also Stephen Gandel, “Lloyd’s CEO: Cyber attacks cost companies $400 billion
every year,” Fortune, January 23, 2015.
129 David Chinn, James Kaplan, and Allen Weinberg, Risk and responsibility in a hyperconnected world:
Implications for enterprises, McKinsey & Company and the World Economic Forum, January 2014.
© Getty Images
Countries cannot afford to shut themselves off from global flows. Given their role in
substantially raising GDP and boosting productivity growth, there is too much value at
stake. But the goal is much broader than simply running a trade surplus. Our analysis finds
that inflows and outflows alike contribute to growth. Narrow export strategies often ignore
the real value of globalization: the flow of ideas, talent, and inputs that allow companies to
innovate in new ways and raise productivity in the economy.
Pursuing this value has never been a straightforward proposition, and today’s more digital
form of globalization makes the calculus even more complex. Trade negotiations will need
to include new dimensions to address issues surrounding cross-border data flows and
the exchange of information and communication technology (ICT) goods. National policy
makers increasingly need a global mindset to avoid erecting barriers that can lead to
competitive disadvantages.
The current wave of churn and transition creates openings for countries to carve out
profitable roles in the global economy. Those opportunities will favor locations that build the
infrastructure, institutions, and business environments that their companies and citizens
need to participate fully. Building these enablers can have the double benefit of boosting
domestic productivity—and without them, the economic impact of flows will be muted.
Realizing the full economic potential of global marketplaces, platforms, and communities
will require a deeper level of international cooperation. It will also depend on whether policy
makers can successful manage the volatility associated with an interconnected and rapidly
evolving digital economy. While it is impossible to anticipate all of the issues that will come
into play, this chapter offers a framework.
POLICY MAKERS NEED A CLEAR AGENDA TO CAPTURE THE FULL POTENTIAL
OF GLOBAL FLOWS
Even as governments try to create the right enabling environments for technology to fuel
growth, digitization is handing them a host of entirely new policy challenges. Many digital
firms have innovative business models that existing regulatory structures never considered.
The digital economy evolves so rapidly that regulators have to take a test-and-learn
approach to keep up with the pace of innovation.
Many of the challenges associated with digitizing
economic activity are now playing out on a
global scale.
6. THE NEW WORLD OF
POLICY CHALLENGES
98 McKinsey Global Institute 6. The new world of policy challenges
Think strategically about the role your country can play in global value chains
Officials building a national agenda to compete successfully in this new era could start by
taking a step back and thinking strategically about how their country can participate in
global flows based on assets they already have or could build. As global value chains shift,
countries can redefine the roles they play within them. The United States, for example, has
long been a major engine of consumer demand for imported goods, but it now plays an
equally important role as the world’s leading producer of digital platforms and content. As
new digital hubs form, the network of global flows may be redrawn in the years ahead.
Automation is narrowing the window of opportunity for developing countries to become the
world’s low-cost manufacturers, and 3D printing could transform how—and where—many
categories of goods are produced. But other types of opportunities exist.
Some countries could build on their geographic proximity to major consumer markets, as
Mexico and countries in Eastern Europe have done. Others may develop successful niches
as global transit hubs, although it is crucial to find ways to add value in addition to serving
as a waypoint. Singapore, for instance, has become central to flows of goods and services,
while Dubai has become a hub of transportation, trade, and finance. Other countries have
used a selective approach, targeting a particular flow or industry to cultivate: China long
ago transformed itself into the world’s manufacturing powerhouse, for instance, and is now
pursuing an active strategy to move up the value chain into more innovative industries. The
Philippines, Morocco, and South Africa have built on the advantages of language to become
global providers of business process outsourcing services. Countries may also build on
pools of talent within their borders, as Italy has done with high-end fashion design and
textiles and as India has done with IT engineering.
Recognizing the value of data flows, many locations are trying to create the “next Silicon
Valley.” But innovation is notoriously hard to orchestrate—and that is not the only way to
participate in the digital global economy. Our research finds that countries benefit from
receiving cross-border digital flows as well as producing them. In other words, countries do
not need to transform themselves into digital content or platform producers to benefit from
data flows.
Address policy and administrative barriers that hinder global flows
For national economies, opening up to all types of global inflows and outflows is crucial for
sustaining growth. Previous MGI research on global productivity trends has underscored
this effect in Brazil, where some sectors have been more exposed to global market
forces and some remain heavily protected. Embraer, the country’s flagship aerospace
company, for example, was privatized and now successfully goes head-to-head with global
competitors. Because Brazil lifted import tariffs on aircraft components, the company is
able to source from global suppliers. By contrast, import tariffs on vehicles have encouraged
foreign carmakers to establish production within Brazil to serve its large consumer market,
but the Brazilian automotive industry has not integrated effectively into global value chains.
Its productivity lags well behind peer economies such as Mexico, which has developed
world-class assembly plants and rapidly gained global market share. Within Mexico, too,
sectors that have privatized, embraced free trade, and welcomed foreign investment
and technology have pulled far ahead of traditional industries in terms of productivity
performance.130
Pursuing bilateral and multilateral trade partnerships is the cornerstone of a more open
approach. Another important step is removing import tariffs, quotas, and subsidies
for national industries, all of which can introduce distortions. Other types of legal and
130 See previous MGI studies: Connecting Brazil to the world: A path to inclusive growth, May 2014, and A tale of
two Mexicos: Growth and prosperity in a two-speed economy, March 2014.
99McKinsey Global Institute Digital globalization: The new era of global flows
administrative barriers can constrain the impact of global flows; these may include limitations
on foreign business ownership and investment, import licensing, regulatory requirements
that deviate from international norms, and limits on immigration. ASEAN, for instance, has
largely eliminated import tariffs among its ten member states, but its ongoing effort to build
a seamless trading bloc involves painstaking multilateral efforts to harmonize many types
of product standards, certification procedures, customs requirements, and cross-border
regulations covering traded services and the movement of labor.131 Removing these types
of barriers can enable large multinational companies, SMEs, entrepreneurs, and individuals
alike to take advantage of opportunities beyond their own borders.
Address the dislocations
Although the overall economic benefits of opening to global flows are clear, they can also
disrupt local industries by exposing them to international competition and new business
models. Some jobs and businesses may be lost even as new opportunities for growth
are created. (See Chapter 4 for more on this topic.) Governments have to consider these
trade-offs and open themselves to global flows at a pace their economies and societies
can absorb.
Labor markets and training systems in most countries have not proven flexible enough to
deal with rapid change on this scale. But providing support to affected workers and creating
a clearer path for them to find new roles deserves greater priority. Wage insurance is one
policy option.132 Another is ensuring that adults who are already in the workforce have
access to short, concentrated training programs for acquiring new skills. Germany may offer
a useful model for policy makers to consider. One of the world’s most connected countries,
ranking fourth in our global index, it has avoided large-scale unemployment by providing
income support and taking a proactive approach to labor market reforms.133
Invest in human capital
The Internet can promote inclusiveness, as long as education and training systems
provide language fluency, basic digital literacy, and other skills so that individuals can take
advantage of the opportunities. But educational systems in most countries are not keeping
up with the demands of a digital world; few mandate computer programming classes in
primary or secondary school.
Even as the new digital era is raising the importance of education, technology also offers
new possibilities for increasing its quality and reaching more people of all ages, whether
through online educational platforms with open access, learning programs that adapt to a
student’s performance, or classroom tools that allow teachers to tailor instruction.
A more digital economy places a new premium on skills, innovation, and adaptability. The
countries that are reaping disproportionate benefits are able to cultivate and attract pools of
highly educated and specialized technical talent. Investment in human capital development
will be a critical determinant of which nations come out on top.
Build the necessary physical infrastructure and close the digital divide
Even in a more digital world, physical infrastructure remains vital for tapping into global flows
of all types. Roads, ports, airports, and rail are the conduits of trade and mobility; investment
that modernizes and maintains these systems can propel economic growth. Many
countries—emerging and advanced economies alike—have paid insufficient attention to
those assets, creating economic inefficiencies and allowing foundational systems to erode.
131 Southeast Asia at the crossroads: Three paths to prosperity, McKinsey Global Institute, November 2014.
132 See, for example, Lori G. Kletzer, “Why the US needs wage insurance,” Harvard Business Review, January
25, 2016.
133 Marco Caliendo, Income support systems, labor market policies and labor supply: The German experience,
IZA discussion paper number 4665, December 2009.
100 McKinsey Global Institute 6. The new world of policy challenges
Today any list of infrastructure priorities also has to include universal, affordable Internet
access. The number of worldwide Internet users now exceeds 3.2 billion, but growth
is slowing. In 2011, the UN Broadband Commission set targets of reaching 60 percent
worldwide Internet penetration by 2015, with 40 percent household penetration in
developing nations. Those goals remain unmet, however: at the end of 2015, 57 percent of
the world’s population, or four billion people, remained offline.134 The enormous digital divide
in the world’s poorest countries and along gender lines remains stubbornly hard to bridge.
As the flow of ideas, information, and innovation becomes more central to participating in
the global economy, access to digital platforms and communication becomes an urgent
development issue.
The value of connecting the offline population to the Internet is significant. The World Bank
has calculated that a 10 percent increase in broadband access is associated with a 1.38
percentage point increase in GDP growth in developing countries and a 1.21 percentage
point increase in advanced economies.135
Our econometric analysis shows that countries with higher Internet penetration reap up
to 25 percent more benefit from cross-border data flows than those with limited Internet
penetration. Tremendous value can be created organically and unexpectedly when
companies and citizens consume data and information—and then combine it with their
own ingenuity.
Create a strong business and institutional environment for the digital economy
to thrive
Just as purchasing IT systems offers no guarantee that a company will be a digital leader,
building Internet infrastructure is not sufficient for countries to capture the full potential
benefits of digital globalization. Their business sectors and consumer populations need to
be able to engage and innovate online.
A recent World Bank report finds that digital technologies have not improved productivity
and reduced inequality to the degree once hoped in countries that lack strong fundamentals
such as education, good governance, and a supportive business environment.136 These
attributes have always been important for attracting foreign investment, and they are even
more critical today. The benefits of digital globalization are heavily concentrated among
countries with those ingredients in place, and lagging countries that fail to make broader
reforms in these areas risk falling even further behind.
The Internet can accelerate development and promote efficiency in emerging economies,
but it is not a shortcut around building good governance. Countries still need healthy
business environments that nurture startups, allow inefficient firms to exit, support research,
and provide a solid legal framework for intellectual property and property rights. Without
these elements, local companies will not be able to use global flows to raise their game, and
foreign investors and companies will be deterred. India, for example, is attempting to tackle
these issues and build a stronger foundation through initiatives such as Startup India and
Digital India.
A recent McKinsey survey found that business executives around the world believe that
government agencies can provide more transparency and information on opportunities
for domestic companies in foreign markets and opportunities for foreign companies
134 The state of broadband 2015: Broadband as a foundation for sustainable development, International
Telecommunication Union and UNESCO Broadband Commission for Digital Development, September 2015.
135 Christine Zhen-Wei Qiang and Carlo M. Rossotto with Kaoru Kimura, “Economic impacts of broadband,”
in Information and communications for development 2009: Extending reach and increasing impact, World
Bank, 2009.
136 World development report 2016: Digital dividends, World Bank, January 2016.
101McKinsey Global Institute Digital globalization: The new era of global flows
within the country. Having a one-stop shop to obtain such information, identify potential
business partners, and understand the regulatory and approval process is becoming
essential. Now that small firms have new avenues for participating in digitally facilitated
global trade, governments can raise awareness of these growth opportunities. Expanding
the information, mentoring, and financing available to small businesses can help them take
advantage of this new shift in cross-border commerce.
Governments can provide another enabler by opening their enormous data sets to
encourage private-sector innovation. Making data more open and widely available in
shareable formats can create substantial economic value, estimated at more than $3 trillion
by MGI. Governments from India, Mexico, New Zealand, Singapore, the United Kingdom,
and the United States are among those that have launched open data initiatives.137
Maintain an open Internet while protecting privacy
Taking an open approach to cross-border data flows can accelerate growth. Yet many
countries are considering limitations on what kind of data can be transmitted across borders
and where data must be stored. Some are moving toward regulations that would require
companies to use servers physically located within their borders to process and store data
generated there. Variations on this type of law exist in Indonesia, Nigeria, Russia, Vietnam,
and elsewhere. Other countries limit certain types of personal data transfers or have unique
consent requirements.138 In 2014, Brazil passed a sweeping “Internet bill of rights”; some
technologists questioned whether its privacy provisions, restrictions on data collection, and
requirements that Brazilians’ data must remain stored on servers within the country could
limit the use of large-scale analytics.139
Privacy has been a major issue in Europe. A 2014 ruling by the European Court of Justice
upheld the “right to be forgotten”—that is, requiring search engines to honor requests from
individual users to remove links to personal, inaccurate, or outdated information. As we went
to press, the future of the “safe harbor” agreement governing data transfers between the
European Union and the United States remained uncertain.
Requirements that data must be sequestered locally raise a host of issues for companies,
including cloud data storage and even personnel data for multinationals with operations and
employees in restricted nations. In particular, some of the business models surrounding the
Internet of Things are predicated on transmitting data to intermediaries, some of which may
be in other countries. Some companies are proceeding with building their own data centers
in locations around the world to cope with these types of requirements. But the compliance
burdens of sequestering data and operating across multiple countries with varying
regulations could limit the economic benefits of cross-border data flows.
Beyond those governments that are acting out of concern for the privacy of their citizens,
others regard the freewheeling nature of the Internet as a challenge to their authority and
have moved to censor content, block websites, or place users under surveillance.
137 Open data: Unlocking innovation and performance with liquid information, McKinsey Global Institute,
October 2013.
138 Data localization: A challenge to global commerce and the free flow of information, Albright-Stonebridge
Group, September 2015.
139 Maria Medrano, “Brazil’s Internet Bill of Rights,” Americas Quarterly, April 2015.
102 McKinsey Global Institute 6. The new world of policy challenges
While legitimate privacy concerns do need to be addressed through the development of
universal standards, the movement toward data localization raises the danger of balkanizing
the Internet. The economic benefits of cross-border data flows could be limited if the
Internet becomes governed by a web of varying country-specific regulations. A study by
the European Centre for International Political Economy examined the impact of recently
proposed or enacted data localization and security regulations in seven economies. It found
that these rules would lower GDP in all seven cases, with Vietnam (-1.7 percent), China
(-1.1 percent), and Indonesia (-0.5 percent) poised for the largest losses.140
Another persistent issue is the tendency of parts of the digital economy to develop natural
monopolies. The biggest platforms have enormous network effects and low marginal
costs precisely because of their enormous size. Some governments are wary of this
type of market power being held by a foreign company. But restricting the biggest global
platforms denies that country’s citizens and small businesses the opportunity to participate.
Our analysis described in Chapter 4 finds that countries benefit significantly from data
consumption, not merely from being home to Internet companies and platform providers.
Governments may want to consider whether their countries can produce their own robust
digital platforms to compete (as China has done). But walling a country off from global
platforms while failing to cultivate its own is a harmful combination.
Make cybersecurity a top priority
A world that runs on data flows is also vulnerable to cyberattacks. Private-sector companies
and government agencies alike have suffered serious data breaches at the hands of
hackers. One study has estimated that cybercrime costs the global economy some
$400 billion in annual losses; these can include consumer data breaches, financial crimes,
market manipulation, and theft of intellectual property.141 Aside from the substantial business
costs, hackers may pose public safety and even national security risks. Governments will
need to work closely with the business community to stay on top of new threats and share
information and new technology solutions. Regulators may need to mandate standards for
securing consumer data, and public agencies need to take additional steps to safeguard
their own assets.
Data privacy and security are thorny issues in almost every area of digital use. They are
central to realizing the full economic value of big data analytics and the Internet of Things,
which are predicated on collecting and sharing data. Governments have to make choices
about data collection, access, usage, and consent, especially for data generated in public
spaces. The dangers that hackers could create in physical settings have to be carefully
considered and guarded against; policy makers can help to address security issues by
creating frameworks for liability.142
Beyond the threat of breaches, governments need to be aware of what is lurking on the
so-called Darknet. The public Internet is vast, but it is dwarfed in size by the “Deep Web”
of non-indexed websites, as we explain in Chapter 1. Much of the Deep Web is simply
private information held by companies, enormous government databases, pay-to-use
databases, or private message boards. It is also used by activists, dissidents, journalists,
whistleblowers, and others who have legitimate needs for maintaining anonymity. But
hidden on the Darknet portion of the Deep Web is an entire online world of criminal activity,
including money laundering, drug trafficking, human trafficking, child pornography, hackers
for hire, and terrorist networks. Some criminal rings have been broken up and their members
140 See Matthias Bauer et al., The costs of data localization: Friendly fire on economic recovery, ECIPE occasional
paper number 3/2014, May 2014. The study also found that investment could drop by 4.2 percent in Brazil,
3.9 percent in the European Union, and 3.1 percent in Vietnam.
141 Net losses: Estimating the global cost of cybercrime, Center for Strategic and International Studies and
McAfee, June 2014.
142 The Internet of Things: Mapping the value beyond the hype, McKinsey Global Institute, June 2015.
103McKinsey Global Institute Digital globalization: The new era of global flows
prosecuted, including the founder of the notorious Silk Road black market. But reining
in the illicit global trade being conducted in cyberspace will require deeper international
coordination.143
•••
Policy makers have to strike the right balance between capturing the benefits of openness
while mitigating the risks—and in a digital world, both opportunities and challenges are
appearing with unprecedented speed. This new version of globalization is creating a faster-
moving and vastly more complex global economy, but it offers new ways to realize the
value of connectedness. This wave of change can accelerate growth for the countries that
approach it with optimism and vision.
143 Daniel Sui, James Caverlee, and Dakota Rudesill, The Deep Web and the Darknet: A look inside the Internet’s
massive black box, Woodrow Wilson International Center for Scholars, October 2015.
© Getty Images
This appendix outlines key points on the methodology in the following sections:
1. Data sources and definitions
2. Econometric model and statistical analyses
3. Academic literature on the relationship between global flows and GDP
4. Methodology for the MGI Connectedness Index
5. Methodology for global connectedness of regions within countries
6. Global survey of startups
1. DATA SOURCES AND DEFINITIONS
Global cross-border flows
We compiled a data set covering five categories of flows for 139 countries from 1980 to
2014 or the latest data available (Exhibit A1). The data set draws on multiple sources that we
describe in more detail later in this section.
For each type of flow, we assembled inflows and outflows for each country individually,
and wherever possible bilaterally. The coverage of bilateral data over time and across
countries varies by flow. Figures for total flows used in this report refer to the broadest
coverage available.
We used subcategories of overall goods, services, and financial flows for specific analyses.
For example, our analysis of knowledge-intensive flows includes only the knowledge-
intensive subcategories of each aggregate flow. The mapping to the goods categories has
been performed based on the United Nations’ six-digit harmonized coding system, HS
2002. We assigned service categories using the 11 chapters in the Comtrade database of
global commodities trade statistics maintained by the United Nations Statistics Division.
We categorized financial flows by the nature of their investment (i.e., FDI, equity, bonds, and
loans) and based them on data from several sources. People flows are not composed of
any single aggregate flow. Instead, we analyzed several components such as international
student flows, long-term migrants, refugees, and overnight visitors. For data flows, we
examined used cross-border bandwidth from TeleGeography.
TECHNICAL APPENDIX
106 McKinsey Global Institute Technical appendix
Exhibit A1
UNCTAD
UNCTAD
Capital-intensive goods 160
UNCTAD
UNCTAD
Migrants
Total financial stock
Portfolio investment
IMF
Travelers
194
EWN1
Bonds
IMF 112
World Bank 185
Reserves
214
260
273
Loans
125IMF
Equity
Remittance
113
World Bank
251
IMF
IMF
265IMFFDI flow
UNCTAD
Cultural and social services
Knowledge-intensive services
Labor-intensive services
146
UNCTAD
IMF
181
138
104
Services total
UNCTAD
Government services
Primary resources
160
UNCTAD
146
135
Capital-intensive services
UNCTAD
160
Total financial flows
UNCTAD
UNCTAD
Labor-intensive goods
218
160R&D-intensive goods
Goods total
Maximum
number of
countriesMajor flows
171TeleGeographyUsed bandwidth
201International students OECD
Source
221UNWTO
The compiled data set contains hundreds of countries and provides comprehensive coverage of the past decade
G
oods
S
ervices
Finance
P
eople
D
ata
SOURCE: McKinsey Global Institute analysis
1 External W ealth of Nations database.
88 90 95 1505 1000
107McKinsey Global Institute Digital globalization: The new era of global flows
Goods flows
This report analyzes the historical growth in the global flow of goods, its dispersion across
countries and regions, and its transformation due to digitization and knowledge intensity.
The primary source is the UNCTAD database, which provides non-bilateral data from 1980
onward. We analyzed more than 5,200 product codes between 2002 and 2014, dividing
them into four categories: capital-intensive manufacturing, labor-intensive manufacturing,
primary resources, and R&D-intensive manufacturing. For bilateral trade, we used data from
the World Bank’s World Integrated Trade Solution database, which is available from 2000
to 2014.
Each of the four categories of goods trade mentioned above has a number of subgroups.
� Capital-intensive manufacturing
— Food, beverages, and tobacco. Includes the production, processing, and
preservation of meat, fish, fruit, vegetables, oils, and fats; the manufacture of dairy
products, grain mill products, and starches and starch products; and the production
of other food products and beverages including spirits, wines, malt liquors, soft
drinks, and mineral waters. Also includes items related to the manufacture of
tobacco products.
— Paper products and publishing. Includes the manufacture of pulp, paper, and
paperboard; the manufacture of corrugated paper and paperboard and containers
made out of those materials; and the manufacture of specialty paper products,
including carbon paper, toilet paper, envelopes, and postcards.
— Manufacturing of petroleum, rubber, plastic, mineral, and metal products. Includes
the production of products related to the commodities of metals, mining, and oil
and gas.
� Labor-intensive manufacturing
— Textiles. Includes the spinning, weaving, and finishing of textiles; the manufacture of
carpets, rugs, rope, twine, and netting; and the manufacture of knitted and crocheted
fabrics and articles.
— Leather, fur products, and apparel. Includes the manufacture of fur and non-fur
apparel, and the dressing and dyeing of fur. Also includes the production of footwear,
the tanning and dressing of leather, and the manufacture of leather products.
— Wood products and furniture. Includes the sawmilling and planing of wood; the
manufacture of wood, cork, and straw; and the manufacture of wood products,
including furniture, musical instruments, sporting goods, and toys.
� Primary resources
— Agriculture, hunting, fishing, and related activities. Includes the growing of crops, the
farming of animals, the hunting and trapping of animals, fishing, and the operation of
fish hatcheries and farms.
— Forestry and logging. Includes goods produced through forestry and logging and
related service activities.
108 McKinsey Global Institute Technical appendix
— Metals. Includes the manufacture of basic iron and steel, basic precious and non-
ferrous metals, and the casting of metals. Also includes other fabricated metal
products, such as tanks, reservoirs, and construction materials.
— Mining. Includes the mining and agglomeration of hard coal and lignite, uranium,
and thorium ores; the mining of ferrous and non-ferrous metal ores; the mining
and quarrying of stone, sand, and clay; and the extraction of crude petroleum and
natural gas.
— Oil and gas. Includes the manufacture of coke oven products and refined petroleum
products as well as the processing of nuclear fuel.
� R&D-intensive manufacturing
— Chemicals and chemical products. Includes the manufacture of basic chemicals,
fertilizers and nitrogen compounds, plastics and rubbers, pesticides and other
agro-chemical products, paints, pharmaceuticals, soaps and detergents, artificial or
synthetic fibers, yarn, and filaments.
— Electrical, telecommunication, and computing machinery. Includes the
manufacture of office, accounting, and computing machinery; electrical machinery
such as motors, generators, transformers, wires and cables, and electrical
equipment; television and radio transmitters and receivers; and sound and video
recording equipment.
— Motor vehicles and other transport equipment. Includes the manufacture of motor
vehicles, trailers and semitrailers, and parts and accessories for motor vehicles and
their engines; and other transport equipment such as ships, railway and tramway
locomotives, aircraft and spacecraft, motorcycles, and bicycles.
— Medical, precision and optical instruments. Includes the manufacture of medical
appliances and instruments, optical instruments, photographic equipment, and
watches and clocks.
— Other machinery and equipment. Includes the manufacture of general-purpose
machinery such as engines, turbines, pumps, compressors, ovens, and lifting and
handling equipment; special-purpose machinery such as agricultural machinery,
weapons and ammunition; machinery for the production of mining and metal
products, food, beverage, and tobacco products; and domestic appliances.
We separately subdivided goods into three categories: finished goods, intermediate goods,
and raw materials. The broad economic categories (BEC) classification system specified by
the United States has three designations: consumer goods, capital goods, and intermediate
goods. We consider both consumer and capital goods as finished goods. We further
subdivide intermediate goods into raw materials and intermediate goods.
� Finished goods. Includes finished capital and consumer goods such as industrial
machines, ships and aircrafts, refined petroleum, sugar, and apparel.
� Intermediate goods. Includes parts used as inputs for making finished products such
as pharmaceutical inputs, vehicle parts, and steel products.
� Raw materials. Includes commodities and processed commodities. Examples of
commodities are coal, corn, cotton, and crude petroleum. Processed commodities
include materials such as animal fats and oils, coffee, and processed metal.
109McKinsey Global Institute Digital globalization: The new era of global flows
Services flows
All non-bilateral flows of services draw on UNCTAD and UN Comtrade data. We subdivided
services flows into five categories: knowledge-intensive, labor-intensive, capital-intensive,
cultural and social, and government. However, the coverage of services data is most
effective at the aggregate level rather than broken into these five categories.
Each of these also has a number of subgroups:144
� Knowledge-intensive services
— Insurance services. The provision of insurance to non-residents by resident insurance
enterprises and vice versa; services provided for freight insurance on goods exported
and imported; services provided for other types of direct insurance including life and
non-life; and services provided for re-insurance.
— Financial services. Financial intermediation services and auxiliary services conducted
between residents and non-residents other than those related to insurance
enterprises and pension funds.
— Computer and information services. Resident and non-resident transactions related
to hardware consultancy, software implementation, information services (i.e., data
processing, data base, news agency), and maintenance and repair of computers and
related equipment.
— Royalties and license fees. Includes receipts (exports) and payments (imports) of
residents and non-residents for the authorized use of intangible non-produced,
non-financial assets and proprietary rights such as trademarks, copyrights, patents,
processes, techniques, designs, manufacturing rights, and franchises; and the
use, through licensing agreements, of produced originals or prototypes such as
manuscripts and films.
— Other business services. Covers merchanting and other trade-related services as
well as operational leasing services; and miscellaneous business, professional, and
technical services.145
� Labor-intensive services
— Travel. Goods and services, including those related to health and education, acquired
by travelers during visits of less than one year. The goods and services are purchased
by, or on behalf of, the traveler or provided, without a quid pro quo, for the traveler to
use or give away.
— Construction services. Construction and installation project work performed on a
temporary basis in the compiling economy or in extraterritorial enclaves by resident
and non-resident enterprises and associated personnel, excluding foreign affiliates.
144 The definitions used for the subgroups are closely based on IMF Balance of Payments Manual definitions.
145 Purchase of a good by a resident of the compiling economy from a non-resident and the subsequent resale
of the good to another non-resident. Value created between purchase and resale is recoded as value of
merchanting service.
110 McKinsey Global Institute Technical appendix
� Capital-intensive services
— Communications services. Communications transactions between residents and
non-residents (i.e., postal, courier, and telecommunications services).
— Transportation. Transportation services performed by residents of one economy
for those of another and vice versa, and that involve the carriage of passengers,
the movement of goods (freight), rentals (charters) of carriers with crew, and related
supporting and auxiliary services.
� Cultural and social services. Audiovisual and related services and other cultural
services provided by residents to non-residents and vice versa.
� Government services. All services (e.g., spending by embassies and consulates)
associated with government sectors or international and regional organizations and not
classified under other items.
Financial flows
Aggregate financial flows comprise the following asset classes:
� FDI. Investments that establish at least a 10 percent stake in a foreign entity. Any
subsequent lending between the direct investor and the financial recipient is also
captured in this category.
� Equity. Any equity or share purchased by an investor in another country that gives the
investor no more than a 10 percent stake.
� Bonds. Any tradable debt security that is purchased by a foreign investor, including
public and corporate (both financial and non-financial) bonds, mortgage-backed
securities, other asset-backed securities, and collateralized debt obligations.
� Loans. Any other assets not classified in the above three categories, primarily loans,
currency, and deposits, and a small share of trade credit.
In addition to these four classes, data on outward investments capture a fifth category
of reserve assets: assets acquired or held by monetary authorities in a foreign currency.
Reserve assets are distinguished from the other four classes to avoid double-counting.
We take all these data from balance of payments statistics from the International
Monetary Fund.
Further, we also look at data on flows of remittances from the World Bank. We do not
include these flows in our core analysis of major financial flows because they either overlap
with other financial flows, such as loans, or are the reverse of goods and services flows.
We also analyze a bilateral data set for FDI, which gives us an indication of the origin and
destination of this flow.
In addition to financial flows data, we take countries’ foreign assets and foreign liabilities
(financial stock data) into consideration. This is sourced from Philip R. Lane’s “External
Wealth of Nations” data set.
111McKinsey Global Institute Digital globalization: The new era of global flows
People flows
Unlike for flows of goods, services, and finance, we do not have additive data sets on people
flows. Instead, we look at overlapping categories and describe people flows from different
angles. All data we collected for people flows are bilateral, indicating both departure and
arrival countries. We use three direct measures:
� Migrants. While all other measures capture or approximate flows of people, our
migration data are stock data, indicating foreign-born residents by country. Data are
from the World Bank and are available for 1980 to 2013.
� Travelers. Arrivals of non-resident visitors or tourists at national borders, drawn from the
UN World Tourism Organization, available from 1995 to 2013.
� Students. International flows of mobile students at the tertiary level (ISCED 5 and 6) from
the UNESCO Institute for Statistics, available from 1999 to 2013,
Data flows
For data flows, we look at cross-border used bandwidth data from TeleGeography,
which provides data by region, country, and key routes from 2005 to 2014. Used capacity
is the sum of all capacity deployed for Internet backbones, private networks, and
switched voice networks. It does not include capacity that is used for restoration and
redundancy purposes.
Knowledge-intensive flows
We define knowledge-intensive flows as cross-border goods, services, finance, people, and
data and communications flows that are rich with ideas, knowledge, and information. The
aim is to approximate a value of global flows that are linked to today’s knowledge economy.
We define the following subcategories of global flows as knowledge-intensive:
� R&D-intensive manufacturing (goods flows). Of all goods flows, those classified
as R&D-intensive manufacturing are considered to have the highest portion of
knowledge involved in production or development. When these goods cross borders,
the knowledge embodied in these products or their development is at least partially
transferred across those borders.
� Knowledge-intensive services (services flows). Knowledge-intensive services are
those requiring the highest skill level of the parties providing the service (e.g., financial
services) or that directly represent the realized value of knowledge or content creation
(e.g., the payment of royalties and license fees).
� FDI (financial flows). Of all financial flows, FDI is most clearly linked to the transfer of
knowledge across borders as companies conducting greenfield FDI transfer knowledge
to the new location. We also consider brownfield FDI as a transfer of knowledge
because companies that acquire other companies either use their own knowledge
and management techniques to improve the business of the acquisition or use the
knowledge embedded in their investment to improve their own business.
� Cross-border telecom revenue (data and communication flows). In our services
database, telecommunications is a financial-intensive service. However, it is an important
proxy for data and communication flows and represents a minimum value of those flows.
Therefore, we have included this revenue in an attempt to capture a significant portion of
the value of data and communication flows. We have elected to include only business,
as opposed to residential, revenue because we believe these flows have the most
substantial knowledge component.
112 McKinsey Global Institute Technical appendix
� Business traveler spending (people flows). In our services database, business travel
is a labor-intensive service. However, it can be used as a proxy for knowledge-intensive
people flows and, at the very least, represents a minimum value of such flows. When
business travelers move across borders, they carry knowledge with them; in fact, these
travelers have likely traveled to a different country to either impart that knowledge or
acquire knowledge that they will carry back to their home country.
Country classifications
For some analyses, we classify each of the 139 countries in our sample as either a
developing or a developed economy. For developed economies, we use the term
“advanced economies” interchangeably. We refer to developing economies as “emerging
markets” or “emerging economies.”
We also assigned each country to one of ten regions, six of which we define as emerging
and four as developed. This classification of countries and their assignment to regions
follows the approach used in previous McKinsey Global Institute reports.146 We define
Singapore, Hong Kong, Taiwan, and Macao as developed economies despite the fact that
they are located in emerging-market regions (Exhibit A2).
146 See, for example, Financial globalization: Retreat or reset? McKinsey Global Institute, March 2013.
Exhibit A2
Classification of countries into regions and development level
SOURCE: McKinsey Global Institute Financial Assets database; McKinsey Global Institute analysis
1 Combined to form “Other Asia” in analysis regarding interregional vs. intraregional trade.
2 Classified as developed despite being located in a region classified as emerging.
Europe, Middle East, and Africa
Western Europe
▪ Austria
▪ Belgium
▪ Denmark
▪ Finland
▪ France
▪ Germany
▪ Greece
▪ Iceland
▪ Ireland
▪ Italy
▪ Malta
▪ Netherlands
▪ Norway
▪ Portugal
▪ Spain
▪ Sweden
▪ Switzerland
▪ United
Kingdom
Eastern Europe and Central Asia
▪ Bulgaria
▪ Czech Republic
▪ Hungary
▪ Kazakhstan
▪ Lithuania
▪ Latvia
▪ Poland
▪ Russia
▪ Slovakia
▪ Turkey
Plus 19 other
countries
Africa and Middle East
▪ Algeria
▪ Angola
▪ Botswana
▪ Cameroon
▪ Egypt
▪ Ghana
▪ Iran
▪ Israel
▪ Jordan
▪ Kenya
▪ Kuwait
▪ Lebanon
▪ Morocco
▪ Nigeria
▪ Saudi Arabia
▪ South Africa
▪ Tunisia
▪ United Arab
Emirates
Plus 26 other
countries
D
eveloped regions
E
m
erging regions
Americas
North America
▪ Canada
▪ United States
Latin America
▪ Argentina
▪ Bolivia
▪ Brazil
▪ Chile
▪ Colombia
▪ Costa Rica
▪ Dominican
Republic
▪ Ecuador
▪ Guatemala
▪ Jamaica
▪ Mexico
▪ Panama
▪ Uruguay
▪ Venezuela
Plus 10 other
countries
Asia
Northeast Asia
▪ Japan ▪ South Korea
Australasia
▪ Australia ▪ New Zealand
China region
▪ China
South Asia1
▪ India
▪ Pakistan
▪ Bangladesh
▪ Sri Lanka
▪ Maldives
Southeast Asia1
▪ Philippines
▪ Malaysia
▪ Cambodia
▪ Indonesia
▪ Thailand
▪ Vietnam
▪ Singapore2
Plus 5 other
countries
113McKinsey Global Institute Digital globalization: The new era of global flows
2. ECONOMETRIC MODEL AND STATISTICAL ANALYSES
In Chapter 4 of this report, we discuss why openness to global flows matters for economic
performance. According to the classical Solow growth model, the main factors determining
GDP growth are the physical capital stock of a nation and its human capital.147 Newer
growth-theory models also include a role for technological progress or innovation. To test
for the additional effect of cross-border flows on GDP growth, which may raise total factor
productivity or increase the utilization of capital and labor, we employ a two-step error
regression model.
Our econometric model
Our model is based on a classic Cobb-Douglas production function of the following form:
where Y represents GDP, K is the fixed capital stock in an economy, L is the labor stock
in an economy, and A is total factor productivity, which includes technological progress
and innovation.
We used a two-step error correction model (ECM). The ECM allows us to incorporate two
aspects into our analysis: first is the inter-temporal relation in GDP growth (i.e., this year’s
growth affects next year’s growth). We can thus estimate the change in GDP growth as a
function of changes in independent variables such as flows. This inter-temporal relationship
also allows us to differentiate between the short-term and long-term effects of the
independent variables of GDP growth.
We use data from 97 countries, from 1995 to 2013 (the latest year for which those data are
available for all flows and all countries). To account for country-specific fixed effects that
may also contribute to GDP growth, we format the data as a panel data set, pooling cross-
sectional and time-series data, and run the regression on it using a fixed-effect model. This
controls for time-invariant effects, such as the legal system or colonial history of the country.
Given that we are using long time series in our analysis, we also pay attention to testing
for cointegration in the data series. This is especially troublesome in building a dynamic
econometric model, as cointegrated data series share a common trend and thus may lead
to a false finding of correlation among the variables. We first used sophisticated methods
to test for and correct cointegration and then used a two-step error-correction model for
the estimation.
To begin, we test the long-term relationship between the data series. We run a unit root
test on the residual from this step and select a model whose residual is cointegrated. In the
second step, we estimate the short-term relationship with the same set of covariates from
the long-run relationship and its residual. Lastly, we control our potential endogeneity in the
data series using instrumental variables. As instruments, we use the difference between
the three-year lag of flows and four-year lags.148 We tested different time lags to see which
instruments are stronger and determined that the three- to four-year lag is most suitable.
147 For an overview of different growth models, see, for example, Robert J. Barro and Xavier Sala-i-Martin,
Economic growth, MIT Press, 2003.
148 Given that our specification includes the one-year lag of GDP, our lagged instrumental variables are at least
t–2 or t–3. Given our panel data, we could have used the GMM method, which increases the number of
lagged instruments. The gain in efficiency was small, so we report the traditional instrumental techniques, with
Hausman test passed for their validity.
Exhibit A15
Formulas
Y = A × Kα × Lβ (1)
Cobb-Douglas production function
logGDPi,t = α + SkβklogCVi,k,t + SiβilogFlowi,j,t–1 + εi,t
Econometric step 1 of 2-step ECM
∆logGDPi,t = α + Skβk∆logCVi,k,t + Sjβj∆logFlowi,j,t +
γlogGDP per capitai,t–2 + δεi,t–1 + ui,t
Econometric step 2 of 2-step ECM
flow’s GDP share in leveli,j,t =
(flowi,j,t)
βj
real GDPi,t
Flow’s share in GDP
flow’s GDP share in growthi,j,t =
(log(flowi,j,t) – log(flowi,j,t–1)) × βj
(log(real GDPi,t) – log(real GDPi,t–1))
Flow’s share in growth
log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 +
θjcentralityi,tlogDataFlowi,j,t + εi,t
Centrality Step 1 of 2-step ECM
log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 +
θjcentralityi,tlogDataFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t
Centrality Step 2 of 2-step ECM
114 McKinsey Global Institute Technical appendix
Using the ECM in this way allows us to make a more credible claim regarding the direction of
causality. We thus examine the impact of the growth of flows in one period on the growth of
GDP for the subsequent period, controlling for unobserved country-specific effect, the noise
from the covariates that are not integrated, and endogeneity among the covariates and GDP.
The transformation of the Cobb-Douglas function to logarithmic scale allows us to
estimate the elasticities of each variable (i.e., by how much does GDP growth change if the
explanatory variable changes by 1 percent). The estimated model takes the following form:
Step 1 of two-step ECM
where i represents country i, t the current year , t–1 the past year, S is the symbol of sum, j
is the index of flow, and k is the index of control variables.
The definition of each variable is as follows:
� logGDPit is the marginal change in real GDP for country i between year t and t–1 in
natural log
� α is the constant term
� logCVi,k,t is the k-th control variable and is the marginal change in the lagged level
of respective control variables for country i in time t. Their coefficient estimates are
interpreted as the long-term elasticity of real GDP with respect to a change in the
control variable.
� logFlowi,j,t–1 is the j-th flow and is the change in the lagged level of the respective flow.
It determines the long-term elasticity.
� εi,t is the residual or error that we will save for the second step, which has to be
cointegrated (i.e., it does not suffer from having a unit root).
Step 2 of two-step ECM:
� ∆logGDPi,t is the change in growth of real GDP for country i between year t and t–1 in
natural log
� ∆logCVi,k,t–1 comprises the set of k control variables and is the change in growth of
the respective control variable from year t–1 to t in natural log. Its coefficient estimates
are interpreted as the short-term elasticity of real GDP growth with respect to the
control variables.
� ∆logFlowi,j,t–1 comprises the set of j flows and is the change in growth of the flows
in natural log. Its coefficient estimates determine the short-term elasticity of real GDP
growth with respect to flows.
Exhibit A15
Formulas
Y = A × Kα × Lβ (1)
Cobb-Douglas production function
logGDPi,t = α + SkβklogCVi,k,t + SiβilogFlowi,j,t–1 + εi,t
Econometric step 1 of 2-step ECM
∆logGDPi,t = α + Skβk∆logCVi,k,t + Sjβj∆logFlowi,j,t +
γlogGDP per capitai,t–2 + δεi,t–1 + ui,t
Econometric step 2 of 2-step ECM
flow’s GDP share in leveli,j,t =
(flowi,j,t)
βj
real GDPi,t
Flow’s share in GDP
flow’s GDP share in growthi,j,t =
(log(flowi,j,t) – log(flowi,j,t–1)) × βj
(log(real GDPi,t) – log(real GDPi,t–1))
Flow’s share in growth
log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 +
θjcentralityi,tlogDataFlowi,j,t + εi,t
Centrality Step 1 of 2-step ECM
log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 +
θjcentralityi,tlogDataFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t
Centrality Step 2 of 2-step ECM
Exhibit A15
Formulas
Y = A × Kα × Lβ (1)
Cobb-Douglas production function
logGDPi,t = α + SkβklogCVi,k,t + SiβilogFlowi,j,t–1 + εi,t
Econometric step 1 of 2-step ECM
∆logGDPi,t = α + Skβk∆logCVi,k,t + Sjβj∆logFlowi,j,t +
γlogGDP per capitai,t–2 + δεi,t–1 + ui,t
Econometric step 2 of 2-step ECM
flow’s GDP share in leveli,j,t =
(flowi,j,t)
βj
real GDPi,t
Flow’s share in GDP
flow’s GDP share in growthi,j,t =
(log(flowi,j,t) – log(flowi,j,t–1)) × βj
(log(real GDPi,t) – log(real GDPi,t–1))
Flow’s share in growth
log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 +
θjcentralityi,tlogDataFlowi,j,t + εi,t
Centrality Step 1 of 2-step ECM
log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 +
θjcentralityi,tlogDataFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t
Centrality Step 2 of 2-step ECM
115McKinsey Global Institute Digital globalization: The new era of global flows
� logGDP per capitai,t–2 is the two-year lagged value of real GDP per capita in natural
log, capturing the catch-up effect of the developing countries on the real GDP growth.
Since this effect does not apply to countries that already have high GDP per capita (i.e.,
advanced countries), its coefficient γ is expected to be negative.
� εi,t–1 is the lagged residual from the step one. Its coefficient δ needs to be negative
and statistically significant for the second step to be statistically valid. Conceptually,
this coefficient captures the rate of the short-term model converging into the long-
term model.
� ui,t is the error term from step two of ECM.
Each of the control variables enters the regression as a stand-alone term. The control
variables CV in the estimation are as follows:
� Human capital, measured by average years of schooling in the adult population
� Real fixed capital stock, derived from the accumulated real fixed investment in the
country after depreciation, to capture capital inputs in the economy
� Employment, to capture labor inputs in the economy.
We test the relationship between flows and GDP by measuring flows in different ways.
First, we use the sum of inflows and outflows for the country, normalized by nominal GDP
(for goods trade and FDI) or population (for migration flows and data flows). Second, we
use each country’s score in the MGI Connectedness Index. The normalized flows used in
the estimation are goods trade, migration, FDI flows, and cross-border data usage. The
connectedness scores used are for goods trade, labor-intensive services trade, travelers,
FDI flow, and used cross-border bandwidth. These variables are selected by assessing their
correlation with each other, Granger causality with GDP, and using a backward/forward
stepwise model selection.
We obtain the signs we expected on each of the flow variables (Exhibit A3), with the
exception of migration flows. The coefficient on migration flows is negative for long-term
elasticity, while we would expect it to be positive. We believe this result is due to loss of
skilled labor in developing countries or their difficulty in absorbing large migrant or refugee
flows. Exhibits A4 and A5 show the short-term and long-term elasticities from the two
measures of global flows described above.
116 McKinsey Global Institute Technical appendix
Exhibit A3
Short-/long-term impact
Name of variable
Granger causality
with real GDP
Expected sign of
coefficient Estimated sign of coefficient
FDI Two-way Positive/positive Positive/positive
Goods trade flow Two-way Positive/positive Positive/positive
Immigration Two-way Positive/positive Insignificant/negative1
Data flows Two-way Positive/positive Positive/positive
Services trade flow Two-way Positive/positive Extended due to correlation with FDI
Fixed capital stock n/a Positive/positive Positive/positive
Employment n/a Positive/positive Positive/positive
Average years of education n/a Positive/positive Insignificant/negative
SOURCE: McKinsey Global Institute analysis
The coefficients from our econometric model have the expected sign
Flow variables
Dependent variable (Log)
Real GDP
Independent variables (Log)
1 Migration flows are negligible or slightly negative at the global level, possibly due to the loss of skilled labor in developing countries or the difficulties of
absorbing a large influx of refugees or migrants. However, migration flows have a positive impact on productivity in advanced economies.
Exhibit A4
GDP impact of global flows, using normalized flow values
SOURCE: McKinsey Global Institute analysis
Elasticities
Long term Short term
Coefficients P-values Coefficients P-values
Flow variables Flow variables
Goods trade 0.05 0.0129 Goods trade 0.0817 0.0002
FDI 0.04 0 FDI 0.0039 0.0761
Migration -0.05 0.0036 Immigration Insignificant n/a
Data 0.02 0 Data usage 0.025 0.0154
Macroeconomic variables Macroeconomic variables
Fixed capital stock 0.48 0 Fixed capital stock 0.76 0
Employment 0.39 0 Employment 0.49 0
Average years of education Insignificant n/a Average years of education Not available n/a
Dependent variable: Real GDP
97 countries, 1995–2013
117McKinsey Global Institute Digital globalization: The new era of global flows
Calculating the impact of flows on GDP output and GDP growth
After we estimate the two-step ECM, we use the short-term and long-term elasticities
(shown in Exhibit A4) to calculate the contribution of flows to both the level of GDP and the
growth rate of GDP for a country i for flow j at time t as follows:
Where βj is the flow j’s GDP long-term elasticity from step one of two-step ECM
Where βj is the flow j’s GDP short-term elasticity from step two of two-step ECM. These
expressions are derived from the two-step ECM because it is a combination of the change
in GDP in level (the long-term model) and the change in GDP growth (the short-term model).
We calculate the above for both normalized flows and for a model specification using
connectedness index scores. The results indicate that flows accounted for 10.1 percent of
global GDP over the past ten years (Exhibit A6).
Exhibit A5
GDP impact of global flows, using connectedness scores for each flow
SOURCE: McKinsey Global Institute analysis
Elasticities
Long term Short term
Coefficients P-values Coefficients P-values
Connectedness scores
Internet traffic 0.0180 0.1202 Insignificant n/a
Goods trade 0.0706 0.0006 0.4264 0.0022
Service trade1 0.0249 0.0498 Insignificant n/a
Travelers 0.0399 0.0407 Insignificant n/a
FDI flow 0.0204 0.0444 Insignificant n/a
Macroeconomic variables
Fixed capital stock 0.4718 0 0.70 0
Employment 0.4685 0 0.48 0
Average years of
education
Insignificant n/a n/a n/a
Dependent variable: Real GDP
97 countries, 1995–2013
1 W e use data on labor-intensive services trade only, not knowledge-intensive services or capital-intensive services,
because the latter are highly correlated with FDI.
Exhibit A15
Formulas
Y = A × Kα × Lβ (1)
Cobb-Douglas production function
logGDPi,t = α + SkβklogCVi,k,t + SiβilogFlowi,j,t–1 + εi,t
Econometric step 1 of 2-step ECM
∆logGDPi,t = α + Skβk∆logCVi,k,t + Sjβj∆logFlowi,j,t +
γlogGDP per capitai,t–2 + δεi,t–1 + ui,t
Econometric step 2 of 2-step ECM
flow’s GDP share in leveli,j,t =
(flowi,j,t)
βj
real GDPi,t
Flow’s share in GDP
flow’s GDP share in growthi,j,t =
(log(flowi,j,t) – log(flowi,j,t–1)) × βj
(log(real GDPi,t) – log(real GDPi,t–1))
Flow’s share in growth
log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 +
θjcentralityi,tlogDataFlowi,j,t + εi,t
Centrality Step 1 of 2-step ECM
log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 +
θjcentralityi,tlogDataFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t
Centrality Step 2 of 2-step ECM
Exhibit A15
Formulas
Y = A × Kα × Lβ (1)
Cobb-Douglas production function
logGDPi,t = α + SkβklogCVi,k,t + SiβilogFlowi,j,t–1 + εi,t
Econometric step 1 of 2-step ECM
∆logGDPi,t = α + Skβk∆logCVi,k,t + Sjβj∆logFlowi,j,t +
γlogGDP per capitai,t–2 + δεi,t–1 + ui,t
Econometric step 2 of 2-step ECM
flow’s GDP share in leveli,j,t =
(flowi,j,t)
βj
real GDPi,t
Flow’s share in GDP
flow’s GDP share in growthi,j,t =
(log(flowi,j,t) – log(flowi,j,t–1)) × βj
(log(real GDPi,t) – log(real GDPi,t–1))
Flow’s share in growth
log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 +
θjcentralityi,tlogDataFlowi,j,t + εi,t
Centrality Step 1 of 2-step ECM
log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 +
θjcentralityi,tlogDataFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t
Centrality Step 2 of 2-step ECM
118 McKinsey Global Institute Technical appendix
Impact on productivity
Flows may increase GDP by raising either the usage of capital and labor, or by raising
productivity. To calculate this impact, we run a separate model specification that includes
GDP/employment, or labor productivity.149 The difference between the coefficient on
this variable and the coefficient of flows on GDP is the residual, which is the impact
on productivity.
The results are shown in Exhibit A7. We find that all flows affect GDP mainly through
productivity. Goods flows, FDI, and data flows all positively increase productivity. Migration
flows have a negative impact on productivity for emerging economies. As noted above, this
may reflect the impact of “brain drain” (the loss of skilled labor) or the difficulties developing
countries encounter when absorbing large migrant or refugee flows. For advanced
economies, we find a positive impact of migration on productivity. Exhibit A7 also shows that
data flows have a positive impact on an economy’s utilization of capital and labor. Thus far,
fears about digital flows reducing employment appear unfounded.
149 We run a similar regression using capital productivity and find similar results.
Exhibit A6
All flows combined contributed 10.1 percent of GDP from 2003 to 2013, with goods and data having largest impact
SOURCE: McKinsey Global Institute analysis
Flows contribution to GDP FLOWS MODEL—ALL COUNTRIES
Shares of output
%
Midpoint 5th percentile 95th percentile
2008–13 2003–13 1998–2013 2008–13 2003–13 1998–2013 2008–13 2003–13 1998–2013
All flows 9.36 10.05 11.03 8.97 9.62 10.55 10.06 10.82 11.90
Goods trade 3.20 3.51 3.97 3.28 3.56 4.03 3.21 3.46 3.91
FDI flow 1.68 1.64 1.52 1.74 1.70 1.58 1.62 1.58 1.46
Immigration 1.74 1.95 2.24 1.16 1.30 1.50 2.61 2.92 3.36
Data flow 2.70 2.95 3.29 2.79 3.06 3.43 2.62 2.85 3.17
Exhibit A7
All flows contribute to raising productivity, but only data flows contribute to increasing
labor and capital inputs
SOURCE: McKinsey Global Institute analysis
Flows model
Flow variables
Long-term elasticity
for real GDP
Impact on
productivity
Impact on
increased inputs
Goods trade 0.05 0.05 0
FDI 0.04 0.05 -0.0023
Migration -0.05 -0.04 -0.0101
Data usage 0.02 0.02 0.0029
119McKinsey Global Institute Digital globalization: The new era of global flows
Network centrality
In a separate analysis, we investigated the importance of a country’s position within the
network of trade flows and within the network of data flows for its GDP growth. More
connections with a greater number of neighbors should reflect a more diverse portfolio
of imports and data. A broader portfolio benefits a country by enriching the consumption
basket, by enabling companies to source ideas and inputs from all over the world, and by
taking advantage of the global competitive landscape to diversify and reduce dependence
on any single partner. Broader network coverage for exporters reflects their competitiveness
and ability to sell in many markets. More routes and a more central position in the network
therefore indicate the presence of highly competitive firms that can participate in global trade
and markets and thereby have a positive impact on their home country’s GDP.
To estimate the impact of centrality, we use two measures: the eigenvector centrality of data
flow and the number of routes. The eigenvector centrality of data flow is the position in a
network from calculating the eigenvector based on a country’s neighbors’ positions. It is the
data-trading partners’ network rather than direct links that affects a country’s eigenvector
centrality. In contrast, the number of routes directly measures the connections of a given
country with other countries.
We use the same model specification in terms of control variables as we do for the
combination of flows. Centrality is introduced as an interaction term with log of data flow for
the long run and with the difference in log of data flow:
Step 1 of two-step ECM
Step 2 of two-step ECM
Correlation between the data flow and its interaction with either measure of centrality is
found to be negligible. When the centrality has a dampening effect on GDP through data
flow, we expect θj to be less βj than for the data flow. We observe that countries on the
periphery of the data flows actually benefit more than those at the center of the data flows.
3. ACADEMIC LITERATURE ON THE RELATIONSHIP BETWEEN GLOBAL FLOWS
AND GDP
A large body of academic literature has examined the impact of different types of global
flows on GDP, productivity, and innovation. Generally, these studies are based on one type
of flow (i.e., trade, financial flows, or immigration) and often on one country. Some have
created their own indexes of globalization including trade and other flows, as well as other
metrics such as trade restrictions, enhanced technology transmission, and improvements in
macroeconomic policy.
From this literature, we find general consensus that trade, FDI, and immigration support
higher levels and growth rates of GDP and also of productivity growth. However, the size
estimates of this impact vary across studies. These estimates are shown in Exhibit A8, along
with our own, which are in a similar range.
Exhibit A15
Formulas
Y = A × Kα × Lβ (1)
Cobb-Douglas production function
logGDPi,t = α + SkβklogCVi,k,t + SiβilogFlowi,j,t–1 + εi,t
Econometric step 1 of 2-step ECM
∆logGDPi,t = α + Skβk∆logCVi,k,t + Sjβj∆logFlowi,j,t +
γlogGDP per capitai,t–2 + δεi,t–1 + ui,t
Econometric step 2 of 2-step ECM
flow’s GDP share in leveli,j,t =
(flowi,j,t)
βj
real GDPi,t
Flow’s share in GDP
flow’s GDP share in growthi,j,t =
(log(flowi,j,t) – log(flowi,j,t–1)) × βj
(log(real GDPi,t) – log(real GDPi,t–1))
Flow’s share in growth
log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 +
θjcentralityi,tlogDataFlowi,j,t + εi,t
Centrality Step 1 of 2-step ECM
log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 +
θjcentralityi,tlogDataFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t
Centrality Step 2 of 2-step ECM
Exhibit A15
Formulas
Y = A × Kα × Lβ (1)
Cobb-Douglas production function
logGDPi,t = α + SkβklogCVi,k,t + SiβilogFlowi,j,t–1 + εi,t
Econometric step 1 of 2-step ECM
∆logGDPi,t = α + Skβk∆logCVi,k,t + Sjβj∆logFlowi,j,t +
γlogGDP per capitai,t–2 + δεi,t–1 + ui,t
Econometric step 2 of 2-step ECM
flow’s GDP share in leveli,j,t =
(flowi,j,t)
βj
real GDPi,t
Flow’s share in GDP
flow’s GDP share in growthi,j,t =
(log(flowi,j,t) – log(flowi,j,t–1)) × βj
(log(real GDPi,t) – log(real GDPi,t–1))
Flow’s share in growth
log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 +
θjcentralityi,tlogDataFlowi,j,t + εi,t
Centrality Step 1 of 2-step ECM
log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 +
θjcentralityi,tlogDataFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t
Centrality Step 2 of 2-step ECM
120 McKinsey Global Institute Technical appendix
Exhibit A8
Academic literature on flows and GDP growth
SOURCE: McKinsey Global Institute analysis
10% increase in flow results in the following % increase in GDP
Flow
Literature
review
MGI
model Reference literature
Goods trade 0.2–1.5 0.5 Frankel and Romer, “Does trade cause growth?” American Economic Review,
1999
Bianjing, Zuoshi, and Jingkui, Endogenous international trade and economic
growth: Empirical study based on 120 Chinese cities, 2011
Wacziarg, “Measuring the dynamics gains from trade,” World Bank Economic
Review, 2001
US Executive Office of the President, The economic benefits of US trade, 2015
De Loecker and Goldberg, “Firm performance in a global market,” Annual Review
of Economics, 2013
FDI flow 0.04–0.5 0.4 Aizenmann, Jinjarak, and Park, “Capital flows and economic growth in the era of
financial integration and crisis,” Open Economies Review, 2013
Bordo, Meissner, and Stuckler, “Foreign currency debt, financial crises and
economic growth: A long-run view,” Journal of International Money and Finance,
2010
Reinhart and Reinhart, “Capital flow bonanzas: An encompassing view of the past
and present,” in NBER International Seminar on Macroeconomics 2008, 2009
Kose, Prasad, and Terrones, Does openness to international financial flows raise
productivity growth? NBER, 2008
Migration1 1.0–5.0 -0.51 Peri, The effect of immigration on productivity: Evidence from US states, 2009
Boubtane, Dumont, and Rault, Immigration and economic growth in the OECD
countries, 1986–2006, IZA discussion paper, 2014
US Executive Office of the President, The economic benefits of fixing our broken
immigration system, 2013
Internet
penetration
0.2–1.4 0.2 Qiang and Rossotto with Kimura, “Economic impacts of broadband,” in Information
and communications for development 2009: Extending reach and increasing
impact, World Bank, 2009
Freund and Weinhold, “The effect of the Internet on international trade,” Journal of
International Economics, 2004
Meijers, “Does the Internet generate economic growth, international trade, or
both?” International Economics and Economic Policy, 2014
Falk and Hagsten, E-commerce trends and impacts across Europe, UNCTAD
discussion paper, 2015
USITC, Digital trade in the US and global economies, part 2, 2014
Onyejiwu, Inter-country variations in digital technology in Africa, WIDER discussion
paper, 2002
1 W e find that migration flows are negligible or slightly negative at the global level, possibly due to the loss of skilled labor in developing countries or the
difficulties of absorbing a large influx of refugees or migrants. However, we find that migration flows have a positive impact on productivity in advanced
economies, consistent with other academic literature.
121McKinsey Global Institute Digital globalization: The new era of global flows
Flows of goods and services
The academic literature provides a wide range of estimates on the impact of trade flows
on GDP growth. Frankel and Romer use bilateral date for 63 countries for 1985 and
estimate that a one percentage point increase in the trade-to-GDP ratio causes almost
a 1.5 percentage point increase in per capita income growth.150 At the other end of the
range, Bianjing, Zuoshi, and Jingkui, based on international trade of 120 cities, suggest that
a 1 percent increase in international trade produces only a 0.19 to 0.22 percentage point
increase in income.151
Wacziarg analyzed data for 57 countries from 1970 to 1989, measuring the impact of
trade policy openness on economic growth, where openness is a function of investment,
enhanced technology transmission, and improvements in macroeconomic policy. The
analysis suggests that an 8.5 percentage point increase in the trade policy measure,
corresponding roughly to one standard deviation, is associated with a 0.6 percentage point
increase in annual GDP growth. Investment is the most important channel, accounting for
almost two-thirds of this effect.152
The US Council of Economic Advisors reviewed the effects of trade and found that the
reduction of trade barriers since World War II has raised US GDP by 7.3 percent, or
approximately $1.3 trillion in 2014. It also found that a 10 percent increase in an industry’s
exports is associated with a 0.2 percent increase in that industry’s labor productivity. Based
on the average industry’s increase in exports, international trade may have been responsible
for about one-quarter of total US productivity growth over the 1990s and 2000s. The paper
also emphasizes that trade spurs both labor productivity and innovation, as measured by
total factor productivity. A 10 percentage point decrease in tariffs corresponds to about a
0.4 percentage point increase in labor productivity growth and about a half percentage point
increase in TFP growth over the two decades.153
Other studies have also examined the impact of trade on productivity growth. De Loecker
and Goldberg review the literature and conclude that there is strong evidence that
globalization raises firm-level productivity.154
The econometric analysis in this report is consistent with the low end of the range. We find
that a 1 percent increase in goods trade to GDP results in a 0.05 percent growth in GDP.
This may reflect the fact that in our analysis, we control for other types of flows, such as data
flows and financial flows, that may accompany trade flows.
150 Jeffrey A. Frankel and David Romer, “Does trade cause growth?” American Economic Review, volume 89,
issue 3, June 1999.
151 Ma Bianjing, Xie Zuoshi, and Li Jingkui, Endogenous international trade and economic growth: An empirical
study based on 120 Chinese cities, available at SSRN, August 2011.
152 Romain Wacziarg, “Measuring the dynamics gains from trade,” World Bank Economic Review, volume 15,
number 3, October 2001.
153 The economic benefits of US trade, Executive Office of the President, May 2015.
154 Jan De Loecker and Pinelopi Koujianou Goldberg, “Firm performance in a global market,” The Annual Review
of Economics, October 2013.
122 McKinsey Global Institute Technical appendix
FDI and other financial flows
The literature on the impact of financial flows on GDP growth is mixed. In general, academic
studies find that FDI has a positive impact on GDP growth, while broader financial flows and
capital account openness have a mixed effect. For instance, Mun et al. find that a 1 percent
increase in FDI flows relative to GDP result in a 0.05 percent increase in GDP growth.155
This is based on analyzing data from Malaysia spanning 1970 to 2005. Similarly, Aizenman,
Jinjarak, and Park examine a sample of 100 countries using data from 1990 to 2010 and find
that a one standard deviation increase in FDI inflow has increases the growth of GDP per
capita by 0.90 to 0.94 percent.156
The impact of broader financial flows (including cross-border lending and portfolio
purchases of equities and bonds) on GDP growth is less clear. Bordo et al. find that flows
of foreign currency debt increase the risk of financial crises, based on their study of foreign
currency debt, financial crises, and short- and long-term output effects from 1880 to 1913
and from 1973 to 2003 for 45 countries.157 Similarly, Reinhart and Reinhart examine a large
group of countries over nearly 50 years and find that episodes of heavy capital inflows are
associated with a higher incidence of banking, currency, and inflation crises, particularly in
developing countries.158 While Kose et al. find that a 10 percentage point increase in ratio of
FDI and equity liabilities to GDP would be associated with a 0.4 percentage point increase
in TFP, a similar increase in ratio of debt liabilities to GDP would lead to a decrease in TFP
growth of 0.2 percentage points.159
Migration flows
The economic literature typically finds a positive impact of immigration on GDP growth for
the destination country receiving migrants. Peri analyzes all 50 US states plus Washington,
DC, in census years between 1960 and 2006; he finds that when immigrants produce a
1 percent increase in employment in a US state, income per worker rises by 0.5 percent.160
Boubtane and Dumont observe that a one percentage point increase in immigration results
in 0.1 percentage point increase in economic growth for 22 OECD countries.161 The US
Council of Economic Advisors finds that enhancing immigration reforms that increase
immigration would increase real GDP relative to current projections by 3.3 percent by 2023
and 5.4 percent by 2033.162
Our research finds a negative, albeit small, impact of immigration on GDP growth. This may
be the result of developing countries losing skilled talent or poorer countries encountering
difficulties in absorbing a large influx of refugees or unskilled immigrants. We find, however,
that migration flows have a positive impact on total factor productivity for advanced
economies, which is consistent with other literature.163
155 Har Wei Mun, Teo Kai Lin, and Yee Kar Man, “FDI and economic growth relationship: An empirical study on
Malaysia,” International Business Research, volume 1, number 2, April 2008.
156 Joshua Aizenman, Yothin Jinjarak, and Donghyun Park, “Capital flows and economic growth in the era of
financial integration and crisis, 1990–2010,” Open Economies Review, volume 24, issue 3, July 2013.
157 Michael D. Bordo, Christopher M. Meissner, and David Stuckler, “Foreign currency debt, financial crises and
economic growth: A long-run view,” Journal of International Money and Finance, volume 29, 2010.
158 Carmen Reinhart and Vincent Reinhart, “Capital flow bonanzas: An encompassing view of the past and
present,” in NBER International Seminar on Macroeconomics 2008, Jeffrey Frankel and Christopher
Pissarides, eds., University of Chicago Press, 2009.
159 M. Ayhan Kose, Eswar S. Prasad, and Marco E. Terrones, Does openness to international financial flows raise
productivity growth? NBER working paper number 14558, December 2008.
160 Giovanni Peri, The effect of immigration on productivity: Evidence from US states, NBER working paper
number 15507, November 2009.
161 Ekrame Boubtane, Jean-Christophe Dumont, and Christophe Rault, Immigration and economic growth in the
OECD countries, 1986–2006, IZA discussion paper number 8681, November 2014.
162 The economic benefits of fixing our broken immigration system, Executive Office of the President, July 2013.
163 For instance, see Ekrame Boubtane, Jean-Christophe Dumont, and Christophe Rault, Immigration and
economic growth in the OECD countries, 1986–2006, IZA discussion paper number 8681, November 2014.
123McKinsey Global Institute Digital globalization: The new era of global flows
Data flows
Most of the literature available on the impact of data flows on economic growth relates to
increase in Internet penetration. For example, Choi and Yi use data from 207 countries from
1991 to 2000 and find that that a one percentage point increase in the Internet user ratio
leads to a 0.057 percentage point increase in GDP.164 Meijers finds that a 10 percentage
point increase of Internet penetration leads to a 0.17 percentage point increase of economic
growth and an increase in international trade.165 A World Bank study for 120 countries from
1980 to 2006 reports that a 10 percent increase in broadband penetration resulted in a
1.38 percent point increase in GDP growth in developing countries and a 1.21 percent point
increase in growth in developed countries.166 Increases in Internet usage can also promote
more trade. One study concludes that a 10 percent increase in Internet access leads to a
0.2 percent increase in exports.167
A recent study of EU firms also found that engaging in e-commerce increases labor
productivity and that e-commerce accounted for 17 percent of EU labor productivity
growth between 2003 and 2010.168 A 2014 study by the US International Trade Commission
(ITC) calculated the productivity gains from the Internet by surveying US businesses and
converting the results into an economic model. The ITC found that the productivity gains
from the Internet have increased US real GDP by 3.4 to 3.5 percent.169
4. METHODOLOGY FOR THE MGI CONNECTEDNESS INDEX
The MGI Connectedness Index ranks countries on the extent of their engagement with the
global economy through inflows and outflows of goods, services, finance, people, and data.
To obtain a more granular picture, the index ranks countries on their connectedness to each
individual flow as well as compiling an aggregate score.
We consider each country’s inflows and outflows of goods, services, finance, people, and
data. Financial flows include FDI, equity, debt, and other flows (mainly cross-border lending).
For people flows, we consider countries’ connectedness in terms of the stock of foreign-
born migrants resident in a given country and that country’s citizens living abroad.
After assigning each country a score for each type of flow, we weight those scores equally
to obtain that country’s overall connectedness score. Our methodology differs from other
globalization indexes in both the weighting and the data used (Exhibit A9). Our 2014 index
assesses 139 countries that provide data for 2014 for each of these flows. For people flows,
however, data are available only through 2013.
164 Changkyu Choi and Myung Hoon Yi, “The effects of the Internet on economic growth: Evidence from cross-
country panel data,” Economic Letters, volume 105, issue 1, October 2009.
165 Huub Meijers, “Does the Internet generate economic growth, international trade, or both?” International
Economics and Economic Policy, volume 11, issue 1, February 2014.
166 Christine Zhen-Wei Qiang and Carlo M. Rossotto with Kaoru Kimura, “Economic impacts of broadband,”
in Information and communications for development 2009: Extending reach and increasing impact, World
Bank, 2009.
167 Caroline L. Freund and Diana Weinhold, “The effect of the Internet on international trade,” Journal of
International Economics, volume 62, 2004.
168 Martin Falk and Eva Hagsten, E-commerce trends and impacts across Europe, UNCTAD discussion paper
number 220, March 2015.
169 United States International Trade Commission, Digital trade in the US and global economies, part 2,
August 2014.
124 McKinsey Global Institute Technical appendix
Exhibit A9
The MGI Connectedness Index measures five types of
inflows and outflows, unlike other studies
SOURCE: McKinsey Global Institute analysis
MGI DHL/Ghemawat E&Y/EIU KDF
Overview 139 countries in
2014
1980–2014
5 dimensions:
goods, services,
finance, people, and
data
140 countries in
2014
2012–14
5 dimensions: goods,
services, finance,
people, and data and
communications
60 countries in 2012
2009–12
6 dimensions: goods,
services, finance, people,
data and communications,
and culture
187 countries in 2015
1970–2015
7 dimensions: goods,
services, finance, people,
data and communications,
culture, and political
globalization
D
im
ensions
Goods 20 Total goods
flows (100%)
35 Total goods
flows (75%)
22 Total goods and
services flows (40%)
Trade openness1
barriers (10%)
Tariff and non-tariff1
barriers (10%)
Ease of trading1
(10%)
Current account
restrictions (10%)
Share of main trading
partners in total trade
(20%)
36 Total goods and
services flows (11%)
Hidden trade barriers
(12%)
Mean tariff (14%)
Taxes on international
trade (13%)
Services 20 Total goods
flows (100%)
Total service
trade (25%)
Financial 20 FDI flows (40%)
Portfolio
investment flows
(10%)
Bank and other
flows (10%)
Foreign
investment
assets and
liabilities (40%)
FDI stocks (25%)
FDI flow (25%)
Portfolio equity
stock (25%)
Portfolio equity
flows (25%)
FDI stock (50%)
Portfolio capital flows
(8%)
Policy toward FDI1
(8%)
Domestic favoritism1
(8%)
Expropriation risk
(8%)
State control (8%)
FDI stocks (13%)
Portfolio investment
stocks (12%)
Income payments to
foreign nationals/GDP
flows (13%)
Restrictions on capital
account (11%)
People 20 Immigrant stock
(80%)
Travelers flow
(20%)
15 Immigrant stock
(33%)
Travelers flow
(33%)
International
student flow
(33%)
19 Net immigration rate
(40%)
Travelers flow (40%)
Hiring of foreign
nationals (20%)
36 Immigrant stock
(21%)
Travelers flow (26%)
International calls flow
(25%)
International letters
flow (25%)
Data and
communi-
cation
20 Cross-border
used Internet
bandwidth
(100%)
15 International
bandwidth stock
(40%)
International
calls flow (40%)
Traded
publications
(20%)
21 ICT goods flows
(30%)
Creative goods flows
(30%)
Broadband subs
stock (20%)
Internet subs stock
(20%)
Internet users stock
(36%)
Television stock
(38%)
Trade in newspapers
flows (26%)
Cultural/
political
17 Cultural integration
Travelers flow (33%)
International fixed
telephone call (33%)
Openness to foreign
culture influence1
(33%)
Cultural globalization
McDonald’s
restaurants (44%)
IKEA stores (44%)
Trade in books (11%)
26 Political globalization
Embassies, member-
ships, UN Security
Council missions,
international treaties
1 Elasticity for TFP with respect to flows is calculated by subtracting the elasticity for labor (or capital) productivity from that for GDP.
Dimension variables (variable weight)
Percentage weight in overall indexXX
125McKinsey Global Institute Digital globalization: The new era of global flows
Normalization and ranking
We assess each country’s connectedness within each type of flow by looking at two
dimensions: flow intensity and flow share.
Flow intensity measures the size of a given flow as a share of a given country’s GDP or
population. As an illustration, Germany’s flow intensity in the goods trade is 71 percent—
that is, the value of all goods imported to and exported from Germany in 2014 was
equivalent to 71 percent of Germany’s GDP. It is important to note that flows in goods and
services are measured in nominal values, while national accounts capture GDP as value
added. This is why the ratio of trade to GDP can easily exceed 100 percent. The intensity of
goods, services, and financial flows is calculated relative to GDP. The intensity of people and
data flows is measured relative to the size of a country’s population.
Relying on flow intensity alone would artificially boost small countries over those with large,
diversified domestic economies in the rankings. To correct for this, we include a measure
of each country’s share of the global total of each flow, which we call flow share. Germany’s
flow share for goods is 7 percent—that is, Germany accounted for 7 percent of all global
inflows and outflows of goods in 2014.
To combine these two measures of connectedness into an index and calculate composite
trade intensity, we use the following methodology:
To smooth the distribution between countries, we use the resulting figure to assign countries
a normalized score relative to other countries on a scale of 1 to 100.
This normalized score can be used to rank all the countries in each of the five types of flows.
Exhibit A16
Moar ormulas
Compound flow intensity score
Compound flow intensity score =
(n–1) × (inflow + outflow)2
GDP × global flow
Normalized compound flow intensity score *** SAME for regions
Normalized compound flow intensity score =
99 × (Compound flow intensity score – min)
(max – min)
+ 1
Regional compound score
Compound flow intensity score =
(nregions –1) × (inflowregion + outflowregion)2
GDPregion × country flow
Regions vs. countries
Compound flow intensity score =
(ncountries –1) × (inflowregion + outflowregion)2
GDPregion × world flow
Exhibit A16
Moar ormulas
Compound flow intensity score
Compound flow intensity score =
(n–1) × (inflow + outflow)2
GDP × global flow
Normalized compound flow intensity score *** SAME for regions
Normalized compound flow intensity score =
99 × (Compound flow intensity score – min)
(max – min)
+ 1
Regional compound score
Compound flow intensity score =
(nregions –1) × (inflowregion + outflowregion)2
GDPregion × country flow
Regions vs. countries
Compound flow intensity score =
(ncountries –1) × (inflowregion + outflowregion)2
GDPregion × world flow
126 McKinsey Global Institute Technical appendix
Subcomponents of the scores on people and financial flows
The scores for people flows consider two components: migrants and travelers. The migrant
and traveler compound intensity scores are calculated separately following the methodology
described above. The overall score for people flows is a weighted average of the migrant
score and the traveler score. We assign the migrant score a weighting of 80 percent, while
travelers account for 20 percent of the overall score.
Financial flows have four components: FDI, portfolio investment flows (equity and bonds),
other financial flows (loans and deposits), and the stock of foreign financial assets and
liabilities. The overall financial flows score is a weighted average of these four flows. FDI
and foreign assets and liabilities stock are given weights of 40 percent each, while portfolio
investment flows and other flows have weights of 10 percent.
A country’s overall connectedness score is calculated by weighing the score on each of
the five types of flows equally and calculating a simple average. This score then determines
each country’s position in the overall rankings.
The MGI Connectedness Index: Full rankings
The abbreviated version of the MGI Connectedness Index that appears in Chapter 3
contains the top 25 countries plus a selection of other major economies. Exhibits A10
through A12, however, list the full rankings for all 139 countries.
As noted in Chapter 3, more countries are participating in global flows today, but we also
observe that flows remain concentrated among a small set of highly connected countries.
We can see this from the connectedness score of each country on each flow. Exhibit A13
shows the top 15 countries in each flow. The score for the top country is normalized to 100,
so the scores for the other countries can be interpreted in relation to the most connected
country. We see that data flows are the most heavily concentrated, with the steepest drop-
offs from the leaders. Service flows are the second most concentrated. Goods flows are
the least concentrated, with country scores declining less rapidly after the leading country.
Financial flows and people flows are similar, and people flows in particular have higher
scores for a larger set of countries.
127McKinsey Global Institute Digital globalization: The new era of global flows
Exhibit A10
MGI Connectedness Index (1/3)
Country connectedness index and overall flows data, 2014
Rank of participation by flow as measured by flow intensity and share of world total
1–10 11–25 26–50 >50Connectivity index rank 100+ <7070–99Flow intensity
Rank Country Score
Connectedness Index rank Flow value1
$ billion
Flow intensity2
% of GDPGoods Services Finance People Data
1 Singapore 64.2 1 2 2 12 6 1,392 452
2 Netherlands 54.3 3 3 6 21 1 1,834 211
3 United States 52.7 7 7 3 1 7 6,832 39
4 Germany 51.9 2 4 8 3 2 3,798 99
5 Ireland 45.9 32 1 1 28 9 559 227
6 United Kingdom 40.8 13 5 5 6 3 2,336 79
7 China 34.2 4 16 4 82 38 6,480 63
8 France 30.1 11 8 9 7 4 2,262 80
9 Belgium 28.0 5 6 33 33 8 1,313 246
10 Saudi Arabia 22.6 20 28 27 2 53 790 106
11 United Arab Emirates 22.2 6 23 17 4 46 789 196
12 Switzerland 18.0 12 11 10 17 13 848 115
13 Canada 17.3 16 22 11 11 18 1,403 79
14 Russia 16.1 21 25 18 5 25 1,059 57
15 Spain 14.4 25 13 19 14 16 1,105 79
16 Korea 14.0 8 12 28 50 44 1,510 107
17 Italy 13.4 17 18 24 16 19 1,587 74
18 Sweden 13.0 29 14 22 31 5 572 100
19 Austria 11.7 26 17 31 20 12 470 108
20 Malaysia 11.6 9 19 25 26 43 610 187
21 Mexico 10.7 14 63 34 18 41 1,022 80
22 Thailand 10.7 10 15 36 44 64 605 162
23 Kuwait 10.6 37 46 13 13 75 306 153
24 Japan 10.5 15 20 12 81 20 2,498 54
25 Kazakhstan 10.0 48 73 41 8 57 176 83
26 Ukraine 9.8 38 39 87 10 34 133 101
27 Australia 9.7 30 34 21 15 33 825 57
28 Denmark 8.9 35 9 32 41 11 369 108
29 Jordan 8.8 73 50 75 9 83 50 138
30 India 8.5 24 10 35 58 70 1,316 64
31 Qatar 7.8 33 35 29 19 59 300 141
32 Czech Republic 7.5 18 33 57 59 15 397 193
33 Malta 7.4 97 26 7 90 50 31 308
34 Poland 7.0 23 31 47 34 22 585 107
35 Hungary 6.8 22 30 26 62 17 287 209
36 Norway 6.0 36 24 20 46 24 458 92
37 Vietnam 5.7 19 54 45 103 61 350 188
38 Lebanon 5.6 82 21 46 22 103 69 151
39 Finland 5.5 46 27 23 70 10 390 144
40 Portugal 5.5 47 36 30 23 31 255 111
41 Turkey 5.1 28 40 53 38 29 521 65
42 Slovak Republic 5.0 27 60 68 67 14 205 205
43 Israel 4.9 51 32 49 24 56 248 82
44 Brazil 4.5 41 38 14 125 30 869 37
45 Chile 4.1 45 58 16 102 27 239 92
46 Belarus 4.1 40 66 101 29 47 92 121
47 Greece 4.1 60 29 54 35 42 160 67
SOURCE: McKinsey Global Institute analysis
1 Flows value represents total goods, services, and financial inflows and outflows.
2 Flow intensity represents the total value of goods, services, and financial flows as a share of the country’s GDP.
128 McKinsey Global Institute Technical appendix
Exhibit A11
MGI Connectedness Index (2/3)
Country connectedness index and overall flows data, 2014
Rank of participation by flow as measured by flow intensity and share of world total
1–10 11–25 26–50 >50Connectivity index rank 100+ <7070–99Flow intensity
Rank Country Score
Connectedness Index rank Flow value1
$ billion
Flow intensity2
% of GDPGoods Services Finance People Data
48 New Zealand 3.9 67 48 61 25 51 130 63
49 Romania 3.9 39 51 83 36 28 194 97
50 Croatia 3.7 76 45 104 27 37 57 100
51 Indonesia 3.4 31 49 38 106 76 504 57
52 Mozambique 3.3 95 70 15 117 110 40 246
53 South Africa 3.3 34 57 52 64 80 277 79
54 Philippines 3.2 54 41 44 52 67 230 81
55 Bulgaria 3.1 49 53 67 48 23 92 165
56 Albania 3.1 114 72 79 30 73 16 117
57 Oman 3.1 44 65 55 54 66 121 148
58 Bosnia and Herzegovina 3.0 86 123 113 32 62 21 117
59 Lithuania 2.8 43 55 112 68 35 87 181
60 Cote d'Ivoire 2.7 80 104 136 37 114 28 82
61 Slovenia 2.7 42 56 64 75 36 105 212
62 Pakistan 2.7 78 91 84 39 88 116 47
63 Azerbaijan 2.6 75 62 37 57 69 92 122
64 Morocco 2.6 58 43 74 56 65 104 97
65 Estonia 2.6 56 47 60 72 21 54 209
66 Bangladesh 2.6 71 99 62 43 113 109 62
67 Serbia 2.5 74 61 103 45 45 52 118
68 Bahrain 2.4 65 118 56 49 58 28 82
69 Moldova 2.4 105 102 102 40 52 12 154
70 Cyprus 2.3 122 37 43 76 55 18 79
71 Jamaica 2.3 115 69 100 42 72 17 113
72 Argentina 2.3 64 68 63 60 32 198 37
73 Egypt 2.2 68 42 69 73 71 158 55
74 Colombia 2.2 61 89 40 83 54 197 52
75 Latvia 2.2 66 67 76 66 26 51 158
76 Armenia 2.2 121 97 99 47 81 12 113
77 Libya 2.2 53 78 59 84 108 65 159
78 Panama 2.1 69 44 42 129 39 74 161
79 Dominican Republic 2.1 94 77 93 53 82 41 64
80 El Salvador 2.1 98 110 94 51 89 26 104
81 Algeria 2.0 52 82 91 91 85 152 71
82 Angola 1.9 50 64 86 134 111 100 76
83 Nigeria 1.9 55 76 48 128 98 268 47
84 Burkina Faso 1.9 123 117 139 55 134 8 67
85 Venezuela 1.9 57 85 71 86 60 172 34
86 Peru 1.8 62 88 51 104 49 122 60
87 Macedonia, FYR 1.8 93 101 124 63 48 18 156
88 Georgia 1.8 106 79 77 65 63 20 123
89 Sri Lanka 1.8 91 81 81 69 93 56 75
90 Guyana 1.7 118 128 116 61 125 4 133
91 Brunei 1.7 89 138 58 74 95 25 146
92 Cambodia 1.7 70 71 72 94 112 35 210
93 Ecuador 1.7 72 112 111 80 79 67 66
SOURCE: McKinsey Global Institute analysis
1 Flows value represents total goods, services, and financial inflows and outflows.
2 Flow intensity represents the total value of goods, services, and financial flows as a share of the country’s GDP.
129McKinsey Global Institute Digital globalization: The new era of global flows
Exhibit A12
MGI Connectedness Index (3/3)
Country connectedness index and overall flows data, 2014
Rank of participation by flow as measured by flow intensity and share of world total
1–10 11–25 26–50 >50Connectivity index rank 100+ <7070–99Flow intensity
Rank Country Score
Connectedness Index rank Flow value1
$ billion
Flow intensity2
% of GDPGoods Services Finance People Data
94 Tunisia 1.7 63 74 108 107 78 53 110
95 Mongolia 1.6 99 98 39 136 86 23 194
96 Kyrgyz Republic 1.6 102 87 88 78 118 13 173
97 Paraguay 1.6 83 133 121 77 94 26 84
98 Iran, Islamic Rep. 1.6 59 86 106 127 96 185 45
99 Costa Rica 1.5 81 75 65 101 74 43 87
100 Lao PDR 1.5 125 131 122 71 120 8 68
101 Ghana 1.5 77 80 126 109 100 32 83
102 Suriname 1.4 116 134 133 79 119 5 83
103 Liberia 1.4 134 121 50 100 139 7 333
104 Bolivia 1.4 84 108 85 93 84 33 98
105 Honduras 1.4 79 100 80 112 99 29 148
106 Yemen 1.4 90 116 114 88 116 26 70
107 Iceland 1.4 107 59 98 123 40 20 120
108 Guatemala 1.4 88 105 92 99 92 44 75
109 Montenegro 1.4 131 111 89 85 77 5 105
110 Uruguay 1.4 103 95 70 95 68 35 60
111 Maldives 1.4 132 52 109 121 126 7 225
112 Nicaragua 1.4 92 113 90 97 90 17 144
113 Gabon 1.4 101 126 130 87 105 13 73
114 Tajikistan 1.4 119 114 78 89 127 11 120
115 Barbados 1.3 136 93 73 108 87 4 97
116 Fm Sudan 1.3 124 129 105 92 97 20 27
117 Mali 1.3 120 106 110 98 131 11 95
118 Kenya 1.3 100 84 127 119 91 35 58
119 Fiji 1.3 113 90 119 111 121 7 163
120 Congo, Dem. Rep. 1.3 110 120 66 118 138 15 46
121 Cape Verde 1.2 137 107 125 105 123 2 114
122 Lesotho 1.2 111 139 129 110 133 3 167
123 Samoa 1.2 138 135 132 96 129 1 121
124 Zambia 1.2 87 124 82 138 115 29 105
125 Senegal 1.2 109 103 117 120 106 15 93
126 Botswana 1.2 85 137 120 133 107 19 121
127 Namibia 1.2 96 122 96 132 101 16 120
128 Uganda 1.2 126 83 95 131 117 18 67
129 Guinea 1.2 130 132 107 113 136 5 82
130 Tanzania 1.2 108 94 97 135 104 25 51
131 Cameroon 1.2 112 96 137 126 124 17 54
132 Benin 1.2 127 125 128 115 132 7 78
133 Rwanda 1.2 135 127 135 114 130 5 64
134 Swaziland 1.1 117 115 138 124 128 5 143
135 Papua New Guinea 1.1 104 92 118 139 137 10 55
136 Belize 1.1 133 119 123 122 102 3 153
137 Grenada 1.1 139 136 134 116 109 1 96
138 Sierra Leone 1.1 128 130 115 130 135 5 96
139 Seychelles 1.1 129 109 131 137 122 3 179
SOURCE: McKinsey Global Institute analysis
1 Flows value represents total goods, services, and financial inflows and outflows.
2 Flow intensity represents the total value of goods, services, and financial flows as a share of the country’s GDP.
130 McKinsey Global Institute Technical appendix
Exhibit A13
25.8
South Korea
45.9
29.6
China
28.6
46.9
Belgium
Switzerland
United Kingdom
87.8
France
24.5
24.5
28.2
Malaysia 30.0
Japan
Mexico
Thailand
44.1
United States
United Arab Emirates
78.4
90.9
Germany 92.6
100.0
Netherlands
Singapore
People Data
SOURCE: McKinsey Global Institute analysis
Flows are concentrated among a few leading countries
The 15 most connected countries within each type of flow, 2014
Connectedness score
Goods Services Finance
100.0Ireland
38.5
Singapore 83.7
12.7
Spain
12.9
20.0
Thailand
Sweden
14.5
South Korea 14.7
India
Switzerland 15.5
16.8
Denmark
France
United States
United Kingdom
29.9
33.5
27.5
34.7
Germany
Netherlands
Belgium
36.2
24.2
66.5
United Kingdom
Switzerland
Germany
Netherlands
68.9
16.2
Malta
57.3
France
14.2
China
Canada
United States
Ireland
18.8
18.3
25.0
25.2
Singapore 83.2
100.0
10.8
Brazil 10.8
Kuwait 11.4
Japan
Mozambique
Jordan
31.4
Singapore
United Kingdom
33.7
Kazakhstan
40.8
Australia 26.6
Spain 30.6
Kuwait
Canada
Ukraine
35.1
36.7
37.0
39.9
France
100.0
51.2
81.7
Russia 44.1
Saudi Arabia
United States
45.7
Germany
43.3
United Arab Emirates
3.2
3.4
3.5
5.8
6.7
6.8
Czech Republic
Slovak Republic
Switzerland
Belgium
Ireland
Denmark
Austria
10.3
14.9
Finland
United States 18.8
34.8
Germany
20.6
France
Sweden
100.0
United Kingdom
24.5
42.7
Netherlands
55.5
Singapore
131McKinsey Global Institute Digital globalization: The new era of global flows
5. METHODOLOGY FOR GLOBAL CONNECTEDNESS OF REGIONS
WITHIN COUNTRIES
In addition to measuring the global connectedness of countries, we have looked at variation
in the global connectedness of states and provinces within five large countries: the United
States, the United Kingdom, Germany, Brazil, and China. We were able to obtain data for
goods flows (imports and exports) for specific regions within these five countries as well
as immigration data. For the United States, we were also able to obtain FDI inflows and
outflows and service exports by state.
We did the regional analysis using two approaches. The first compared the global
connectedness of different regions within a given country. The second treated each region
as a country to see how it would compare if inserted into the MGI Connectedness Index
for 2014.
Regional comparison within each country
For regional comparison within a country, we used a formula for compound intensity similar
to that used in the MGI Connectedness Index:
The key differences are that here regions are treated as the entities. Hence n becomes
the total number of regions in the country, GDP is the GDP of the region, and inflows and
outflows are the global flows for the region. We use the same method to normalize the flow
intensity scores, i.e.:
In this case, we use the maximum and minimum compound intensity scores among the
regions within the country. For the United Kingdom, for example, we analyze 12 regions.
The southeast has the highest compound intensity score, whereas Northern Ireland has the
lowest compound intensity score.
The results show considerable variation in the connectedness of regions within countries,
particularly for China and the United States. Germany, in contrast, has the least variation
among regions in international goods trade.
Exhibit A16
Moar ormulas
Compound flow intensity score
Compound flow intensity score =
(n–1) × (inflow + outflow)2
GDP × global flow
Normalized compound flow intensity score *** SAME for regions
Normalized compound flow intensity score =
99 × (Compound flow intensity score – min)
(max – min)
+ 1
Regional compound score
Compound flow intensity score =
(nregions –1) × (inflowregion + outflowregion)2
GDPregion × country flow
Regions vs. countries
Compound flow intensity score =
(ncountries –1) × (inflowregion + outflowregion)2
GDPregion × world flow
Exhibit A16
Moar ormulas
Compound flow intensity score
Compound flow intensity score =
(n–1) × (inflow + outflow)2
GDP × global flow
Normalized compound flow intensity score *** SAME for regions
Normalized compound flow intensity score =
99 × (Compound flow intensity score – min)
(max – min)
+ 1
Regional compound score
Compound flow intensity score =
(nregions –1) × (inflowregion + outflowregion)2
GDPregion × country flow
Regions vs. countries
Compound flow intensity score =
(ncountries –1) × (inflowregion + outflowregion)2
GDPregion × world flow
132 McKinsey Global Institute Technical appendix
Comparison of regions with countries
To compare regions with countries, the relevant formula for compound intensity
score becomes:
World flow is used instead of country flow to normalize the regional flow, since the region
is treated as a separate country. The n continues to be the overall number of countries.
Since the region is already included as a part of the country, the total world flow remains
the same, and we keep n constant as the number of countries. To calculate the normalized
compound intensity score, we use a similar technique as for countries, using the maximum
and minimum compound intensity scores among all countries for normalizing.
The results show that if states and provinces were countries, many would be among the
most connected in the world. For instance, California would rank fourth in the world in
people flows, while Guangdong would rank sixth in the world in goods flows.
6. GLOBAL SURVEY OF STARTUPS
Background and methodology
In collaboration with 1776, a global incubator and venture fund, and its Challenge Cup
competition and Startup Federation programs, MGI undertook a survey and set of
interviews to understand the extent to which it is now possible for startups to form global
connections from their inception.
The survey included:
� Demographic questions (industry, company stage, home country, company age)
� One core question regarding the types of global activities in which respondents
participate (i.e., users in other countries, talent hired from other countries, funding
from other countries, inputs from other countries, mentors/advisers in other countries,
incubators in other countries)
� Several targeted follow-up questions based on responses to the core question (e.g.,
for those who responded that they had users in other countries, the survey asked the
number of countries with users, the share of user base that was international, and what
that share would likely be in five years)
� A final question asking respondents to rank the barriers to additional global activities.
Exhibit A16
Moar ormulas
Compound flow intensity score
Compound flow intensity score =
(n–1) × (inflow + outflow)2
GDP × global flow
Normalized compound flow intensity score *** SAME for regions
Normalized compound flow intensity score =
99 × (Compound flow intensity score – min)
(max – min)
+ 1
Regional compound score
Compound flow intensity score =
(nregions –1) × (inflowregion + outflowregion)2
GDPregion × country flow
Regions vs. countries
Compound flow intensity score =
(ncountries –1) × (inflowregion + outflowregion)2
GDPregion × world flow
133McKinsey Global Institute Digital globalization: The new era of global flows
The research was conducted in October and November 2015 through two primary
methods. The first involved e-mail outreach to 1776’s global community of startups
and Startup Federation partners, Global Accelerator Network, and current and former
participants in 1776’s Challenge Cup competition to encourage them to participate in the
survey online. This resulted in 168 responses (62 percent of all responses). The second
method involved in-person interviews of startup founders attending Challenge Cup events
in Montreal (October 22, 2015), Washington, DC (October 29), Bogotá (November 5), Beijing
(November 10), Budapest (November 12), Pretoria (November 18), and Mumbai (November
21). These discussions resulted in 103 responses (38 percent of all responses).
� The sectors represented include education, health, cities and transportation, energy
and sustainability, technology and communication, enterprise software, e-commerce,
entertainment, and financial technology, among others.
� Company stages include startups in concept phase (16 percent), product launch
(24 percent), customer validation (21 percent), and growth/scaling phase (39 percent).
� A majority of respondents (59 percent) said their startups were founded in 2014 or 2015,
with the remainder founded between 2006 and 2013.
� US startups accounted for 56 percent of respondents. Also represented were Asia and
Australia (13 percent), Europe (11 percent), Africa and the Middle East (11 percent), and
South America (5 percent).
Despite this diversity, it is important to note that respondents are not typical of the average
startup. Participation in cross-border networks such as the Startup Federation and pitch
competitions like 1776’s global Challenge Cup suggest that respondents are likely more
sophisticated, internationally minded, and technologically savvy than entrepreneurs or small
businesses focused on delivering local services. Even with that caveat, however, the data
suggest that there is a significant and influential subset of digitally empowered startups that
are now global from inception.
The 271 survey participants are a diverse group across a range of metrics (Exhibit A14).
134 McKinsey Global Institute Technical appendix
Exhibit A14
MGI Global Startup Survey 2015: Respondent demographics
SOURCE: McKinsey Global Institute Global Startup Survey 2015
100% = 271 respondents
NOTE: Numbers may not sum due to rounding.
Health
6
6
14
E-commerce
Energy and
sustainability
6
Enterprise software
and security
Technology and
communication
12
Cities and
transportation
7
21
Education
18
Other
Financial technology
4
6Entertainment
Concept
44
(16%)
Customer
validation
66
(24%)
Growth/
scaling105
(39%)
56
(21%)
Product
launch
36
(13%)
2013
2006–11
2012 31
(11%)
45
(17%)
2014
80
(30%)
2015
79
(29%)
Eastern Europe
Other
United
States
4
7
India
South America
2
Israel
4
3
5
China
5
Canada
Western Europe
6
Africa
7
56
In-person
interview 103
(38%)
168
(62%) Online
survey
Sector
%
Stage
Number (%)
Company
age
Number (%)
Geography
%
Survey
method
Number (%)
135McKinsey Global Institute Digital globalization: The new era of global flows
© Getty Images
© Getty Images
A
Abeliansky, Ana L., Imnaculada Martínez-Zarzoso, and Klaus Prettner, The impact of 3D
printing on trade and FDI, Center for European, Governance and Economic Development
Research discussion paper number 262, October 2015.
Aghion, Philippe, and Jeffrey G. Williamson, Growth, inequality, and globalization: Theory,
history, and policy, Cambridge University Press, 1998.
Ahmed, Shaghil and Andrei Zlate, “Capital flows to emerging market economies: A brave
new world?” Journal of International Money and Finance, volume 48, November 2014.
Aizenman, Joshua, Yothin Jinjarak, and Donghyun Park, “Capital flows and economic
growth in the era of financial integration and crisis, 1990–2010,” Open Economies Review,
volume 24, issue 3, July 2013.
Akamai, State of the Internet Q1 2015 report, volume 8, number 1, 2015.
Albright-Stonebridge Group, Data localization: A challenge to global commerce and the free
flow of information, September 2015.
AliResearch and Accenture, Cross-border B2C e-commerce market trends, June 2015.
Autor, David H., David Dorn, and Gordon H. Hanson, The China shock: Learning from
labor market adjustment to large changes in trade, NBER working paper number 21906,
January 2016.
Autor, David H., David Dorn, and Gordon H. Hanson, “The China syndrome: Labor market
effects of import competition in the United States,” American Economic Review, volume
103, number 6, 2013.
B
Bahar, Dany, Ricardo Hausmann, and César A. Hidalgo, “Neighbors and the evolution of
comparative advantage of nations: Evidence of international knowledge diffusion?” Journal
of International Economics, volume 92, number 1, 2014.
Barnett, George A., and Han Woo Park, “Examining the international Internet using multiple
measures: New methods for measuring the communication base of globalized cyberspace,”
Quality and Quantity, volume 48, issue 1, January 2014.
Barro, Robert J., and Xavier Sala-i-Martin, Economic growth, MIT Press, 2003.
Bauer, Matthias, Hosuk Lee-Makiyama, Erik van der Marel, and Bert Verschelde, The costs
of data localisation: Friendly fire on economic recovery, European Centre for International
Political Economy (ECIPE) occasional paper number 3/2014, May 2014.
Bednárová, Pavla, “Global competitiveness of developed market economies and its relation
to dimensions of globalization,” in Proceedings of the 11th International Conference Liberec
Economic Forum, 2013.
Bednárová, Pavla, Sárka Laboutková, and Ales Kocourek, “On the relationship between
globalization and human development,” in Proceedings of the 10th International Conference
Liberec Economic Forum, 2011.
Bell, Jim, and Sharon Loane, “‘New-wave’ global firms: Web 2.0 and SME
internationalisation,” Journal of Marketing Management, volume 26, issue 3–4, April 2010.
BIBLIOGRAPHY
138 McKinsey Global Institute Bibliography
Bergh, Andreas, and Therese Nilsson, “Do liberalization and globalization increase income
inequality?” European Journal of Political Economy, volume 26, issue 4, December 2010.
Bernard, Andrew B., and J. Bradford Jensen, “Exceptional exporter performance: Cause,
effect, or both?” Journal of International Economics, volume 47, issue 1, February 1999.
Bernard, Andrew B., J. Bradford Jensen, and Peter K. Schott, “Trade costs, firms, and
productivity,” Journal of Monetary Economics, volume 53, issue 5, July 2006.
Bianjing, Ma, Xie Zuoshi, and Li Jingkui, Endogenous international trade and economic
growth: An empirical study based on 120 Chinese cities, available at SSRN, August 2011.
Boubtane, Ekrame, Jean-Christophe Dumont, and Christophe Rault, Immigration and
economic growth in the OECD countries, 1986–2006, Institute for the Study of Labor (IZA)
discussion paper number 8681, November 2014.
Borchert, Ingo, Batshur Gootiiz, and Aaditya Mattoo, Policy barriers to international trade in
services: New empirical evidence, World Bank policy research working paper number 6109,
June 2012.
Busse, Matthias, and Jens Königer, Trade and economic growth: A re-examination of the
empirical evidence, Hamburg Institute of International Economics research paper number
123, 2012.
C
Castro, Daniel, and Alan McQuinn, Cross-border data flows enable growth in all industries,
Information Technology and Innovation Foundation, February 2015.
Center for Strategic and International Studies and McAfee, Net losses: Estimating the global
cost of cybercrime, June 2014.
Centre for Entrepreneurship, SMEs and Local Development, Fostering SMEs’ participation
in global markets: Final report, July 2013.
Chakravorti, Bhaskar, Christopher Tunnard, and Ravi Shankar Chaturvedi, Digital
planet: Readying for the rise of the e-consumer, The Fletcher School, Tufts University,
September 2014.
Chang, Chun-Ping, Chien-Chiang Lee, and Meng-Chi Hsieh, “Does globalization promote
real output? Evidence from quantile cointegration regression,” Economic Modelling, volume
44, 2015.
Chinn, David, James Kaplan, and Allen Weinberg, Risk and responsibility in a
hyperconnected world: Implications for enterprises, McKinsey & Company and the World
Economic Forum, January 2014.
Choi, Changkyu, and Myung Hoon Yi, “The effects of the Internet on economic growth:
Evidence from cross-country panel data,” Economic Letters, volume 105, issue 1,
October 2009.
Cisco, Cisco Visual Networking Index: Forecast and methodology, 2014–2019, May 2015.
Cisco, Cisco Visual Networking Index: Global mobile data traffic forecast update, 2014–
2019, February 2015.
D
De Loecker, Jan, and Pinelopi Koujianou Goldberg, “Firm performance in a global market,”
Annual Review of Economics, October 2013.
Dewhurst, Martin, Jonathan Harris, and Suzanne Heywood, “Understanding your
‘globalization penalty,’” McKinsey Quarterly, July 2011.
139McKinsey Global Institute Digital globalization: The new era of global flows
Dutta, Soumitra, Bruno Lanvin, and Sacha Wunsch-Vincent, eds., The global innovation
index 2015: Effective innovation policies for development, Cornell University, Insead, and
World Intellectual Property Organization, 2015.
E
eBay, Small online business growth report: Towards an inclusive global economy,
January 2016.
Economist Intelligence Unit, The ICT globalisation index, 2014.
Economist Intelligence Unit, Networked manufacturing: The digital future, April 2014.
Ellison, Nicole, “Connection strategies: Social capital implications of Facebook-enabled
communication practices,” New Media & Society, volume 13, number 6, September 2011.
Evenett, Simon J., and Johannes Fritz, The tide turns? Trade, protectionism, and slowing
global growth, Centre for Economic Policy Research, 18th Global Trade Alert Report, 2015.
Executive Office of the President, “The economic benefits of US trade,” May 2015.
F
Falk, Martin, and Eva Hagsten, E-commerce trends and impacts across Europe, UNCTAD
discussion paper number 220, March 2015.
Francois, Joseph, Bertin Martens, and Fan Yang, The macro-economic impact of cross-
border ecommerce in the EU, European Commission Joint Research Centre, 2014.
Frankel, Jeffrey A., and David Romer, “Does trade cause growth?” American Economic
Review, volume 89, issue 3, June 1999.
Friedman, Thomas L., The world is flat 3.0: A brief history of the twenty-first century,
Picador, 2007.
G
George, Katy, Sree Ramaswamy, and Lou Rassey, “Next-shoring: A CEO’s guide,”
McKinsey Quarterly, January 2014.
Ghemawat, Pankaj, and Steven A. Altman, Depth Index of Globalization 2013: And the big
shift to emerging economies, IESE Business School, University of Navarra, 2013.
Ghemawat, Pankaj, and Steven A. Altman, DHL Global Connectedness Index 2014:
Analyzing global flows and their power to increase prosperity, DHL, 2014.
Google, Google economic impact report, United States 2014, August 2014.
Grossman, Gene M., and Elhanan Helpman, “Globalization and growth,” American
Economic Review, volume 105, number 5, May 2015.
H
Hatton, Timothy J., “The economics of international migration: A short history of the debate,”
Labour Economics, volume 30, 2014.
Hausmann, Ricardo, César A. Hidalgo, Sebastian Bustos, Michele Coscia, Alexander
Simoes, and Muhammed A. Yildirim, The atlas of economic complexity: Mapping paths to
prosperity (new edition), Center for International Development, Harvard University, 2013.
Hofheinz, Paul, and Michael Mandel, Uncovering the hidden value of digital trade: Towards
a 21st century agenda of transatlantic prosperity, Lisbon Council and Progressive Policy
Institute interactive policy brief issue 19/2015, July 2015.
140 McKinsey Global Institute Bibliography
I
IAB Europe, Consumers driving the digital uptake: The economic value of online advertising-
based services for consumers, September 2010.
International Post Corporation, IPC cross-border e-commerce report, 2010.
International Telecommunication Union and UNESCO, The state of broadband 2015,
September 2015.
J
Jensen, J. Bradford, Global trade in services: Fear, facts, and offshoring, Peterson Institute
for International Economics, September 2011.
K
Kaplan, James M., and Kayvaun Rowshankish, Addressing the impact of data location
regulation in financial services, Global Commission on Internet Governance paper series
number 14, Centre for International Governance Innovation and Chatham House, May 2015.
Kelly, Sanja, Madeline Earp, Laura Reed, Adrian Shahbaz, and Mai Truong, Freedom on
the Net 2014: Tightening the Net: Governments expand online controls, Freedom House,
October 2014.
Kilic, Cuneyt, Effects of globalization on economic growth: Panel data analysis for
developing countries, University of Ploiesti Bulletin, volume 67, 2015.
Kletzer, Lori G., “Why the US needs wage insurance,” Harvard Business Review, January
25, 2016.
Koepke, Robin, Capital flows to emerging markets, Institute of International Finance,
January 2015.
Kose, M. Ayhan, Eswar S. Prasad, and Marco E. Terrones, Does openness to international
financial flows raise productivity growth? NBER working paper number 14558,
December 2008.
Kumar, Nirmalya, and Phanish Puranam, “Emerging markets: Have you restructured for
global success?” Harvard Business Review, October 2011.
M
Mandel, Michael, Beyond goods and services: The (unmeasured) rise of the data-driven
economy, Progressive Policy Institute, October 2012.
Mandel, Michael, Data, trade, and growth, Progressive Policy Institute, August 2013.
Martens, Pim, Marco Caselli, Philippe De Lombaerde, Lukas Figge, and Jan Aart Scholte,
“New directions in globalization indices,” Globalization, volume 12, issue 2, 2015.
McAnaney, Patrick, Innovation that matters: How city networks drive civic entrepreneurship,
1776, US Chamber of Commerce Foundation, and Free Enterprise, May 2015.
McKinsey Global Institute, Digital America: A tale of the haves and have-mores,
December 2015.
McKinsey Global Institute, Financial globalization: Retreat or reset? March 2013.
McKinsey Global Institute, Global flows in a digital age: How trade, finance, people, and data
connect the world economy, April 2014.
McKinsey Global Institute, The Internet of Things: Mapping the value beyond the hype,
June 2015.
141McKinsey Global Institute Digital globalization: The new era of global flows
McKinsey Global Institute, A labor market that works: Connecting talent with opportunity in
the digital age, June 2015.
McKinsey Global Institute, Manufacturing the future: The next era of global growth and
innovation, November 2012.
McKinsey Global Institute, Playing to win: The new global competition for corporate profits,
September 2015.
McKinsey Global Institute, Urban world: The shifting global business landscape,
October 2013.
McKinsey Technology, Media and Telecom Practice, Offline and falling behind: Barriers to
Internet adoption, September 2014.
Meijers, Huub, “Does the Internet generate economic growth, international trade, or both?”
International Economics and Economic Policy, volume 11, issue 1, February 2014.
Melitz, Marc J., and Gianmarco I. P. Ottaviano, “Market size, trade, and productivity,” Review
of Economic Studies, volume 75, issue 1, 2008.
Meltzer, Joshua P., The importance of the Internet and transatlantic data flows for US and
EU trade and investment, Global Economy and Development working paper number 79,
Brookings Institution, October 2014.
Meltzer, Joshua P., Maximizing the opportunities of the Internet for international trade, The
E15 Initiative policy options paper, January 2016.
Meltzer, Joshua P., Using the Internet to promote services exports by small-and medium-
sized enterprises, Global Economy and Development working paper number 83, Brookings
Institution, February 2015.
Mettler, Ann, and Anthony D. Williams, The rise of the micro-multinational: How freelancers
and technology-savvy start-ups are driving growth, jobs and innovation, Lisbon Council
policy brief, June 2012.
Miller, Ben, and Robert D. Atkinson, Digital drag: Ranking 125 nations by taxes and tariffs on
ICT goods and services, Information Technology and Innovation Foundation, October 2014.
Mun, Har Wei, Teo Kai Lin, and Yee Kar Man, “FDI and economic growth relationship:
An empirical study on Malaysia,” International Business Research, volume 1, number 2,
April 2008.
N
Neto, Delfim Gomes, and Francisco José Veiga, “Financial globalization, convergence and
growth: The role of foreign direct investment,” Journal of International Money and Finance,
volume 37, October 2013.
Nicholson, Jessica R., and Ryan Noonan, Digital economy and cross-border trade: The
value of digitally-deliverable services, US Department of Commerce, ESA issue brief number
01-14, January 2014.
O
OECD, Interconnected economies: Benefiting from global value chains, May 2013.
OECD, OECD digital economy outlook 2015, 2015.
OECD, Measuring the digital economy: A new perspective, December 2014.
OECD, Supporting investment in knowledge capital, growth and innovation, October 2013.
Olarreaga, Marcelo, and Sidley Austin LLP, Enabling traders to enter and grow on the global
stage, eBay, 2012.
142 McKinsey Global Institute Bibliography
Osterhammel, Jürgen, and Niels P. Petersson, Globalization: A short history, Princeton
University Press, 2009.
P
Peri, Giovanni, The effect of immigration on productivity: Evidence from US states, NBER
working paper number 15507, November 2009.
Potrafke, Niklas, “The evidence on globalisation,” World Economy, volume 38, number 3,
March 2015.
Q
Qiang, Christine Zhen-Wei and Carlo M. Rossotto with Kaoru Kimura, “Economic impacts
of broadband,” in Information and communications for development 2009: Extending reach
and increasing impact, World Bank, 2009.
R
Rao, B. Bhaskara, Artur Tamazian, and Krishna Chaitanya Vadlamannati, “Growth effects of
a comprehensive measure of globalization with country-specific time series data,” Applied
Economics, volume 43, issue 5, 2011.
Reinhart, Carmen, and Vincent Reinhart, “Capital flow bonanzas: An encompassing view
of the past and present,” in NBER International Seminar on Macroeconomics 2008, Jeffrey
Frankel and Christopher Pissarides, eds., University of Chicago Press, 2009.
S
Samuelson, Paul A., “Where Ricardo and Mill rebut and confirm arguments of mainstream
economists supporting globalization,” Journal of Economic Perspectives, volume 18,
number 3, summer 2004.
Sarac, Aysegul, Nabil Absi, and Stéphane Dauzère-Pérès, “A literature review on the impact
of RFID technologies on supply chain management,” International Journal of Production
Economics, volume 128, number 1, November 2010.
Shahbaz, Muhammad, “Does trade openness affect long run growth? Cointegration,
causality and forecast error variance decomposition tests for Pakistan,” Economic
Modelling, volume 29, 2012.
Smick, David, “Could globalization crack up?” International Economy, fall 2012.
Spence, Michael, The next convergence: The future of economic growth in a multispeed
world, Farrar, Straus and Giroux, 2011.
Spence, Michael, and Sandile Hlatshwayo, The evolving structure of the American economy
and the employment challenge, Council on Foreign Relations working paper, March 2011.
Sui, Daniel, James Caverlee, and Dakota Rudesill, The Deep Web and the Darknet: A look
inside the Internet’s massive black box, Woodrow Wilson International Center for Scholars,
October 2015.
T
Tang, Kam Ki, “Correcting the size bias in trade openness and globalization measures,”
Global Economy Journal, volume 11, issue 3, 2011.
TeleGeography, Global Bandwidth Research Service; Global Bandwidth Forecast Service;
Global Internet Geography, www.telegeography.com.
Terziaa, Nuray, “The impact of e-commerce on international trade and employment,”
Procedia Social and Behavioral Sciences, volume 24, 2011.
http://www.telegeography.com
143McKinsey Global Institute Digital globalization: The new era of global flows
Timmis, Jonathan, The Internet and international trade in goods, University of Nottingham
School of Economics discussion paper, August 2012.
U
UNCTAD, Global value chains and development: Investment and value added trade in the
global economy, February 2013.
UNCTAD, Information economy report 2015: Unlocking the potential of ecommerce for
developing countries, March 2015.
UN High Commissioner on Refugees, UNHCR global trends report: World at war,
June 2015.
United Parcel Service, Change in the (supply) chain, 2015.
United States International Trade Commission, Digital trade in the US and global economies,
part 2, August 2014.
V
Villaverde, José, and Adolfo Maza, “Globalisation, growth and convergence,” World
Economy, volume 34, issue 6, June 2011.
W
Wacziarg, Romain, “Measuring the dynamic gains from trade,” World Bank Economic
Review, volume 15, number 3, October 2001.
Westmore, Ben, International migration: The relationship with economic and policy factors in
the home and destination country, OECD economics department working paper, July 2014.
World Bank, “Potential macroeconomic implications of the Trans-Pacific Partnership,” in
Global economic prospects: Spillovers amid weak growth, World Bank Group Flagship
Report, January 2016.
World Bank, World development report 2016: Digital dividends, January 2016.
World Bank, World development report 2016: Internet for development, concept note,
report number 91877, December 2014.
World Trade Organization, World trade report 2011: The WTO and preferential trade
agreements: From co-existence to coherence, 2011.
RELATED MGI
AND MCKINSEY RESEARCH
Digital America: A tale of the haves and
have-mores (December 2015)
While the most advanced sectors, companies,
and individuals push the boundaries of
technology use, the US economy as a whole is
realizing only 18 percent of its digital potential.
Playing to win: The new global competition
for corporate profits (September 2015)
A 30-year period of unprecedented corporate-
profit growth could be drawing to a close.
Competition is intensifying as emerging-market
companies go global and technology-enabled
firms make rapid moves into new sectors.
The Internet of Things: Mapping the value
beyond the hype (June 2015)
If policy makers and businesses get it right,
linking the physical and digital worlds could
generate up to $11.1 trillion a year in economic
value by 2025.
Global flows in a digital age: How trade,
finance, people, and data connect the world
economy (April 2014)
The movement of goods and services, finance,
and people has reached previously unimagined
levels. Global flows are creating new degrees of
connectedness among economies—and playing
an ever-larger role in determining the fate of
nations, companies, and individuals.
Financial globalization: Retreat or reset?
(March 2013)
Cross-border capital flows remain far below
their pre-crisis peak, and growth in financial
assets around the world has stalled. Continued
retrenchment could jeopardize investment unless
policy makers can “reset” the financial system
for a healthier flow of financing that supports
economic growth.
Disruptive technologies: Advances that will
transform life, business, and the global
economy (May 2013)
This report identifies 12 technologies that could
drive massive economic transformations in the
coming years. It looks at how these technologies
could change our world, as well as their benefits
and challenges, and offers guidelines to help
business and policy leaders respond.
www.mckinsey.com/mgi
E-book versions of selected MGI reports are available at MGI’s website,
Amazon’s Kindle bookstore, and Apple’s iBooks Store.
Download and listen to MGI podcasts on iTunes or at
www.mckinsey.com/mgi/publications/multimedia/
Cover images all © Getty Images.
Contents page images all © Getty Images.
http://www.mckinsey.com/insights/high_tech_telecoms_internet/digital_america_a_tale_of_the_haves_and_have_mores
http://www.mckinsey.com/insights/high_tech_telecoms_internet/digital_america_a_tale_of_the_haves_and_have_mores
http://www.mckinsey.com/insights/corporate_finance/the_new_global_competition_for_corporate_profits
http://www.mckinsey.com/insights/corporate_finance/the_new_global_competition_for_corporate_profits
http://www.mckinsey.com/insights/business_technology/the_internet_of_things_the_value_of_digitizing_the_physical_world
http://www.mckinsey.com/insights/business_technology/the_internet_of_things_the_value_of_digitizing_the_physical_world
http://www.mckinsey.com/insights/globalization/global_flows_in_a_digital_age
http://www.mckinsey.com/insights/globalization/global_flows_in_a_digital_age
http://www.mckinsey.com/insights/globalization/global_flows_in_a_digital_age
http://www.mckinsey.com/insights/global_capital_markets/financial_globalization
http://www.mckinsey.com/insights/global_capital_markets/financial_globalization
http://www.mckinsey.com/insights/business_technology/disruptive_technologies
http://www.mckinsey.com/insights/business_technology/disruptive_technologies
http://www.mckinsey.com/insights/business_technology/disruptive_technologies
McKinsey Global Institute
March 2016
Copyright © McKinsey & Company
www.mckinsey.com/mgi
@McKinsey_MGI
McKinseyGlobalInstitute
www.mckinsey.com/mgi
Broadening participation Boosting productivity
and GDP
Changing the way
companies go global
857343
HIGHLIGHTS
MARCH 2016
DIGITAL GLOBALIZATION:
THE NEW ERA OF GLOBAL FLOWS
Copyright © McKinsey & Company 2016
In the 25 years since its founding, the McKinsey Global Institute (MGI) has sought to develop
a deeper understanding of the evolving global economy. As the business and economics
research arm of McKinsey & Company, MGI aims to provide leaders in the commercial,
public, and social sectors with the facts and insights on which to base management and
policy decisions. We are proud to be ranked the top private-sector think tank, according
to the authoritative 2015 Global Go To Think Tank Index, an annual report issued by the
University of Pennsylvania Think Tanks and Civil Societies Program at the Lauder Institute.
MGI research combines the disciplines of economics and management, employing the
analytical tools of economics with the insights of business leaders. Our “micro-to-macro”
methodology examines microeconomic industry trends to better understand the broad
macroeconomic forces affecting business strategy and public policy. MGI’s in-depth reports
have covered more than 20 countries and 30 industries. Current research focuses on six
themes: productivity and growth, natural resources, labor markets, the evolution of global
financial markets, the economic impact of technology and innovation, and urbanization.
Recent reports have assessed global flows; the economies of Brazil, Mexico, Nigeria, and
Japan; China’s digital transformation; India’s path from poverty to empowerment; affordable
housing; the effects of global debt; and the economics of tackling obesity.
MGI is led by three McKinsey & Company directors: Richard Dobbs, James Manyika,
and Jonathan Woetzel. Michael Chui, Susan Lund, Anu Madgavkar, and Jaana Remes
serve as MGI partners. Project teams are led by the MGI partners and a group of senior
fellows, and include consultants from McKinsey & Company’s offices around the world.
These teams draw on McKinsey & Company’s global network of partners and industry and
management experts. In addition, leading economists, including Nobel laureates, act as
research advisers.
The partners of McKinsey & Company fund MGI’s research; it is not commissioned by
any business, government, or other institution. For further information about MGI and to
download reports, please visit www.mckinsey.com/mgi.
http://www.mckinsey.com/mgi
James Manyika | San Francisco
Susan Lund | Washington, DC
Jacques Bughin | Brussels
Jonathan Woetzel | Shanghai
Kalin Stamenov | New York
Dhruv Dhingra | New York
MARCH 2016
DIGITAL GLOBALIZATION:
THE NEW ERA OF GLOBAL FLOWS
PREFACE
The web of global economic connections is growing deeper, broader, and
more intricate. Yet much of the public discussion surrounding globalization
is stuck on the narrow topic of trade surpluses and deficits. This lens fails to
take into account the new and more complex reality of a digitally connected
global economy. While the global goods trade and financial flows have
flattened since the Great Recession, cross-border flows of data are surging.
They now tie the world economy together just as surely as flows of traditional
manufactured goods.
Two years ago, the McKinsey Global Institute (MGI) set out to paint a
comprehensive picture of how globalization is evolving. The resulting report,
Global flows in a digital age: How trade, finance, people, and data connect the
world economy, assessed the network of cross-border inflows and outflows
of trade, services, finance, people, and data and its influence on economic
growth. Building on that earlier work, this report provides a more detailed
analysis of how global flows are continuing to evolve. It offers new insights
into how companies and countries are participating in the web of flows and
extends our econometric analysis, drawing on improved data and employing
more sophisticated methodology. We find even stronger evidence of the
economic value of participating in global flows—and we further find that
data flows account for a substantial portion of that impact. Both inflows and
outflows matter for growth as they circulate ideas, research, technologies,
talent, and best practices around the world.
Today’s more digital form of globalization is changing who is participating, how
business is done across borders, how rapidly competition moves, and where
the economic benefits are flowing. Even though advanced economies in
general continue to be the leaders in most flows, the door has opened to more
countries, to small companies and startups, and to billions of individuals. Our
previous research found the biggest benefits of trade flows go to countries at
the center of the global network. Interestingly, this report finds that countries at
the periphery of the network of data flows stand to gain even more than those
at the center. The convergence of globalization and digitization means that
business leaders and policy makers will need to reassess their strategies—and
given that we are only in the very early stages of this phenomenon, enormous
opportunities are still at stake.
This research was led by James Manyika, a director of the McKinsey Global
Institute based in San Francisco; Susan Lund, an MGI partner based in
Washington, DC; Jacques Bughin, a McKinsey director based in Brussels
who is a core leader of the Firm’s High Tech, Telecom, and Media Practice,
a current member of the MGI Council, and an incoming director of MGI;
and Jonathan Woetzel, an MGI director based in Shanghai. The project
team, led by Kalin Stamenov and Dhruv Dhingra, included Laura Cappellin,
Ritesh Jain, Ayush Mittal, Katie Ramish, Soyoko Umeno, and Amber Yang.
Esteban Arias, Joana Carreiro, Carlos Molina, Moira Pierce, and Vivien Singer
provided valuable research and analytics support. Lisa Renaud served as
senior editor. Sincere thanks go to our colleagues in operations, design,
production, and external relations, including Tim Beacom, Marisa Carder,
Matt Cooke, Deadra Henderson, Richard Johnson, Julie Philpot, Mary Reddy,
Rebeca Robboy, Margo Shimasaki, and Patrick White.
We thank McKinsey colleagues Jörg Bromberger, Michael Chui, Diaan-Yi Lin,
and Sree Ramaswamy for sharing their expertise and insights. Special thanks
go to the volunteers who helped us conduct our survey of global startups:
Ricardo Bernal, Mayank Bishnoi, Oleksandr Bondarenko, Tamas Csikai,
Karol Dolega, Rishika Garg, Thomas Grandin, Catherine Hart, Valeria Laszlo,
Mohato Lekena, Thandi Luzuka, Victoria Muwanga-Zake, Mohit Narotam, and
Siyi Amy Shi.
Our academic advisers provided valuable insights and challenged our
thinking. We are grateful to Matthew J. Slaughter, the Paul Danos Dean of the
Tuck School of Business at Dartmouth; Michael Spence, Nobel laureate and
William R. Berkley Professor in Economics and Business at NYU Stern School
of Business; and Laura Tyson, professor of business administration and
economics at the Haas School of Business, University of California, Berkeley.
Merit E. Janow, dean of the School of International and Public Affairs at
Columbia University, offered new perspectives and directions for our research.
Philip R. Lane, Governor of the Central Bank of Ireland and former Whatley
Professor of Political Economy at Trinity College, Dublin, generously shared
data and perspectives on the international investment positions of countries.
A number of individuals and organizations generously contributed their
time, data, and expertise. For their support in surveying startups, we thank:
Donna Harris, Patrick McAnaney, Morgan Gress, and Kaitlin Walls of 1776,
a global incubator and venture fund dedicated to accelerating innovation
in areas of essential human need. We are also grateful to Molly Jackman
of Facebook; Usman Ahmed of PayPal; Alan Elias of eBay; Robert Pepper
of Cisco; Jarrad Hubbard of TeleGeography; and Uwe Deichmann,
Deepak Mishra, and Daria Taglioni of the World Bank. Without them, this
report would not have been possible.
This report contributes to MGI’s mission to help business and policy leaders
understand the forces transforming the global economy, identify strategic
locations, and prepare for the next wave of growth. As with all MGI research,
this work is independent and has not been commissioned or sponsored in
any way by any business, government, or other institution. We welcome your
comments on the research at [email protected]
Richard Dobbs
Director, McKinsey Global Institute
London
James Manyika
Director, McKinsey Global Institute
San Francisco
Jonathan Woetzel
Director, McKinsey Global Institute
Shanghai
March 2016
© Shutterstock
CONTENTS
HIGHLIGHTS
Soaring cross-border
data flows
The MGI
Connectedness Index
How regions and
cities participate
30
57
63
In brief
Executive summary Page 1
1. A new era of digital globalization Page 23
2. Digital platforms open the door to new participants Page 43
3. How countries, cities, and regions are connecting Page 55
4. Global flows boost economic growth Page 73
5. Competing in a digital global landscape Page 85
6. The new world of policy challenges Page 97
Technical appendix Page 105
Bibliography Page 137
IN BRIEF
DIGITAL GLOBALIZATION:
THE NEW ERA OF GLOBAL FLOWS
The rapidly growing flows of international trade and finance that characterized the 20th century have flattened or
declined since 2008. Yet globalization is not moving into reverse. Instead digital flows are soaring—transmitting
information, ideas, and innovation around the world and broadening participation in the global economy.
� The world is more interconnected than ever. For the first time in history, emerging economies are counterparts
on more than half of global trade flows, and South-South trade is the fastest-growing type of connection.
� While flows of goods and finance have lost momentum, used cross-border bandwidth has grown 45 times
larger since 2005. It is projected to grow by another nine times in the next five years as digital flows of
commerce, information, searches, video, communication, and intracompany traffic continue to surge.
� Digital platforms change the economics of doing business across borders, bringing down the cost of
international interactions and transactions. They create markets and user communities with global scale,
providing businesses with a huge base of potential customers and effective ways to reach them.
� Small businesses worldwide are becoming “micro-multinationals” by using digital platforms such as eBay,
Amazon, Facebook, and Alibaba to connect with customers and suppliers in other countries. Even the
smallest enterprises can be born global: 86 percent of tech-based startups we surveyed report some
type of cross-border activity. The ability of small businesses to reach new markets supports economic
growth everywhere.
� Individuals are participating in globalization directly, using digital platforms to learn, find work, showcase their
talent, and build personal networks. Some 900 million people have international connections on social media,
and 360 million take part in cross-border e-commerce.
� Over a decade, global flows have raised world GDP by at least 10 percent; this value totaled $7.8 trillion
in 2014 alone. Data flows now account for a larger share of this impact than global trade in goods. Global
flows generate economic growth primarily by raising productivity, and countries benefit from both inflows
and outflows.
� The MGI Connectedness Index offers a comprehensive look at how countries participate in inflows and
outflows of goods, services, finance, people, and data. Singapore tops the latest rankings, followed by the
Netherlands, the United States, and Germany. China has surged from No. 25 to No. 7.
� Although more nations are participating, global flows remain concentrated among a small set of leading
countries. The gaps between the leaders and the rest of the world are closing very slowly, but catch-up
growth represents a major opportunity for lagging countries. Some economies could grow by 50 percent or
more over the long term by accelerating participation.
� Many companies grew more complex and inefficient as they expanded across borders. But digital
technologies can tame complexity and create leaner models for going global. This is a moment for companies
to rethink their organizational structures, products, assets, and competitors.
Countries cannot afford to shut themselves off from global flows, but narrow export strategies miss the real value
of globalization: the flow of ideas, talent, and inputs that spur innovation and productivity. Digital globalization
makes policy choices even more complex. Value chains are shifting, new hubs are emerging, and economic
activity is being transformed. This transition creates new openings for countries to carve out profitable roles in the
global economy. Those opportunities will favor locations that build the infrastructure, institutions, and business
environments that their companies and citizens need to participate fully.
Increase in world GDP,
worth $7.8T in 2014
GDP increase from data
flows, larger impact than
goods trade
Global flows increase economic growth
10% $2.8T
Potential GDP boost for some
countries by increasing
participation in
global flows
~50%
Digital technologies are changing how business is done
across borders and broadening participation
Large multinationals
Attain truly global scale with new
markets and suppliers
New strategies for products,
assets, organization
Individuals
New ways to work, learn, and
communicate across borders
>900M have international
connections on social media
SMEs
Use digital platforms to find
customers and suppliers abroad
50M on Facebook, 10M on Alibaba,
2M on Amazon
Startups
>80% of tech-based startups are
“born global”
Foreign customers, financing,
suppliers from day one
Global flows of trade and finance are flattening, while data flows are soaring
1980 2014
DATATRADE FINANCE
45X
growth in
data flows
2005–2014
The new era of digital globalization
© Getty Images
EXECUTIVE SUMMARY
Somewhere in Kenya, a girl logs on for a personalized math lesson from California-based
Khan Academy. Thousands of Syrian refugees rely on Facebook updates for the latest
information to guide their journey through Europe. A multinational energy giant launches
plans to use sensors on 4,000 oil wells around the world to monitor production remotely.
A manufacturer in Australia buys components from a Chinese supplier on Alibaba, and a
clinical trial in India transmits patient data to US pharmaceutical researchers.
The world has become more intricately connected than ever before. Back in 1990, the total
value of global flows of goods, services, and finance amounted to $5 trillion, or 24 percent
of world GDP. There were some 435 million international tourist arrivals, and the public
Internet was in its infancy. Fast forward to 2014: some $30 trillion worth of goods, services,
and finance, equivalent to 39 percent of GDP, was exchanged across the world’s borders.
International tourist arrivals soared above 1.1 billion. And the Internet is now a global network
instantly connecting billions of people and countless companies around the world.
Flows of physical goods and finance were the hallmarks of the 20th-century global
economy, but today those flows have flattened or declined. Twenty-first-century
globalization is increasingly defined by flows of data and information. This phenomenon now
underpins virtually all cross-border transactions within traditional flows while simultaneously
transmitting a valuable stream of ideas and innovation around the world.1
Digitization changes the economics of globalization in several ways. As digital platforms
become global in scope, they are driving down the cost of cross-border communications
and transactions, allowing businesses to connect with customers and suppliers in any
country. Globalization was once for large multinational corporations, but platforms reduce
the minimum scale needed to go global, enabling small business and entrepreneurs around
the world to participate. As a result, new types of competitors can emerge rapidly from any
corner of the world, increasing pressure on industry incumbents.
More than ever before, companies and countries cannot afford to ignore the opportunities
beyond their own borders. Our econometric research indicates that global flows of
goods, foreign direct investment, and data have increased current global GDP by roughly
10 percent compared to what would have occurred in a world without any flows. This
value was equivalent to $7.8 trillion in 2014 alone. Data flows account for $2.8 trillion of this
effect, exerting a larger impact on growth than traditional goods flows. This is a remarkable
development given that the world’s trade networks have developed over centuries but
cross-border data flows were nascent just 15 years ago.
1 This research builds on the 2014 McKinsey Global Institute report Global flows in a digital age: How trade,
finance, people, and data connect the world economy.
The shift to a more digital form of globalization
changes who is participating, how business is done
across borders, and where the economic benefits
are flowing.
2 McKinsey Global Institute Executive summary
Global flows support growth by raising productivity and creating more efficient markets with
truly global scale. But not all countries are making the most of this potential. Our updated
MGI Connectedness Index ranks countries on inflows and outflows of goods, services,
finance, people, and data. Advanced economies are still the most globally connected.
Although more developing countries are deepening their participation, they are narrowing
the gap with the leading advanced economies only very slowly over time.
Accelerating catch-up growth is a major opportunity for the developing world. Our 2014
report showed that countries in the center of trade networks derive more benefit from goods
flows than countries with few connections. But our new research shows that data flows offer
stronger economic benefits to countries on the periphery of the world’s digital networks.
The new age of digital globalization also poses challenges. Companies can enter new
markets, but they are exposed to pricing pressures, aggressive global competitors, and
disruptive digital business models. Data has to be protected against cybercrime. Students
can educate themselves online from anywhere on earth, but their view into other societies
can heighten their impatience with bleak job prospects at home. Social media creates
global communities but also allows networks of extremists to connect. It will take more
international coordination to deal with many of these issues. Today’s version of globalization
is vastly more complex and fast-paced, but connectedness can be a path to growth.
A NEW ERA OF DIGITAL GLOBALIZATION HAS BEGUN
The world has never been more deeply connected by commerce, communication,
and travel than it is today. But the pattern of globalization is shifting. Trade was once
dominated by tangible goods and was largely confined to advanced economies and
their large multinational companies. Today global data flows are surging, and digital
platforms allow more countries and smaller enterprises to participate. This shift has far-
reaching implications.
After a 20-year period of growing roughly twice as fast as the world economy, global flows of
goods, services, and finance hit roughly $30 trillion in 2007, peaking at 53 percent of global
GDP. But this rapid expansion has stopped in its tracks. Growth in global goods trade has
flattened, financial flows have fallen sharply, and trade in services has posted only modest
growth. These flows have finally regained their pre-recession levels in terms of dollar value,
but they are now just 39 percent of world GDP (Exhibit E1).
Many observers point to this trend as evidence that globalization has stopped.2 We have
a different view: globalization has instead entered a new era defined by data flows that
transmit information, ideas, and innovation. Digital platforms create more efficient and
transparent global markets in which far-flung buyers and sellers find each other with a few
clicks. The near-zero marginal costs of digital communications and transactions open new
possibilities for conducting business across borders on a massive scale.
2 See, for example, David Smick, “Could globalization crack up?” International Economy, fall 2012; Joshua
Cooper Ramo, “Globalism goes backward,” Fortune, November 20, 2012; and Jeffrey Rothfeder, “The great
unraveling of globalization,” Washington Post, April 24, 2015.
Soaring cross-border data flows now generate
more economic value than traditional flows of
traded goods.
3McKinsey Global Institute Digital globalization: The new era of global flows
Traditional flows of goods, services, and finance have flattened
For two decades, the world’s trade in goods (including commodities, finished goods, and
intermediate inputs) grew roughly twice as fast as global GDP as major multinationals
expanded their supply chains and established new bases of production in countries with
low-cost labor. Global trade in goods soared from 13.8 percent of world GDP in 1986 to
26.6 percent in 2008 on the eve of the Great Recession. After a sharp decline and short-
lived rebound, however, the goods trade has been growing more slowly than world GDP
in recent years, puzzling economists and business leaders alike. Some of this decline is
cyclical. Our analysis suggests that weak demand and plummeting prices for commodities
account for nearly three-quarters of the decline in trade.
But trade in both finished and intermediate manufactured goods has also declined, thanks
to several structural forces. The makers of many finished goods are beginning to place less
importance on labor costs and more on speed to market and non-labor costs. As a result,
some production is moving closer to end consumers. Trade is also declining for many
intermediate goods such as chemicals, paper, textile fabrics, and communications and
electrical equipment. This suggests that global value chains may be shortening, at least in
part because of the cost of managing complex, lengthy supply chains.
In the decade ahead, the global goods trade may continue to decline relative to world GDP.
At a minimum, it is unlikely to resume rapid growth. Not only are factor costs changing, but
3D printing and other technologies also have the potential to transform how—and where—
goods such as electronics, vehicle parts, other transportation equipment, machinery and
electrical equipment, medical instruments, and apparel are produced.
Cross-border financial flows—which include lending, foreign direct investment (FDI), and
purchases of equities and bonds—link together national financial markets, connecting
borrowers and savers from different countries. They grew from $0.5 trillion in 1980
(4.1 percent of global GDP) to $11.9 trillion in 2007 (20.7 percent of global GDP). But 2007
proved to be the height of a global credit bubble. Since then financial flows have fallen to less
Exhibit E1
39
3738
4141
53
37
32
27
24
2322
24
2221
2324
26
2014
18
28
3029
9
655
8
29
109
32
6
29
1990
24
5
10
643 433 333
1980
3
10
32
2000
12 11
49
28
3735
22
17
41
32
46
2007
30 31
24
21
13
26
SOURCE: UNCTAD; IMF Balance of Payments; W orld Bank; McKinsey Global Institute analysis
Finance
Goods
Services
After 20 years of rapid growth, traditional flows of goods, services, and finance have declined relative to GDP
Flows of goods, services, and finance, 1980–2014
$ trillion, nominal All flows as % of GDP
-14 p.p.
Global flows 2
ES
mc 0307
4 McKinsey Global Institute Executive summary
than half their previous value ($5.2 trillion in 2014); they are only one-third as high relative to
global GDP.3 A decline in cross-border lending accounts for the majority of the overall drop
in financial flows and may reflect a return to long-term trend. But other types of portfolio
investment and FDI have also fallen, raising concerns about financing for emerging markets.
Accelerating flows of data and information are changing the dynamics
of globalization
While global flows of trade and finance have lost momentum, the volume of data being
transmitted across borders has surged, creating an intricate web that connects countries,
companies, and individuals (Exhibits E2 and E3).4
Global flows of data primarily consist of information, searches, communications,
transactions, video, and intracompany traffic. They underpin and enable virtually every other
kind of cross-border flow. Container ships still move products to markets around the world,
but now customers order them online, track their movement using RFID codes, and pay
for them via digital transactions. Although videos use a majority of Internet bandwidth, the
Internet of Things and other business applications are gaining importance. Indeed, Cisco
estimates that machine-to-machine connections will account for more than 40 percent of
global devices and connections by 2019.5
3 Financial globalization: Retreat or reset? McKinsey Global Institute, March 2013.
4 To measure these flows, we track used cross-border bandwidth, which is highly correlated with Internet traffic.
5 Cisco Visual Networking Index: Forecast and methodology, 2014–2019, Cisco, May 2015.
Exhibit E2
Cross-border data flows are surging and connecting more countries
SOURCE: TeleGeography, Global Internet Geography; McKinsey Global Institute analysis
Used cross-border bandwidth
2005
100% = 4.7 Terabits per second (Tbps)
NA
United States and Canada
LA
Latin America
AF
Africa
EU
Europe
ME
Middle East
OC
Oceania
AS
Asia
Regions
Bandwidth
Gigabits per second (Gbps) <50 50–100 100–500 500–1,000 1,000–5,000 5,000–20,000 >20,000
2014
100% = 211.3 Tbps
NOTE: Lines represent interregional bandwidth (e.g., between Europe and North America) but exclude intraregional cross-border bandwidth (e.g., connecting
European nations with one another).
45x larger
NA
EU
LA
ME
AF
AS
OC
NA
EU
LA
ME
AF
AS
OC
5McKinsey Global Institute Digital globalization: The new era of global flows
Tangible flows
of physical goods
Flows mainly between
advanced economies
Capital- and labor-
intensive flows
Transportation
infrastructure is
critical for flows
Multinational
companies
drive flows
Flows mainly of
monetized
transactions
Ideas diffuse slowly
across borders
Intangible flows of
data and information
Greater participation by
emerging economies
More knowledge-
intensive flows
Digital infrastructure
becomes equally
important
Growing role of
small enterprises
and individuals
More exchanges of
free content and
services
Instant global access
to information
Innovation flows
from advanced to
emerging economies
Innovation flows in
both directions
Globalization: Then vs. now
Exhibit E3
6 McKinsey Global Institute Executive summary
Digital platforms are key to this new era of globalization. Over the past two decades, the
largest corporations built their own digital platforms to manage suppliers, connect to
customers, and enable internal communication and data sharing for employees around
the world. But a diverse set of public Internet platforms has emerged to connect anyone,
anywhere. These include operating systems, social networks, digital media platforms,
e-commerce websites, and all kinds of online marketplaces. Their use of automation and
algorithms drives the marginal costs of adding new interactions practically to zero, allowing
the biggest platforms to support hundreds of millions of global users (Exhibit E4). Now users
can more easily see details on products, services, prices, and alternative choices. This
removes some information asymmetries so that markets function more efficiently, although it
can disrupt some intermediaries in the process.
Exhibit E4
The biggest online platforms have user bases on par with the populations of the world’s
biggest countries
Active users of online platforms vs. country population
Million
205
256
300
300
320
321
400
407
650
1,000
1,000
1,314
1,372
1,590
China
YouTube
India
Alibaba
Skype
Brazil
Amazon
Indonesia
United States
Countries2
Online platforms1
1 4Q15 or latest available.
2 2015 population.
SOURCE: Facebook; Twitter; Alibaba; Fortune; Statista; Population Reference Bureau; McKinsey Global Institute analysis
7McKinsey Global Institute Digital globalization: The new era of global flows
Approximately 12 percent of the global goods trade is conducted via international
e-commerce, with much of it driven by platforms such as Alibaba, Amazon, eBay, Flipkart,
and Rakuten. Beyond e-commerce, digital platforms for both traditional employment
and freelance assignments are beginning to create a more global labor market.6 Some
50 percent of the world’s traded services are already digitized.7
Digitization also enables instantaneous exchanges of virtual goods. E-books, apps, online
games, MP3 music files and streaming services, software, and cloud computing services
can all be transmitted to customers anywhere in the world there is an Internet connection.
Many major media websites are shifting from building national audiences to global ones; a
range of publications, including The Guardian, Vogue, BBC, and BuzzFeed, attract more
than half of their online traffic from foreign countries. By expanding its business model
from mailing DVDs to selling subscriptions for online streaming, Netflix has dramatically
broadened its international reach to more than 190 countries. While media, music, books,
and games represent the first wave of digital trade, 3D printing could eventually expand
digital commerce to many more product categories.
Finally, “digital wrappers” are digital add-ons that enable and raise the value of other types
of flows. Logistics firms, for example, use sensors, data, and software to track physical
shipments, reducing losses in transit and enabling more valuable merchandise to be
shipped and insured. Online user-generated reviews and ratings give many individuals
the comfort level needed to make cross-border transactions, whether they are buying a
consumer product on Amazon or booking a hotel room halfway around the world on Airbnb,
Agoda, or TripAdvisor.
DIGITIZATION IS MAKING GLOBAL FLOWS MORE INCLUSIVE
Globalization was once driven almost exclusively by governments, large multinational
corporations, and major financial institutions. Today artisans, entrepreneurs, app
developers, freelancers, small businesses, and even individuals can participate directly on
digital platforms with global reach.
SMEs can be micro-multinationals, and digital startups are born global
Small and medium-sized enterprises (SMEs) worldwide are using the “plug-and-play”
infrastructure of Internet platforms to put themselves in front of an enormous global
customer base and become exporters. Amazon, for instance, now hosts some two million
third-party sellers. In countries around the world, the share of SMEs that export is sharply
higher on eBay than among offline businesses of comparable size. PayPal enables cross-
border transactions by acting as an intermediary for SMEs and their customers. Participants
from emerging economies are senders or receivers in 68 percent of cross-border PayPal
transactions. Microenterprises and projects in need of capital can turn to platforms such as
Kickstarter, where nearly 3.3 million people representing nearly all countries made pledges
in 2014.
Facebook estimates that 50 million SMEs are on its platform, up from 25 million in 2013; on
average 30 percent of their fans are from other countries. To put this number in perspective,
consider that the World Bank estimated there were 125 million SMEs worldwide in 2010.
For small businesses in the developing world, digital platforms are a way to overcome
constraints in their local markets. The ability of SMEs to reach global audiences supports
economic growth everywhere.
6 A labor market that works: Connecting talent with opportunity in the digital age, McKinsey Global Institute,
June 2015.
7 Daniel Castro and Alan McQuinn, Cross-border data flows enable growth in all industries, Information
Technology and Innovation Foundation, February 2015.
12%
of the global goods
trade is
e-commerce
8 McKinsey Global Institute Executive summary
The increasing globalization of small businesses is starting to show up in national statistics.
It is most clearly seen in the United States, where the share of exports by large multinational
corporations dropped from 84 percent in 1977 to 50 percent in 2013. Among SMEs that
export, the smallest (those with fewer than 50 employees) are gaining share the fastest. An
analysis of export data for 16 OECD countries shows mixed evidence, with the SME share of
total exports growing in ten of the countries.8
Even new startups can form global connections and market to international customers
from their inception. We surveyed 271 startups worldwide through a partnership with
1776, a global incubator and venture fund. By working with 1776 and its Startup Federation
program, we were able to expand the reach of the survey to 19 countries. While these
startups represent a more tech-savvy cross-section than the broader universe of
entrepreneurs, the results show that even the smallest and youngest enterprises can
execute a global vision if their business model is built on digital technologies. A surprising
86 percent of survey respondents pointed to at least one cross-border activity. Almost two-
thirds have customers or users in other countries, and almost half reported sourcing talent
from other countries.
Individuals can participate directly in globalization, with significant
economic impact
Thanks to social media and other Internet platforms, individuals are forming their own cross-
border connections. We estimate that 914 million people around the world have at least
one international connection on social media, and 361 million participate in cross-border
e-commerce (Exhibit E5). These figures are growing rapidly. On Facebook, 50 percent of
users now have at least one international friend. This share is even higher—and growing
faster—among users in emerging economies.
The business and economic implications of individual participation are significant. Digital
platforms provide a huge built-in base of potential customers and effective ways to market
to them directly. As social media exposes consumers from around the world to what is
available, products can go viral on a scale that has never been seen before. In 2015, Adele’s
song “Hello” racked up 50 million views on YouTube in its first 48 hours, and her album 25
sold a record 3.38 million copies in the United States in its first week alone, more than any
other album in history. In 2012, Michelle Obama wore a dress from British online fashion
retailer ASOS in a photo that was retweeted 816,000 times and shared more than four
million times on Facebook; it instantly sold out.
Digital platforms offer individuals new ways to learn, collaborate, and acquire new skills—
and then to showcase their talents to potential employers. Some 44 million people around
the world find freelance work on Freelancer.com, Upwork, and other digital platforms; nearly
400 million have posted their professional profiles on LinkedIn. Individuals with creativity and
drive can propel themselves onto a global stage in ways that would have been unimaginable
in the pre-digital world. A number of previously unknown singers have been discovered after
posting videos on YouTube. The Weeknd, spotted on YouTube by Drake, dominated the
Billboard charts in 2015 and recently earned an Oscar nomination for best original song.
8 Some countries where SME share of exports declined were those suffering from a post-crisis credit crunch,
such as Portugal.
86%
of surveyed
startups report at
least one
cross-border
activity
9McKinsey Global Institute Digital globalization: The new era of global flows
GLOBAL FLOWS DRIVE ECONOMIC GROWTH, BUT COUNTRY PARTICIPATION
IS UNEVEN
In this report, we set out to develop more robust estimates of whether global flows
contribute to economic growth, using an expanded and improved data set and more
sophisticated statistical methods than in our last report on this topic, in 2014.9 We find even
stronger evidence that global flows increase GDP in the long term by raising productivity
and that data flows have as much impact as goods trade. But we also find that country
participation varies widely, and every type of flow remains dominated by a small group of
leading countries. There is enormous value at stake for lagging countries in catching up.
9 We first test for cointegration in the data and then use an error-correction econometric model. Our data cover
1995–2013 and 97 countries. See the technical appendix for a comprehensive discussion of the econometric
model, different statistical tests, and the variables and data used.
Exhibit E5
Individuals are participating in globalization, and 914 million have cross-border social media connections
SOURCE: Facebook; AliResearch; US Department of Commerce; OECD; W orld Bank; McKinsey Global Institute analysis
Students
studying abroad
5 million
Cross-border
e-commerce shoppers
361 million
International
travelers
429 million
People living outside
home country
244 million
Cross-border
online students
13 million
Cross-border
online workers
44 million
Social networking users with
at least one foreign connection
914 million
NOTE: Numbers adjusted to account for overlap between platforms and for individuals making multiple international trips in the same year.
REPEATS
in report
10 McKinsey Global Institute Executive summary
Global flows raised world GDP growth by 10 percent, or $7.8 trillion,
in 2014 alone
Our econometric analysis finds robust evidence that global flows of goods, FDI, people, and
data contribute structurally to economic growth by increasing productivity.10 It breaks new
ground by testing the impact of all types of flows together, both inflows and outflows, and
considering how countries are positioned in each web of flows.
Our results indicate that over a decade, global flows have raised world GDP by roughly
10 percent over what would have resulted in a world in without any flows. In 2014 alone, they
generated roughly $7.8 trillion in value. Flows of goods and FDI account for about half of this
impact, while data flows, the hallmark of 21st-century globalization, account for $2.8 trillion.
All types of global flows boost productivity growth, and data flows additionally appear to
increase the amount of labor and capital used in the economy.
We also examine how a country’s position in the network of flows affects the benefits it
receives. Countries in the center of the global network of goods trade benefit more than
those at the periphery. The network of cross-border data flows, by contrast, is still rather
new and less dense. The United States and Europe are at the center of the world’s digital
networks, facilitating links to other countries. But we find that countries at the periphery of
this digital network stand to gain even more than those at the center. For economies that
have been relatively disconnected, the arrival of new digital platforms and cross-border data
flows can be transformational.
Overall, our analysis underscores the value of connectedness—and the benefits are much
broader and more nuanced than a simple accounting of net exports can capture. Countries
that participate in global flows gain exposure to ideas, research, technologies, talent,
and best practices from around the world. The most connected economies can draw on
these flows to enhance their own competitiveness, innovation, and efficiency, positioning
themselves to take advantage of growth opportunities in global markets. However, countries
also need to have supporting institutions and policies in place to realize this potential.
Although more countries are participating, global flows remain concentrated
among a relatively small group of leading countries
Today global connections link a larger and more diverse range of countries than ever. For
the first time in history, emerging economies are counterparts on more than half of global
trade flows, and South-South trade between these countries is the fastest-growing type
of connection. The value of traded goods and services plus financial flows exceeded
80 percent of GDP for only 72 countries (mainly developed ones) in 1990; by 2014, that was
true for 121 countries. But while more countries are participating in global flows, their level of
participation varies widely.
10 We include only the FDI component of total financial flows, since those have been shown by other research
to be correlated with GDP growth. The impact of other forms of financial flows on growth is mixed. We do
not include service flows in our econometric analysis because they are highly correlated with FDI and with
goods trade.
We find strong evidence that global flows increase
GDP over the long term by raising productivity. Both
inflows and outflows matter for growth.
11McKinsey Global Institute Digital globalization: The new era of global flows
The MGI Connectedness Index offers a comprehensive look at how countries participate in
inflows and outflows of goods, services, finance, people, and data (Exhibit E6).11 Our index
takes into account the size of each flow for a country relative to its own GDP or population
(flow intensity) as well as its share of each total global flow. Combining these measures
avoids making large and diversified economies appear closed simply due to the extent of
economic activity taking place within their own borders.
Singapore, a small country that punches far above its weight in all types of global flows, tops
this year’s rankings. It is followed by the Netherlands (one of Europe’s main digital hubs),
the United States, Germany, Ireland, and the United Kingdom. China’s surge is particularly
noteworthy; it has climbed from 25th in our previous index to the No. 7 spot.
However, the world is still far from fully globalized. Advanced economies in general
remain more connected than developing countries, and the top countries have far
higher connectedness scores than the rest of the world Exhibit E7). All types of flows are
concentrated among a small set of countries. The top 15 countries in traded goods account
for 63 percent of the global total; that share is 62 percent in services and 79 percent in FDI.
We use statistical tests of convergence to see if the gaps between country participation in
global flows are closing over time. Our results indicate that lagging countries are catching up
to leading countries—but extremely slowly, given that the global flows of leading countries
continue to rise. At current trends, cutting the gap in half would take eight years in the
goods trade and 13 years in FDI flows. For data flows, we do not see any sign that laggards
are catching up to leaders, perhaps reflecting that digitization has a long way to go in all
countries and it is a relatively young phenomenon.
Lagging countries could realize tremendous growth potential by accelerating their
participation in well-targeted ways. We find that countries in the top quartile increased their
flow of goods relative to GDP at an average of 3 percent annually, for example, while goods
flows grew at only 1 percent for the bottom quartile. The top-quartile countries increased
FDI flows by 5 percent of GDP annually during this period, while those flows shrank by
8 percent annually for countries in the bottom quartile. If countries in the bottom three
quartiles had increased participation in flows at the same rate as the top quartile over the
past decade, global GDP would be an additional $10 trillion, or 13 percent, higher today.
In other words, limited participation in global flows by many countries had a real cost to
the world economy. For some individual countries, GDP would be more than 50 percent
higher today.
Countries have taken different routes to become more globally connected. Top-ranked
Singapore emerged decades ago as Southeast Asia’s global shipping hub. It subsequently
mapped out an explicit strategy to become a regional hub for finance and services by
attracting skilled international talent and establishing incentives and promotional efforts
to attract FDI. The Netherlands is a major hub for Europe’s data traffic as well as a port
for traded goods. Like Ireland, it has created tax and regulatory regimes to attract many
subsidiaries, headquarters, and holding companies for multinational corporations. In
contrast, the United States and Germany both follow a generalist model with strength
across all five flows. The United Kingdom also has broad participation across flows, with a
spike in cross-border service and financial flows, a reflection of London’s role as a global
financial hub.
11 Several other indexes measure the degree to which countries are connected to global activity, although they
use different data and weighting. These include the DHL Global Connectedness Index produced by Pankaj
Ghemawat and Steven A. Altman and globalization indexes from Ernst & Young, A. T. Kearney, and the Swiss
Economic Institute. See, for example, Pankaj Ghemawat and Steven A. Altman, Depth Index of Globalization
2013: And the big shift to emerging economies, IESE Business School, University of Navarra, 2013.
12 McKinsey Global Institute Executive summary
Exhibit E6
MGI Connectedness Index
Country connectedness index and overall flows data, 2014
Rank of participation by flow as measured by flow intensity and share of world total
1–10 11–25 26–50 >50Connectedness index rank 100+ <7070–99Flow intensity Rank Country Score Connectedness Index rank Flow value1 $ billion Flow intensity2 % of GDPGoods Services Finance People Data 1 Singapore 64.2 1 2 2 12 6 1,392 452 2 Netherlands 54.3 3 3 6 21 1 1,834 211 3 United States 52.7 7 7 3 1 7 6,832 39 4 Germany 51.9 2 4 8 3 2 3,798 99 5 Ireland 45.9 32 1 1 28 9 559 227 6 United Kingdom 40.8 13 5 5 6 3 2,336 79 7 China 34.2 4 16 4 82 38 6,480 63 8 France 30.1 11 8 9 7 4 2,262 80 9 Belgium 28.0 5 6 33 33 8 1,313 246 10 Saudi Arabia 22.6 20 28 27 2 53 790 106 11 United Arab Emirates 22.2 6 23 17 4 46 789 196 12 Switzerland 18.0 12 11 10 17 13 848 115 13 Canada 17.3 16 22 11 11 18 1,403 79 14 Russia 16.1 21 25 18 5 25 1,059 57 15 Spain 14.4 25 13 19 14 16 1,105 79 16 Korea 14.0 8 12 28 50 44 1,510 107 17 Italy 13.4 17 18 24 16 19 1,587 74 18 Sweden 13.0 29 14 22 31 5 572 100 19 Austria 11.7 26 17 31 20 12 470 108 20 Malaysia 11.6 9 19 25 26 43 610 187 21 Mexico 10.7 14 63 34 18 41 1,022 80 22 Thailand 10.7 10 15 36 44 64 605 162 23 Kuwait 10.6 37 46 13 13 75 306 153 24 Japan 10.5 15 20 12 81 20 2,498 54 25 Kazakhstan 10.0 48 73 41 8 57 176 83 26 Ukraine 9.8 38 39 87 10 34 133 101 27 Australia 9.7 30 34 21 15 33 825 57 28 Denmark 8.9 35 9 32 41 11 369 108 29 Jordan 8.8 73 50 75 9 83 50 138 30 India 8.5 24 10 35 58 70 1,316 64 32 Czech Republic 7.5 18 33 57 59 15 397 193 34 Poland 7.0 23 31 47 34 22 585 107 35 Hungary 6.8 22 30 26 62 17 287 209 36 Norway 6.0 36 24 20 46 24 458 92 37 Vietnam 5.7 19 54 45 103 61 350 188 39 Finland 5.5 46 27 23 70 10 390 144 40 Portugal 5.5 47 36 30 23 31 255 111 41 Turkey 5.1 28 40 53 38 29 521 65 43 Israel 4.9 51 32 49 24 56 248 82 44 Brazil 4.5 41 38 14 125 30 869 37 45 Chile 4.1 45 58 16 102 27 239 92 47 Greece 4.1 60 29 54 35 42 160 67 48 New Zealand 3.9 67 48 61 25 51 130 63 51 Indonesia 3.4 31 49 38 106 76 504 57 53 South Africa 3.3 34 57 52 64 80 277 79 54 Philippines 3.2 54 41 44 52 67 230 81 64 Morocco 2.6 58 43 74 56 65 104 97 73 Egypt 2.2 68 42 69 73 71 158 55 83 Nigeria 1.9 55 76 48 128 98 268 47 86 Peru 1.8 62 88 51 104 49 122 60 118 Kenya 1.3 100 84 127 119 91 35 58 SOURCE: McKinsey Global Institute analysis 1 Flows value represents total goods, services, and financial inflows and outflows. 2 Flow intensity represents the total value of goods, services, and financial flows as a share of the country’s GDP. 13McKinsey Global Institute Digital globalization: The new era of global flows Exhibit E7 A small group of leading countries are much more connected than the rest of the world 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 0 805030 40 7060 1402010 Connectedness score, 2014 United Kingdom Singapore Sweden United Arab Emirates Mexico Russia Saudi Arabia Turkey Australia Belgium Switzerland Spain Indonesia Ireland India Norway Austria Japan South Korea France China Canada Italy United States Per capita GDP, 2014 $ thousand, purchasing power parity, current international dollar Netherlands Germany Portugal Kuwait Denmark Ukraine QatarFinland Czech RepublicGreece Emerging Developed Correlation coefficient (r) = 0.54 Size of circle represents $ value of flows in 2014 SOURCE: IMF; McKinsey Global Institute analysis Vietnam Brazil Thailand Malaysia 14 McKinsey Global Institute Executive summary Although our report mainly assesses the global connectedness of countries, nation-states are not the only lens through which to observe globalization. Cities, regions within countries, and broader blocs of countries are connecting with the global economy in myriad ways and to varying degrees. For instance, our previous report found that the world had only eight truly “global cities” with strong connections in at least four of the five major flows: New York, London, Tokyo, Los Angeles, San Francisco, Singapore, Hong Kong, and Dubai. This year Tokyo drops off the list due to a decline in goods trade, while Shanghai takes its place. Within countries there can be very different patterns of globalization. In the United Kingdom and Germany, for instance, the variation across regions is modest. China, by contrast, has a handful of highly connected coastal provinces and largely unconnected inland provinces. Some highly connected states and provinces rank as economic powerhouses in their own right: China’s booming province of Guangdong would rank sixth globally in terms of goods flows, while California would rank fourth in the world for people flows. We also look at the patterns of trade among neighbors and trading blocs. Europe is the most integrated region; more than 60 percent of its trade in goods is intraregional. But the corresponding shares are sharply lower in Africa, Latin America, and South Asia. This indicates a significant opportunity for developing countries to increase their participation in flows by trading with their neighbors (Exhibit E8). Exhibit E8 35 65 While much of the world’s trade in goods is long distance, roughly half or more of other global flows move within the same region Distribution of flows between intraregional (short haul) vs. interregional (long haul), 20141 % of world flow SOURCE: UNCTAD; UN W orld Tourism Organization; TeleGeography, Global Internet Geography; IMF; McKinsey Global Institute analysis 1 For goods, services, FDI, and travelers we have divided the world into 10 regions; for data flows we have used TeleGeography’s six regions. 2 Distribution of services flows for 2014 estimated based on 2011 data; 2013 bilateral traveler data used for people flows. NOTE: Numbers may not sum due to rounding. 36 64 Long haul (interregional) Short haul (intraregional) 53 47 54 46 67 33 Services2 FDI Data Goods People2 15McKinsey Global Institute Digital globalization: The new era of global flows COMPANIES MAY NEED TO REINVENT THEMSELVES TO WIN IN A DIGITAL GLOBAL MARKETPLACE The new era of digital globalization offers unprecedented opportunities for companies to achieve both global scale and efficiency, but it also calls for reevaluating existing strategies, business models, and operations. Business leaders in all industries should consider the following issues: � Do your footprint and organizational structure make sense in a more digital world? As companies expanded across borders, many encountered a “globalization penalty” due to the costs of rising complexity.12 But now digital technologies allow companies to globalize in a leaner way. Digital tools for remote collaboration and instant communication mean that it is possible to centralize some global functions, such as back-office operations or R&D; to create virtual global teams that span borders; or even to forgo having one global headquarters location. Digitization is also enabling business models that are less capital-intensive. Rather than establishing a large physical presence in many countries, some companies focus local offices on sales and marketing only. Those that deliver digital goods and services can enter new international markets without establishing a physical presence at all. � Should you offer one brand and one product line around the world, or customize for local markets? In some industries, product tailoring is driven by local regulatory requirements or language differences. In others, companies that sell into many global markets have expanded their product portfolios to appeal to local consumer preferences and price points. But others take a different approach: offering products that are the same everywhere in the world. Apple, for instance, offers just three models of its iPhone and iPad, all with consistent design and branding wherever they are sold. Facebook, Uber, and Airbnb have simply scaled up their digital platforms in country after country with limited customization. Many global automakers are attempting to strike a balance by whittling down the number of platforms used across their international manufacturing operations (that is, using fewer underlying designs that can be customized by swapping certain components to create differentiated models). The media and consumer technology industries are shifting to simultaneous global product launches, since consumers around the world can see instantaneously what is offered in other countries. � Do you have the right suppliers and customer channels? Digital tools can orchestrate a multitude of vendors around the globe with greater precision and efficiency. But even as technology enables more complex global value chains, the importance of different factor costs is shifting. Until relatively recently, many companies were willing to fully outsource manufacturing and other functions to locations with low-cost labor. Today many are reevaluating those decisions and giving greater weight to energy prices, distance to market, infrastructure, ease of doing business, and risk. According to a recent UPS survey, approximately one-third of high-tech companies are moving manufacturing or assembly closer to end-user markets; this number is up by 25 percentage points from 2010.13 As China’s labor costs rise and the country moves into higher-value-added industries, more of the world’s manufacturing business is up for grabs. Businesses will have to consider whether their suppliers and customer channels should change. 12 Martin Dewhurst, Jonathan Harris, and Suzanne Heywood, “Understanding your ‘globalization penalty,’” McKinsey Quarterly, July 2011. 13 Change in the (supply) chain, United Parcel Service, 2015. 16 McKinsey Global Institute Executive summary � Do you have the right assets to compete digitally and globally? Building digital platforms, online customer relationships, and data centers may be critical for a growing range of companies, far beyond the Internet giants. GE, for example, is transforming its core manufacturing capabilities to establish itself as a leader in Internet of Things technology. Businesses in all industries need to take a fresh look at their assets, including customer relationships and market data, and consider whether there are new ways to monetize them. Alibaba has a vast pool of transactional data on the vendors that operate on its platform, and it has built on it to move into new areas such as mobile payments and small business financing. The insurance industry could similarly harness its sophisticated data pools on different forms of risk to create new products and services. � Are you ready for a new era of digitally accelerated global competition? Competition is intensifying and product cycles are shortening due to the confluence of three trends. First, emerging-market giants are going global. Many of them are aggressive, deep-pocketed, and able to operate with different time horizons and financial targets. By 2025, MGI estimates that companies headquartered in emerging markets will make up 45 percent of the global Fortune 500, up from 26 percent today.14 Second, tech companies are expanding into new industries. Some of the truly disruptive players are siphoning value out of industries and giving it away for free to consumers as a way to build their positions. Finally, the largest Internet platforms allow millions of SMEs and startups to go head-to-head with incumbents. These new forms of competition have unleashed pricing pressures and industry disruptions. The Internet and international competition have cut into the window of exclusivity companies once enjoyed on new products and services; “copycat” versions can be launched in new markets even before the originator has time to scale up. It is more important than ever to stay alert to new competitive threats. � Are you prepared for new risks? As the world grows more dependent on information systems, the private sector is also becoming more vulnerable to cyberattacks. It is difficult to stay ahead of increasingly sophisticated hackers, but companies can prioritize their information assets, test continuously, and work with frontline employees to emphasize basic protective measures. If a breach does occur, a decisive and forthright response from marketing, public affairs, and customer service functions can be critical to restoring customer trust.15 Maintaining data security has to be a top priority for CEOs in every industry. POLICY MAKERS FACE A NEW WORLD OF CHALLENGES Countries cannot afford to shut themselves off from global flows, given the value at stake in raising productivity and long-term GDP growth. Pursuing this opportunity requires a new policy agenda that includes the issues outlined below. � Thinking strategically about the role your country can play. Policy makers should carefully consider how to build on their country’s comparative advantages. Many countries are trying to develop the next Silicon Valley, but innovation is notoriously difficult to orchestrate. Meanwhile, developing nations may face a shrinking opportunity to become low-cost manufacturers for the world as automation advances. But other opportunities exist. Some countries can build on their geographic proximity to major consumer markets, as Mexico and Eastern Europe have done. Others may develop a successful niche as global transit hubs, as Dubai has done in transportation and trade flows. Other countries have targeted a particular flow or industry to cultivate, 14 See Playing to win: The new global competition for corporate profits, McKinsey Global Institute, September 2015, and Urban world: The shifting global business landscape, McKinsey Global Institute, October 2013. 15 Risk and responsibility in a hyperconnected world: Implications for enterprises, McKinsey & Company and the World Economic Forum, January 2014. 17McKinsey Global Institute Digital globalization: The new era of global flows perhaps building on pools of talent within their borders (as India has done with business process outsourcing). � Addressing policy and administrative barriers that hinder global flows. Pursuing bilateral and multilateral trade partnerships is the cornerstone of a more open approach. Another important step is removing import tariffs, quotas, and subsidies for national industries, all of which can introduce distortions. Other types of legal and administrative barriers also have to be dismantled to make the most of global flows; these may include limitations on foreign business ownership and investment, import licensing, regulatory requirements that deviate from international norms, and limits on immigration. The Association of South East Asian Nations (ASEAN), for instance, has largely eliminated import tariffs among its ten member states, but its ongoing effort to build a seamless trading bloc involves harmonizing product standards, certification procedures, customs requirements, and cross-border regulations covering traded services and the movement of labor.16 � Addressing dislocations. Even though their net global effect is ultimately positive, global flows can cause job losses and displacement in the short run. Governments have to consider these trade-offs and open to global flows at a pace their economies and societies can absorb. Few countries have adequately supported the workers and communities affected by exposure to international competition and disruptive business models. But these workers will need a clearer path to new roles—and the societal cost of neglecting this issue grows over time. It will take a much more proactive response to ensure that labor markets and training systems can deal with rapid change. � Investing in human capital. The Internet can promote inclusiveness, but only if education and training systems provide language fluency, basic digital literacy, and other skills so that individuals can take advantage of the opportunities. Investment in human capital development will be a critical determinant of which nations come out on top. � Building the necessary infrastructure and closing the digital divide. Even in a more digital world, roads, ports, airports, and rail remain vital as the conduits of trade and mobility. But today any list of infrastructure priorities also has to include universal, affordable Internet access. At the end of 2015, 57 percent of the world’s population, or four billion people, remained offline, and only 15 percent had access to broadband.17 The value of connecting these people is significant. Our own econometric analysis shows that countries with higher Internet penetration reap up to 25 percent more benefit from cross-border data flows than those with limited Internet penetration. � Creating a strong business and institutional environment. A recent World Bank report finds that in many developing countries, the economic benefits of digital technologies have been limited by a lack of strong fundamentals such as education and good governance.18 To capture the full growth potential of digital globalization, countries need to cultivate a healthy business environment that nurtures startups, allows inefficient firms to exit, ensures a level playing field, and establishes a solid legal framework for intellectual property and property rights. � Protecting data privacy while maintaining an open Internet. Many countries have enacted or are considering limitations on what kind of data can be transmitted across borders; this may include requirements that companies use servers physically located 16 Southeast Asia at the crossroads: Three paths to prosperity, McKinsey Global Institute, November 2014. 17 The state of broadband 2015, International Telecommunication Union and UNESCO, September 2015. For more on policy approaches to addressing this issue, see Offline and falling behind: Barriers to Internet adoption, McKinsey Technology, Media and Telecom Practice, September 2014. 18 World development report 2016: Digital dividends, World Bank, January 2016. 18 McKinsey Global Institute Executive summary within their borders to process and store data generated there. As we went to press, for example, the future of the “safe harbor” agreement governing data transfers between the European Union and the United States remained uncertain. Legitimate privacy concerns need to be addressed through thoughtful frameworks, but data localization and fragmented regulation may have real economic costs.19 � Making cybersecurity a top priority. One study has estimated that cybercrime costs the global economy some $400 billion in annual losses through consumer data breaches, financial crimes, market manipulation, and theft of intellectual property.20 Hackers may also pose public safety and even national security risks. While companies are often at the forefront of ensuring cybersecurity, governments can invest in research, share information, model good security practices, and craft thoughtful rules. Governments will need to work closely with their global counterparts and with the business community to stay on top of new threats and share technology solutions. Regulators may need to mandate standards for securing consumer data, and public agencies need to safeguard their own assets. ••• Many of the challenges associated with digitizing economic activity are now playing out on a global scale. Even measuring digital globalization in statistics has become a more complex undertaking, since much of the value being generated winds up as consumer surplus. Our analysis provides strong evidence of the economic value of openness—and it shows that both inflows and outflows matter, as they expose an economy to ideas, research, technologies, talent, and best practices from around the world. For countries that have been slow to participate, the opportunities for catch-up growth are too substantial to ignore. 19 Matthias Bauer et al., The costs of data localization: Friendly fire on economic recovery, ECIPE occasional paper number 3/2014, May 2014, analyzes recently proposed or enacted data localization rules in seven economies. It found that these rules would lower GDP in all seven cases, with Vietnam (-1.7 percent), China (-1.1 percent), and Indonesia (-0.5 percent) poised for the largest losses. 20 Net losses: Estimating the global cost of cybercrime, Center for Strategic and International Studies and McAfee, June 2014. 19McKinsey Global Institute Digital globalization: The new era of global flows © Getty Images 20 McKinsey Global Institute Executive summary SOURCE: UNCTAD; McKinsey Global Institute analysis 1 Estimated from 2011 bilateral services flows data and 2014 services trade data from UNCTAD. NOTE: For cross-border data flows, see Exhibit E2. % of global GDP 0.01–0.05 0.05–0.10 0.10–0.25 0.25–0.50 >0.50
NA
United States.
Canada, and Mexico
LA
Latin
America
AF
Africa
EE
Eastern Europe
and Central Asia
ME
Middle
East
OA
Other
Asia
CH
China
region
WE
Western
Europe
AU
Austral-
asia
NE
Northeast
Asia
% of global GDP
0.02–0.05 0.05–0.10 0.10–0.25 0.25–0.50 0.50–1.00 >1.00
NA
United States
and Canada
LA
Latin
America
AF
Africa
EE
Eastern Europe
and Central Asia
ME
Middle
East
OA
Other
Asia
CH
China
region
WE
Western
Europe
AU
Australasia
NE
Northeast
Asia
GOODS FLOWS
SERVICES FLOWS
1980
100% = $1.8 trillion (18.6% of GDP)
NA WE
LA
ME
AF
CH
AU
EE
NE
OA
10.5x larger2014
100% = $19 trillion (24.6% of GDP)
NA WE
LA
ME
AF
CH
AU
EE
NE
OA
2002
100% = $1.6 trillion (4.9% of GDP)
NA WE
LA
ME
AF
CH
AU
EE
NE
OA
3.1x larger20141
100% = $4.9 trillion (6.4% of GDP)
NA WE
LA
ME
AF
CH
AU
EE
NE
OA
21McKinsey Global Institute Digital globalization: The new era of global flows
SOURCE: IMF CDIS; UN W orld Tourism Organization; McKinsey Global Institute analysis
1 Estimated from bilateral FDI stock data.
NOTE: For cross-border data flows, see Exhibit E2.
% of global GDP
0.02–0.05 0.05–0.10 0.10–0.25 0.25–0.50 0.50–1.00
FINANCIAL FLOWS (FDI)1
PEOPLE FLOWS
NA
United States,
Canada, and Mexico
LA
Latin
America
AF
Africa
EE
Eastern Europe
and Central Asia
ME
Middle
East
OA
Other
Asia
CH
China
region
WE
Western
Europe
AU
Austral-
asia
NE
Northeast
Asia
Million cross-border travelers
<1 1–5 5–10 10–50 >50
NA
United States
and Canada
LA
Latin
America
AF
Africa
EE
Eastern Europe
and Central Asia
ME
Middle
East
OA
Other
Asia
CH
China
region
WE
Western
Europe
AU
Australasia
NE
Northeast
Asia
2002
100% = $0.7 trillion (2.1% of GDP)
NA WE
LA
ME
AF
CH
AU
EE
NE
OA
2.3x larger2014
100% = $1.65 trillion (2.1% of GDP)
NA WE
LA
ME
AF
CH
AU
EE
NE
OA
2002
100% = 650 million
NA WE
LA
ME
AF
CH
AU
EE
NE
OA
1.6x larger2013
100% = 1.03 billion
NA WE
LA
ME
AF
CH
AU
EE
NE
OA
© Getty Images
For decades, the movement of traded goods, services, and finance defined our image
of globalization and deepened the connections between nations. Today, however, those
traditional flows have lost their momentum. At least part of this shift appears to be structural
rather than a temporary cyclical dip. But this does not mean that globalization has moved
into reverse. Enormous streams of data are transmitted across borders every minute, and
they are growing exponentially in both volume and variety. Today globalization is being
accelerated and redefined by flows of data that embody ideas, information, and innovation.
Increasingly, the World Wide Web provides the ties that bind the global economy together.
Consider that some 50 percent of the world’s traded services are already digitized.21
Approximately 12 percent of the global goods trade is conducted via international
e-commerce. Cross-border Skype calls equal 46 percent of the volume of traditional
international calls. Across 18 countries analyzed by eBay, anywhere from 88 to 100 percent
of the SMEs that use its platform are exporters.
By lowering the costs of communication and transactions, digitization opens new
possibilities for conducting business across borders. As digital platforms grow in scale and
sophistication, they are creating more efficient and transparent global markets in which
far-flung buyers and sellers find each other with a few clicks. They provide businesses
with enormous built-in customer bases and effective ways to connect with them—and
they enable even microenterprises to participate directly in global flows. Digital flows are
also shifting globalization into a faster gear as information ricochets around the world and
collaboration spans time zones. Taken together, these shifts create economic value by
increasing innovation, competition, and productivity.
Many of the challenges associated with digitizing national economies are now playing out on
a global scale. Industries are being disrupted by new entrants, value chains are re-forming,
and profit pools are shifting. Much of the value of digitization is going into unpriced benefits
for consumers that are not captured in official GDP statistics. While the digital world may
be more inclusive and closely connected, it is not truly “flat”—and digitization tends to
accentuate disparities.
Digitization changes the economics of globalization in several ways, as we discuss in this
chapter and the ones that follow. First, in contrast to the last era of globalization, we find
that countries on the periphery of the network of global data flows benefit even more than
the digital content producers at the center. In addition, the types of companies involved
are different: instead of waiting for the benefits of globalization to trickle down from large
corporations, SMEs can become micro-multinationals in their own right, and startups can
be “born global.” Finally, digitization is fueling competition as it enables innovative business
models and allows companies to scale up quickly. This chapter brings the new and more
digital version of globalization into focus.
21 Daniel Castro and Alan McQuinn, Cross-border data flows enable growth in all industries, Information
Technology and Innovation Foundation, February 2015.
45X
increase in used
cross-border
bandwidth since
2005
1. A NEW ERA OF
DIGITAL GLOBALIZATION
24 McKinsey Global Institute 1. A new era of digital globalization
GROWTH IN TRADITIONAL FLOWS OF GOODS, SERVICES, AND FINANCE
HAS FLATTENED
In the 20th-century version of globalization, the world built deeper and more intricate ties
as the goods trade and cross-border finance grew in volume and scope. But both types of
flows took sharp tumbles during the financial crisis and the Great Recession. Since then,
global trade bounced back but is now flattening—and capital flows remain at a fraction of
the heights reached during the bubble years. These traditional types of flows still form an
important part of the global economy, but flows of data and information are providing the
real momentum.
Global trade in goods has slowed dramatically since 2008, reflecting
structural shifts
Between 1985 and 2007, the world’s trade in goods grew roughly twice as fast as global
GDP. This reflects major multinationals expanding their supply chains and establishing
new bases of production to tap into the enormous pools of low-cost labor in emerging
economies. Global trade in goods rose from 13.8 percent of world GDP in 1985 ($2 trillion)
to 26.6 percent of GDP ($16 trillion) on the eve of the Great Recession. Since its post-crisis
rebound, however, growth in goods trade has flattened—and it has even receded when
measured relative to GDP (Exhibit 1).
Much of the growth in goods trade since 2000—and much of its subsequent deceleration—
is related to commodity prices. As emerging economies rapidly urbanized and
industrialized, their appetite for raw materials such as steel, copper, and agricultural goods
boosted trade volumes and sent commodity prices soaring to new heights. From 2000
to 2011, the price of many commodities doubled or even tripled. But today the picture is
Exhibit 1
24.6
26.1
21.7
26.6
20.1
13.8
12
14
16
18
20
22
24
26
28
05 20142000 10
-2.0
Global goods trade, 1980–2014
% of GDP
1985 90 95
After decades of steady growth relative to GDP, trade in goods has been declining since its post-recession rebound
SOURCE: UNCTAD; McKinsey Global Institute analysis
2.0
Global trade
$ trillion
GDP growth
rate
%
1995–2005 2005–141985–95
10.63.5 5.2 6.5 15.4 19.0
7.5 6.710.1
Global flows 2
Report
mc 0225
25McKinsey Global Institute Digital globalization: The new era of global flows
remarkably different. Prices have declined sharply over the past few years, and the volume
of commodities being traded has also flattened (Exhibit 2). From June to December 2014,
the price of Brent crude fell from $112 a barrel to $62, and the price of copper has fallen by
half since its peak in 2011.22 We calculate that this slowdown in commodities accounts for
nearly three-quarters of the decline in goods trade as a share of global GDP.
Yet there is more behind the slowdown in global goods trade than a commodities cycle.
Trade in manufactured goods has also been flat to declining for both finished goods and
intermediate inputs. Global container shipping volumes grew by 7.8 percent from 2000 to
2005, but from 2011 to 2014, growth was markedly slower, at only 2.8 percent.23
Multiple cyclical factors have sapped momentum in the trade of manufactured goods.
Many of the world’s major economies—notably China, Europe, and Japan—have been
experiencing slowdowns. China, for example, posted almost 18 percent annual growth in
both imports and exports from 2000 to 2011. But since then its export growth has slowed to
4.6 percent, and imports have actually shrunk.
However, there may be structural reasons in global manufacturing that explain decelerating
growth in traded goods. Our analysis find that global consumption growth is outpacing
trade growth for some types of finished goods, such as automobiles, pharmaceuticals,
fertilizers, and plastic and rubber goods. This indicates that more production is happening
in the countries where the good is consumed. This may reflect the “reshoring” of some
manufacturing to advanced economies as well as increasing consumption in emerging
markets where these goods are produced.
22 Oil prices from US Energy Information Administration data, January 2015. Our analysis extends through 2014,
but commodity prices have continued their sharp decline since then.
23 Data from IHS.
Exhibit 2
The commodities slump partly explains the loss of momentum in goods trade,
but finished and intermediate goods have declined as well
SOURCE: IHS; UNCTAD; McKinsey Global Institute analysis
Value of goods trade, 2002–14
% of world GDP
0
1
2
3
4
5
6
7
8
9
10
11
12
2002
-1.1
201408
0
1
2
3
4
5
6
7
8
9
10
11
12
-0.4
2002 08 2014
0
1
2
3
4
5
6
7
8
9
10
11
12
2014
-0.5
082002
Processed and raw materials Intermediate goods Finished goods
Trade value
$ trillion
4.6 6.1 8.5
26 McKinsey Global Institute 1. A new era of digital globalization
For intermediate goods, declines in trade are more widespread across product categories,
including chemicals, paper, textile fabrics, and communications and electrical equipment.
In fact, the value of trade declined in roughly half of the categories of intermediate
goods between 2011 and 2014 (Exhibit 3). This could indicate that global value chains
are shortening.
The current slowing of trade growth may or may not reverse in the years ahead. The
development of 3D printing has not yet had a clear effect on global trade, but if this
technology is widely adopted by global manufacturers, it could reduce global trade volumes
Exhibit 3
SOURCE: IHS; McKinsey Global Institute analysis
Chemicals
Telephones, microphones, etc.
-16
-72
Pharmaceutical inputs
Fertilizers -11
-10
Vehicle parts
Total increasing
-13Communications equipment
38
Other declining categories
38
Electrical equipment
Total declining
201
Textile fabrics
-67
16
Paper
Other increasing categories
-212
Office machines
75
33
Aircraft parts
-7
Steel products
-8
-8
Trade has declined in half of intermediate goods categories, reflecting shorter global value chains
47%
of categories
have posted
trade declines
since 2011
53%
of categories
have posted
trade increases
since 2011 but
grew more
slowly than
GDP
3.5
13.0
3.5
2.2
13.0
2.8
1.2
1.8
1.2
2.0
0.6
2.3
22.0
Change in trade in categories of intermediate goods products, 2011–14
$ billion
% of intermediate
goods trade, 2014
Declining
categories
Increasing
categories
31.0
NOTE: Numbers may not sum due to rounding.
27McKinsey Global Institute Digital globalization: The new era of global flows
as more products are “printed” where they are consumed. There are already examples
of this at work. Consider GE Aviation, which is beginning to use 3D printing to produce
fuel nozzles for its new Leap engine. A fuel nozzle made the traditional way consists of 20
different components, with a supply chain that spans countries. But 3D printing allows the
company to produce best-quality nozzles in one piece, at one location, eliminating the need
to ship intermediate parts across borders.
Examining a wide range of both R&D-intensive and labor-intensive products, we find
significant potential to transform how—and where—many categories of goods are
produced with 3D printing in the years ahead.24 The applications are particularly relevant
for electronics, vehicle parts, other transportation equipment, machinery and electrical
equipment, medical instruments, and apparel.
Cross-border financial flows have fallen sharply since 2008 and show no sign
of recovery
Cross-border capital flows—which include lending, foreign direct investment, and
purchases of equities and bonds—link national financial markets, connecting borrowers
and savers from different countries. For 25 years prior to the 2008 financial crisis, these
flows grew faster than global GDP, rising from $0.5 trillion in 1980 to $11.9 trillion in 2007
(Exhibit 4).
24 For more on 3D printing technology and its economic potential, see Disruptive technologies: Advances that
will transform life, business, and the global economy, McKinsey Global Institute, May 2013.
Exhibit 4
SOURCE: IMF Balance of Payments; Economist Intelligence Unit; Bank for International Settlements; Institute of International Finance; McKinsey Global
Institute analysis
1 Includes foreign direct investment, purchases of foreign bonds and equities, and cross-border loans and deposits.
2 Includes trade credits, loans, currency, and deposits.
NOTE: Numbers may not sum due to rounding.
Cross-border lending accounts for 70 percent of the drop in global financial flows,
reflecting new banking regulation
6.8
20.7
12.2
4.5
4.1
0
2
4
6
8
10
12
14
16
18
20
22
1980 200720001990 2014E
-14 p.p.
5.7
0.9
2.6
1.7
2.7
1.6
0.9
1.0
2007
11.9
FDI
Equity
5.2
-6.7
Loans2
Bonds
2014
2
-7
-6
-23
Compound
annual
growth rate,
2007–14
%
Global cross-border capital inflows-to-GDP ratio1
%
Global capital inflows by type
$ trillion
28 McKinsey Global Institute 1. A new era of digital globalization
But since their peak in 2007, financial flows have contracted sharply, dropping from
21 percent of global GDP in 2007 to just 7 percent in 2014. Much of the decline is in cross-
border lending. Facing new regulations on capital and liquidity, as well as pressures from
shareholders and regulators to reduce risk, many banks in advanced economies are
winnowing down the geographies and business lines in which they operate. From early
2007 through the end of 2012, commercial banks sold off more than $722 billion in assets
and operations, with foreign operations accounting for almost half of this total.25 There is no
sign of a reversal in this trend, and the sharp decline could indicate a reversion to a longer-
term trend prior to the credit bubble years. Overall, the decline in cross-border lending
explains 72 percent of the total drop in cross-border financial flows since 2007.
Beyond the retrenchment in cross-border lending, international investment flows in bonds,
equities, and FDI are also flat or down. Cross-border bond and FDI flows have declined
41 percent and 35 percent, respectively, in absolute terms between the end of 2007 and the
end of 2014. Cross-border equity flows are essentially flat in value but have also declined
relative to global GDP. Preliminary data for 2015 show that global financial flows declined
further across a broad range of developing countries.26
The only financial flows that have continued to grow since the Great Recession are
remittances sent from global migrants to their home countries. These have grown 7 percent
annually over the past five years and are now worth $583 billion annually. Although steady
in nature, remittances are significantly smaller than equity flows (which totaled $1 trillion in
2014) and bond flows ($1.6 trillion in 2014). Growth in remittances reflects the increasing
flows of migrants and other people flows (see Box 1, “People on the move”).
Global service trade continues to grow, albeit slowly
Global trade in services is a much smaller flow than trade in goods. It has grown slowly but
steadily over the years, rising from some $400 billion in 1985 to approximately $5 trillion
in 2014, or from 3.4 percent to 6.3 percent of global GDP. Compared with the $19 trillion
goods trade, global trade in services remains small. Its compound annual growth rate of
8.8 percent since 1985 has outpaced global GDP growth over that period.
In the past, trade in services often involved people traveling around the world to deliver
expertise, but today financial services, IT support, R&D, engineering and design, and
many other services can be delivered digitally. Emerging economies such as Costa Rica,
India, Morocco, the Philippines, and South Africa, for example, have relied on technology
to build flourishing business process outsourcing industries that offer call center and
technical support services to global clients. Trade in digitally deliverable services has more
than doubled over the past decade, reaching $2.4 trillion in 2014. This amounts to almost
50 percent of total services exports. Advanced economies accounted for 81 percent of total
digitally deliverable service exports in 2014. India and the Philippines were the only emerging
economies ranking in the top ten net exporters of such services.
In the years ahead, the continued expansion of digital technologies, cross-border Internet
connections, and global online marketplaces for freelance services could potentially
increase traded services. Still, compared to the value of global goods trade, the global
services trade is likely to remain a far smaller cousin.
25 Financial globalization: Retreat or reset? McKinsey Global Institute, March 2013.
26 Institute of International Finance data.
7%
financial flows as a
share of world GDP
in 2014, down from
21% in 2007
29McKinsey Global Institute Digital globalization: The new era of global flows
Box 1. People on the move
People are more mobile than ever—and digital
technologies may partly facilitate this trend. We find that
all types of people flows across borders are growing
faster than the global population (Exhibit 5). Roughly a
quarter of a billion people, or 3.4 percent of the world’s
population, lived outside the country of their birth in 2013,
compared with 120 million, or 2.7 percent of the global
population, in 1980. People can now use digital platforms
to find work abroad and then stay closely connected with
friends and family back home through voice over Internet
protocol (VoIP) or Skype calling, instant messaging,
and social media. New platforms can even help with
logistics such as managing foreign bank accounts and
remittances; TransferWise, for instance, offers users
instant international monetary transfers without hefty
currency conversion fees.
Other people have been forced from their homelands by
conflict. After a decade of slight decline, the number of
refugees worldwide jumped from 16.7 million in 2013 to
19.5 million in 2014—a spike that worsened in 2015 with
the escalation of the Syrian refugee crisis.1 Many of these
1 UNHCR global trends report: World at war, UN High Commissioner
on Refugees, June 2015. Note that this number does not include
the 38.2 million people who are internally displaced by war and
persecution, nor the 1.8 million people who are awaiting the
outcome of asylum claims. Both of these numbers are up sharply
from 2013.
recent refugees have been relying on real-time social
media updates to guide their journey (for more on this, see
Chapter 2).
The number of international tourist arrivals hit 1.1 billion in
2014, continuing a trend of steady growth.2 As incomes
rise in emerging economies, the citizens of these
countries are eager to experience in person the world they
have seen online. Recent years have brought a huge influx
of Chinese visitors to destinations ranging from Australia
to the United States to Europe. Digital platforms enable
these flows: online travel sites make it easier than ever for
users to compare and book airfares, while sites such as
Airbnb help them find the exact accommodations that
suit their needs. Having guidance, user reviews, and GPS
mapping at their fingertips has given travelers the ability to
navigate unfamiliar destinations with greater confidence.
Additionally, OECD statistics show that some 4.5 million
international students traveled abroad to study in 2012.
The robust growth rate in this number may be a hopeful
sign that a new generation is embracing the opportunity
to become global citizens in a more mobile world.
2 UN World Tourism Organization statistics. Note that the “people
flows” metric in the MGI Connectedness Index adjusts this number
down to account for individuals making multiple trips within a
given year.
Exhibit 5
50
100
150
200
250
300
350
400
450
10 2014059590851980 2000
All types of people flows are outpacing global population growth
Students1
Refugees
Travelers
World
population
Migrants1
Major change
1 Latest data available are from 2012 for students and 2013 for migrants; 2012–13 growth was used for linear extrapolation to 2014.
SOURCE: OECD; W orld Bank; UN W orld Tourism Organization; UN High Commissioner on Refugees; UN Population Division; McKinsey Global Institute
analysis
1980–90 1990–2000 2000–10 2010–14 1980–2014
Compound annual
growth rate (%)
People flows
across borders
Index:
100 = 1980
1.67.5 -3.5 -1.4Refugees 8.0
4.41.7 8.7 3.0Students 3.8
1.51.8 1.4 1.2World population 1.2
4.24.7 4.6 3.8Travelers 4.9
1.70.2 2.1 2.9Migrants 4.4
Box 1
30 McKinsey Global Institute 1. A new era of digital globalization
DIGITIZATION IS USHERING IN A NEW ERA OF GLOBALIZATION
As of the end of 2015, some 3.2 billion people around the world—accounting for
43.4 percent of the global population—were online.27 The expansion of the Internet,
combined with the introduction of digital platforms and other types of digital tools, has
opened a new chapter in the story of globalization.
Cross-border data flows are the hallmarks of 21st-century globalization. Not only do they
transmit valuable streams of information and ideas in their own right, but they also enable
other flows of goods, services, finance, and people. Virtually every type of cross-border
transaction now has a digital component. Container ships still move products to markets
around the world, but now customers order them on digital platforms, track their movement
using RFID codes, and pay for them via digital transactions. Massive online platforms such
as Alibaba, Amazon, eBay, and Facebook link businesses and customers anywhere in the
world. By reducing the cost of transactions and allowing digital goods, services, and capital
to change hands instantly, digitization is creating a more hyperconnected, hyperspeed era
of global flows.
Cross-border data flows are soaring and connecting more countries
As Internet usage continues to grow within individual economies, users are rapidly forming
and deepening international connections. In 2015, 50 percent of Facebook users had at
least one international friend, up from just 16 percent in 2012. Cross-border used bandwidth
has grown 45 times larger over the past decade. In absolute terms, it has grown from 4.7
terabits per second (Tbps) in 2005 to 211.3 Tbps in 2014, for an annual growth rate of
52 percent.28 Over the next five years, total Internet Protocol (IP) traffic is projected to triple,
while cross-border used bandwidth is projected to post a ninefold increase (Exhibit 6).29
Most international Internet traffic travels via an extensive cable network found on the world’s
ocean floors, running along coastlines and between continents. Cross-border capacity
expanded by 38 percent annually from 2007 to 2014 as new submarine cables were built
and old ones were upgraded. Emerging economies are becoming more integrated into this
network (Exhibit 7). In 2005, 75 countries used more than 1 gigabit per second of cross-
border bandwidth; by 2014, that number was up to 164. Emerging economies started from
a small base, but they have outpaced advanced economies in the growth of used cross-
border bandwidth over the past decade.
27 The state of broadband 2015: Broadband as a foundation for sustainable development, International
Telecommunication Union and UNESCO Broadband Commission for Digital Development, September 2015.
28 TeleGeography, Global Internet Geography.
29 Projections of total IP traffic from Cisco Visual Networking Index: Forecast and methodology, 2014–
2019, Cisco, May 2015; projection of cross-border bandwidth from TeleGeography, Global Bandwidth
Forecast Service.
Virtually every type of cross-border transaction now
has a digital component.
50%
share of Facebook
users with at least
one international
friend
31McKinsey Global Institute Digital globalization: The new era of global flows
Exhibit 6
397
1,020
18E
30
744
19E15E14 20E
1,397
17E
543
211
13
147
16E10
1,914
70
2021E
290
1211
46
0908
11
2005
5 19
0706
7
101
SOURCE: TeleGeography, Global Bandwidth Forecast Service; McKinsey Global Institute analysis
Used cross-border bandwidth, global
Terabits per second
Cross-border bandwidth has grown 45 times larger over the past decade—
and may grow another nine times larger by 2021
ForecastActual
45x
>9x
Exhibit 7
Cross-border data flows are surging and connecting more countries
SOURCE: TeleGeography, Global Internet Geography; McKinsey Global Institute analysis
Used cross-border bandwidth
2005
100% = 4.7 Terabits per second (Tbps)
NA
United States and Canada
LA
Latin America
AF
Africa
EU
Europe
ME
Middle East
OC
Oceania
AS
Asia
Regions
Bandwidth
Gigabits per second (Gbps) <50 50–100 100–500 500–1,000 1,000–5,000 5,000–20,000 >20,000
2014
100% = 211.3 Tbps
NOTE: Lines represent interregional bandwidth (e.g., between Europe and North America) but exclude intraregional cross-border bandwidth (e.g., connecting
European nations with one another).
45x larger
NA
EU
LA
ME
AF
AS
OC
NA
EU
LA
ME
AF
AS
OC
32 McKinsey Global Institute 1. A new era of digital globalization
However limitless the public Internet might seem, the portion of the Internet that has been
indexed and can be navigated by anyone using standard search engines is only the surface
of a much larger structure. The “Deep Web” cannot be accessed in the same way, and
it is estimated to be hundreds of times larger than the public Internet. It includes private
company networks, some enormous publicly accessible topic databases (such as climate
data from the US National Oceanic and Atmospheric Administration), libraries and archives,
private chat rooms, the underlying operations of social media sites and other platforms,
and much more. Private data networks have been growing faster than the public Internet as
technology giants expand their dedicated long-haul networks. The share of private networks
in international used bandwidth has increased from 20 percent in 2009 to 35 percent in
2014.30 Much of the Deep Web is legitimate and benign, but it does have some shadowy
corners, collectively known as the “Darknet,” where criminal trade flourishes.31 This report
measures data flows by analyzing used cross-border bandwidth. It therefore captures traffic
of all types—public and private, legitimate and illicit—since these cannot be disaggregated.
Data flows include a huge variety of business and personal communications, transactions,
information, videos, and other digital media content, gaming, and much more. We also
analyzed cross-border digital calls, which have more than doubled from 274 billion call
minutes in 2005 to 569 billion call minutes in 2014. This rising volume is primarily attributable
to the expanded use of voice over Internet protocol (VoIP) technology. Since 2005, VoIP call
minutes have grown by 19 percent per year, while traditional call minutes have grown by
4 percent. Additionally, cross-border computer-to-computer Skype communications have
soared, with call minutes increasing by some 500 percent over the past five years.32 In 2014,
computer-to-computer Skype call minutes were equal to 46 percent of traditional phone
call minutes.
Data flows—both within countries and between them—reflect the activities of individuals
and of businesses. Many people assume that the Internet is dominated by individuals
viewing YouTube and other streaming videos, trading e-mails, and posting on social media.
But a large share of Internet traffic is also driven by companies interacting with their foreign
operations, suppliers, and customers. The business aspect of data flows is likely to take
on a deeper dimension in the near future as more companies embed monitors, sensors,
and tracking devices into their physical assets. As the Internet of Things is more widely
adopted, Cisco estimates that machine-to-machine connections will account for more than
40 percent of global devices and connections by 2019. It could account for more than half
not long after that (Exhibit 8).33 These connections generate very small and intermittent data
bursts that account for only a small share of IP traffic. But those flows represent a great
deal of economic value for companies since they are directly related to making machines,
processes, and supply chains more efficient.
30 TeleGeography, Global Bandwidth Research Service.
31 For an in-depth discussion of this topic, see Daniel Sui, James Caverlee, and Dakota Rudesill, The Deep Web
and the Darknet: A look inside the Internet’s massive black box, Woodrow Wilson International Center for
Scholars, October 2015.
32 TeleGeography, Global Bandwidth Research Service.
33 Cisco Visual Networking Index: Forecast and methodology, 2014–2019, Cisco, May 2015.
Digital platforms are creating more efficient and
transparent global markets. They provide businesses
with enormous built-in customer bases and effective
ways to reach them.
33McKinsey Global Institute Digital globalization: The new era of global flows
Digitization transforms global flows in three ways
Globalization has a very different look today in part because digitization has introduced
three new phenomena into the equation. First and foremost, large-scale Internet platforms
have driven down the cost of cross-border interactions and transactions. Second, purely
digital goods and services are now traded virtually and instantly. And finally, the addition of
“digital wrappers” to traditional products is enhancing their value.
Digital platforms connect people around the world
Digital platforms include e-commerce marketplaces, operating systems (such as Google’s
Android and Apple’s iOS), social networks (such as Facebook, Instagram, Twitter, WeChat,
and QQ), and digital media platforms (such as YouTube, Uvideos, Spotify, Hulu, and Netflix).
Virtual global marketplaces now match job seekers with employers (LinkedIn), freelancers
with assignments (Upwork), borrowers with lenders (Kiva), creative projects with funders
(Kickstarter), travelers with accommodations (Airbnb), and students with education
providers (Khan Academy).
The biggest platforms are creating truly global markets and user communities on a scale
that has never been seen before (Exhibit 9). Facebook’s monthly active user base, for
example, has surpassed the size of China’s population. As of early 2015, creators filming in
YouTube Spaces have produced more than 10,000 videos that have generated more than
one billion views. Alibaba recorded more than $14 billion in sales on its platforms in just 24
hours during its 2015 “Singles Day” promotion, smashing its record set the previous year.
(See Chapter 2 for more on the scope of individual participation around the world.)
Exhibit 8
4
16
6
43
12
19
SOURCE: Cisco; McKinsey Global Institute analysis
By 2019, machine-to-machine connections are expected to account for more than 40 percent
of global devices and connections
Connections,
2019
PCs TabletsTVsMachine-to-machine
(M2M)
OtherSmartphones
22
0
19
33
3
23
Global devices and connections Global IP traffic by devices
100% =
25 billion
100% =
168 exabytes
per month
34 McKinsey Global Institute 1. A new era of digital globalization
The largest corporations can build their own e-commerce sites or open innovation
platforms. But the biggest and most widely recognized public platforms are open
ecosystems that host an entire universe of diverse participants. E-commerce marketplaces
such as Alibaba, Amazon, eBay, Flipkart, and Rakuten, for example, support millions
of vendors, creating enough product variety and price competition to attract enormous
global customer bases. These platforms give smaller enterprises exporting capabilities by
providing them with payment infrastructure, logistics support, and global visibility.
The influence of e-commerce marketplaces on international trade is significant—and still
growing. Today some 16 percent of B2C e-commerce transactions are cross-border, and
that share is projected to reach almost 30 percent by 2020, when international sales could
hit $1 trillion (Exhibit 10). Cross-border B2B e-commerce is even bigger. In 2014, it was
an estimated $1.8 trillion to $2 trillion market. Together, the roughly $2.2 trillion of cross-
border e-commerce in 2015 is equal to approximately 12 percent of global goods trade.
While growth in the overall goods trade has flattened, the portion enabled by e-commerce
is growing.
The size of these platforms, combined with their use of automated processes driven by
algorithms, lowers the marginal costs for platform operators practically to zero.34 Platforms
make it possible for users to research products, services, prices, and alternative choices.
This removes some information asymmetries so that markets function more efficiently,
although it may disrupt traditional intermediaries in the process.
34 Michael Chui and James Manyika, “Competition at the digital edge: ‘Hyperscale’ businesses,” McKinsey
Quarterly, March 2015.
Exhibit 9
SOURCE: Facebook; Twitter; Freelancer; Upwork; Mashable; Fortune; Statista; McKinsey Global Institute analysis
Active users on select platforms, 4Q15 or latest available
Million
Digital platforms are connecting billions of people around the world
YouTube
1,000
650
Alibaba
407
Twitter Skype Amazon PayPal eBay LinkedIn
1,000
1,590
400
320 300 300
179 162
100
35McKinsey Global Institute Digital globalization: The new era of global flows
For businesses, digital platforms provide a huge built-in base of potential customers and
effective ways to market to them directly and launch new products. As social media exposes
hundreds of millions of consumers from around the world to what is available, products can
launch globally and go viral in unprecedented ways. In the fashion industry, for example,
bloggers, vloggers, Instagram, and Twitter are accelerating trends by highlighting what
celebrities wear (from Beyonce’s “kale” sweatshirt to virtually any outfit Kate Middleton
appears in). In 2012, Michelle Obama wore an affordable red-and-white checked dress from
British online fashion retailer ASOS in a photo that was retweeted 816,000 times on Twitter
and shared more than four million times on Facebook; it instantly sold out.
Digital platforms enable small businesses, entrepreneurs, and individuals to connect across
borders, as we discuss in the next chapter.
Digital goods and services can be delivered instantaneously at very little cost
Today there is growing trade in digital goods and services that have near-zero transmission
costs. McKinsey research shows that global spending on digital media grew at 18 percent
annually from 2008 to 2013, compared with negative 5 percent growth in spending on
Exhibit 10
By 2020, some 940 million online shoppers are expected to spend
almost $1 trillion on cross-border e-commerce transactions
SOURCE: AliResearch; McKinsey Global Institute analysis
0.82.5
0.5
2.8
1.8
15
2.4
1.0
19E 2020E
3.1
2.2
18E
2.1
0.7
3.4
17E
1.9
16E
2.2
0.41.9
2014
0.31.6
1.6
0.2
1.4
1.2
16E
0.5
0.6
19E
1.2
2020E
1.2
0.9
1.2
0.7
17E 18E
1.2
0.8
15
1.11.0
0.3
0.4
2014
1.9
1.6
1.8
2.0
1.3
2.1
1.5
Global B2C e-commerce shoppers
Billion
Global B2C e-commerce transaction volume
$ trillion
Cross-
border
% of
total
+27% p.a.
+21% p.a.
Cross-border
Domestic
Forecast
NOTE: Numbers may not sum due to rounding.
15 16 18 21 24 27 29 23 25 28 33 37 42 45
36 McKinsey Global Institute 1. A new era of digital globalization
traditional media.35 E-books, apps, online games, MP3 music files and streaming services,
software, and cloud computing services can be transmitted instantaneously to customers
anywhere in the world there is an Internet connection.
Major media websites, for example, are building global audiences (Exhibit 11). The Wall
Street Journal’s international site traffic grew from 21 percent of its total online readership
in 2013 to 33 percent in 2015. The international share of readers rose from 29 percent to
50 percent for BuzzFeed, and from 27 percent to 61 percent for Yahoo Sports. Roughly
three-quarters of The Guardian’s online readership is outside the United Kingdom.36 Netflix
expanded its business model from mailing DVDs to selling subscriptions for online streaming
in 2007, and by the end of 2015, it had subscribers in more than 190 countries.
35 The McKinsey global media report defines this as spending on over-the-top (OTT) transactional digital
video, OTT subscription digital video, digital recorded music downloads, digital recorded music-streaming
subscriptions, consumer magazine digital circulation and advertising spend, daily newspaper digital
circulation and advertising spend, electronic consumer books, online video games, mobile video games, and
digital learning material. Traditional media consists of spending on physical home video sales and rentals,
physical recorded music, consumer magazine print circulation and advertising spend, daily newspaper print
circulation and advertising spend, print consumer books, and boxed-console and PC video games, and print
education material.
36 All international site traffic shares based on monthly site-traffic data from SimilarWeb, comparing shares in
October 2013 and October 2015.
Exhibit 11
Online traffic from outside country of origin as share of total traffic1
Digital media is attracting global audiences
SOURCE: Similar W eb; McKinsey Global Institute analysis
1 Based on monthly site traffic data from Similar W eb.
October 2015October 2013
91
77
73
27
40
27
29
24
22
24
21
22
78
56
New York Times 25
39
41
50
60
33
75
92
61
41
Business Week
Wall Street Journal
Vogue
Time
BBC News
Financial Times
The Economist
Netflix
Buzzfeed
The Guardian
Yahoo Sports
37McKinsey Global Institute Digital globalization: The new era of global flows
Companies can take advantage of digital platforms to create international buzz and
momentum for global product launches. In 2015, Adele’s song “Hello” racked up 50 million
views on YouTube in its first 48 hours, for instance. The following month, her smash album
25 was made available on CD and as a digital download (although not through streaming
services) featuring the song. It sold 900,000 downloads on the first day, and during its first
week of release, it was No. 1 on the download list of iTunes stores in 110 countries.37 In the
United States, 3.38 million copies of the album sold in the first week, the most since Nielsen
began tracking point-of-sale music purchases in 1991.
“Digital wrappers” enhance and enable other types of flows
Adding a digital component to traditional types of flows can enhance their value. Automotive
manufacturers are racing to develop “connected cars,” with features ranging from voice
recognition and smartphone functionality to preventive maintenance alerts, hazard reaction,
and self-driving capabilities. Logistics companies are using sensors, data, and software to
track physical shipments, reducing the volume of goods lost in transit. FedEx, for example,
allows customers to monitor the progress of packages almost continuously by placing small
tracking devices in them; it also uses sensors to monitor temperature, humidity, barometric
pressure, and light exposure for sensitive cargo.38
This type of continuous data availability is invaluable for companies that operate long and
complex supply chains. One study found that RFID technology can help to reduce inventory
costs by up to 70 percent while improving efficiency. Case studies in Germany, including the
logistics centers of Hewlett-Packard and BMW, found that the technology reduced losses
in transit by 11 to 14 percent.39 Rio Tinto, for example, transmits data continuously from
its mines, processing plants, and vehicle fleets around the world to “excellence centres”
located in Brisbane, Australia.
Online user-generated reviews and ratings are another type of digital wrapper. They give
many individuals the comfort level they need to conduct a cross-border transaction, whether
they are buying a consumer product on Amazon or booking a hotel room thousands of
miles away on Airbnb, Agoda, or TripAdvisor. TripAdvisor, for example, has more than
250 million reviews and opinions from travelers on more than 5.2 million businesses and
properties globally. A recent survey by the UN World Tourism Organization revealed that 70
to 92 percent of travelers in various advanced economies considered online guest reviews
important or very important for their hotel booking decisions—meaning that these reviews
influence billions of dollars in cumulative spending.40
37 Clarisse Loughrey, “Adele’s new album 25 is No. 1 on iTunes in almost every country in the world,” The
Independent, November 26, 2015.
38 Disruptive technologies: Advances that will transform life, business, and the global economy, McKinsey Global
Institute, May 2013.
39 Aysegul Sarac, Nabil Absi, and Stéphane Dauzère-Pérès, “A literature review on the impact of RFID
technologies on supply chain management,” International Journal of Production Economics, volume 128,
number 1, November 2010.
40 Online guest reviews and hotel classification systems: An integrated approach, UN World Tourism
Organization, 2014.
50M
YouTube views of
Adele’s “Hello” in
the first 48 hours
38 McKinsey Global Institute 1. A new era of digital globalization
DIGITAL GLOBALIZATION IS CREATING NEW CHALLENGES AS WELL AS NEW
CHANNELS FOR GROWTH
Just as digitization is transforming individual economies and the business models of
individual companies around the world, it is altering the broader global economy—and these
shifts are even bigger and more complex since they are taking place across nations with
different factor costs, levels of development, and regulatory regimes.
Opening to all types of flows, and particularly data flows and global platforms, has the
potential to disrupt traditional industries even as it creates new channels for growth. (For
more on the issue of jobs, see Chapter 3.) The ease of comparison shopping on digital
platforms, for instance, encourages companies to compete on price. This shift works in the
consumer’s favor but creates pressure on the bottom line for companies. It is becoming a
more competitive world in other ways as well. Digital platforms enable small enterprises and
foreign competitors to move into new markets, and technology-powered companies are
demonstrating the ability to add new business lines with ease.
Information can be transmitted halfway around the world in the blink of an eye, but so can
disruptions. Students can educate themselves online from anywhere on earth, but their view
into how other societies live can heighten their impatience with bleak job prospects at home.
Social media creates global communities in a positive sense, but it also allows networks of
extremists to form and strategize.
A world that runs on data flows is also more vulnerable to grid failures and cybercrime. One
study has estimated that cybercrime costs the global economy some $400 billion in annual
losses; these can include consumer data breaches, financial crimes, market manipulation,
and theft of intellectual property.41 This is not only a business risk; it can even pose
national security risks. The issue of combating global cybercrime is discussed more fully in
Chapter 5.
THE WORLD IS STILL FAR FROM FLAT
The Internet has enabled the creation of global markets, but it cannot fully erase the barriers
of geography. Distance still matters. A significant share of each of the major types of global
flows move within well-established regions rather than between them. This is particularly
true of people and data flows. But even in the goods trade, where transportation costs
have fallen dramatically over the past 25 years, 35 percent of global trade is still regional
(Exhibit 12).
41 Net losses: Estimating the global cost of cybercrime, Center for Strategic and International Studies and
McAfee, June 2014.
Information can be transmitted halfway around the
world in the blink of an eye, but so can disruptions.
39McKinsey Global Institute Digital globalization: The new era of global flows
A look at each country’s major trading partner (including exports and imports) reveals that
for many countries, nearby neighbors are their largest trade partners. Canada trades heavily
with the United States, Argentina and Brazil are major partners, and Germany trades with
the rest of Europe (Exhibit 13). There are exceptions to this pattern, however. Notably, China
has become the largest trade partner for large swaths of Latin America and Africa, while the
United States is China’s largest trade partner.
In e-commerce, the distance separating buyers and sellers reduces the volume of
transactions that are completed, although to only about half the degree seen in physical
trade. This effect may be due to the trust factor, language barriers, the presence of familiar
payment systems, and shipping costs. A recent study based on more than ten billion online
transactions found that a 0.5 percent increase in the distance between the two countries
lowers the volume of online trade between them by 1 percent. Sharing a land border also
significantly affects the volume of trade between two countries.42
The Internet itself is not a seamless global web, in part because the whole world is not
connected. Six billion people do not have high-speed broadband, almost four billion do not
have any Internet access at all, and nearly two billion do not have a mobile phone. Digital
divides persist across income, age, geography, and gender. In Africa, the richest 60 percent
are almost three times as likely to have Internet access as the bottom 40 percent, and the
young and urban have more than twice the access of older and rural citizens.43 These gaps
42 Bo Cowgill and Cosmina Dorobantu, Worldwide gravity in online commerce, August 2014.
43 World development report 2016: Digital dividends, World Bank, January 2016.
Exhibit 12
35
65
While much of the world’s trade in goods is long distance,
roughly half or more of other global flows move within the same region
Distribution of flows between intraregional (short haul)
vs. interregional (long haul), 20141
% of world flow
SOURCE: UNCTAD; UN W orld Tourism Organization; TeleGeography, Global Internet Geography; IMF; McKinsey Global
Institute analysis
1 For goods, services, FDI, and travelers we have divided the world into 10 regions; for data flows we have used
TeleGeography’s six regions.
2 Distribution of services flows for 2014 estimated based on 2011 data; 2013 bilateral traveler data used for people flows.
NOTE: Numbers may not sum due to rounding.
36
64
Long haul (interregional)
Short haul (intraregional)
53
47
54
46
67
33
Services2 FDI
Data
Goods
People2
40 McKinsey Global Institute 1. A new era of digital globalization
are hard to erase. Consider that electricity has been in use for more than a century—and yet
some 1.2 billion people, or 17 percent of the global population, still lacked electricity as of
2013.44
Even among those who are connected, cross-border Internet traffic tends to be regional in
nature. In 2014, two-thirds of used cross-border bandwidth was intraregional, reflecting to
some degree the fact that the world’s digital networks have a hub structure. Some academic
research has characterized the Internet not as a global web but as a series of small worlds.
Barnett and Park examine the network structure of the global Internet by looking at
international hyperlink connections, cross-border bandwidth, and shared website use (from
a website and national perspective). They find that the global Internet is very concentrated
among a few core countries that serve as hubs for broader regions drawn together by
shared language, cultural similarities, and historical ties. Their analysis of shared website
use (web perspective) resulted in a Gini coefficient of .930, indicating that the Web-based
network is very centralized among a small group of core countries: the United States, the
United Kingdom, China, Germany, Brazil, France, India, Italy, Japan, Spain, and Russia.45
44 World energy outlook 2015, International Energy Agency, November 2015.
45 George A. Barnett and Han Woo Park, “Examining the international Internet using multiple measures: New
methods for measuring the communication base of globalized cyberspace,” Quality and Quantity, volume 48,
issue 1, January 2014.
Exhibit 13
China, the United States, or Germany is the major trading partner for most countries
SOURCE: UNCTAD; McKinsey Global Institute analysis
Largest trading partner in goods (exports and imports combined), 2014
NOTE: Data omitted for some small nations as indicated in gray.
41McKinsey Global Institute Digital globalization: The new era of global flows
Another reason that the Internet is not yet fully global is that content production is
concentrated in a few advanced economies. Hollywood, for example, has ruled the global
movie box office and the world’s television screens for decades, creating entertainment
that attracts hundreds of millions of viewers worldwide. The United States continues to
play that role in a more digital world. It accounts for more than 50 percent of online content
consumed in all regions except Europe (Exhibit 14).46 The Internet is a democratizing force in
many regards, but some of the legacy structures and disparities that exist in the offline world
persist online.
Still, the rise of cross-border data flows and a truly global Internet infrastructure is still in its
early days. Just 15 years ago, cross-border data flows were negligible. As the underlying
infrastructure continues to expand and as more users around the world join, the barriers of
distance, languages, and cultural norms could erode, creating a more unified world.
•••
Immense flows of goods, services, finance, people, and data move across the world’s
borders, creating a more global world. The pervasiveness of Internet connectivity and
the spread of digital technologies enable data and information to travel around the world
instantaneously, and this capability is transforming all other types of flows. These forces
have unleashed an accelerating wave of change and intensifying global competition. As
digital platforms create new global marketplaces, they are making globalization a more
inclusive phenomenon. Chapter 2 will discuss how individuals and small businesses—in
advanced and emerging economies alike—are using these digital platforms to form their
own global connections.
46 TeleGeography, Global Internet Geography.
>50%
US share of digital
content consumed
in all regions
except Europe
Exhibit 14
88
66
56
51
36
10
16
32
17
61
18
12
32
3
United States
and Canada 2
Latin America
Europe
Africa
Asia Pacific
SOURCE: TeleGeography, Global Internet Geography; Pingdom; McKinsey Global Institute analysis
Content host
42
15
12
31
100% =
1 million
websites
Europe Asia Pacific OtherUnited States1
The United States is the largest producer of digital content for Internet users across the globe
Location of top 100 websites requested by users
% by user region, as of April 2015
Hosting location of
top 1 million websites, 20132
%
1 Includes United States and Canada for location of top 100 websites requested by users.
2 Based on Pingdom analysis of Alexa top 1 million websites.
© Getty Images
Globalization is no longer just the purview of the world’s largest multinational corporations.
Today digitization has erased many of the barriers that once prevented small and medium-
sized enterprises (SMEs), entrepreneurs, and ordinary citizens from making cross-
border connections.
Companies once had to grow to substantial size before they could afford the resources
needed to export, but digitization has dramatically reduced the minimum scale required
to do business across borders. Small businesses are joining the biggest e-commerce
marketplaces to connect with customers and suppliers anywhere in the world. Capital is
available for microenterprises on platforms such as Kickstarter, where close to 3.3 million
people representing nearly all countries made pledges in 2014.47 More than nine million
freelancers from 180 countries have connected with clients on Upwork for assignments
such as web development, graphic design, and marketing.48
The more inclusive nature of digital globalization has significant implications for businesses
and economies, particularly in developing countries. In these nations, companies and
individuals can use digital platforms as a way to overcome constraints in their local markets
and tap into global customers, suppliers, financing, and talent. Instead of waiting for the
benefits of globalization to trickle down from large corporations, SMEs can become micro-
multinationals in their own right, and startups can be “born global.” Individuals can discover
opportunities, information, and ideas from anywhere in the world.
SMALL AND MEDIUM-SIZED BUSINESSES ARE BECOMING
MICRO-MULTINATIONALS
The ability of SMEs to reach global audiences supports economic growth. Digitization
has empowered many to transform themselves into “micro-multinationals.” Digital
platforms provide small firms with “plug-and-play” infrastructure and the opportunity to put
themselves in front of an enormous built-in global customer base.
Consider all of the tools and platforms that a small Chinese manufacturer has at its
disposal when it becomes a Taobao merchant. The company can open and customize
a Taobao “storefront” for free using a mobile app and upload its merchandise for sale. It
can communicate with customers using an instant messaging service, handle payments
through Alipay, choose an Alibaba-affiliated logistics company for shipping, place targeted
digital ad buys through Alimama, and get a small loan instantly from an Alibaba microfinance
subsidiary that can evaluate the merchant’s credit based on data about its business
47 Kickstarter, 2014: Year in Review presentation.
48 Elance-oDesk annual impact report, 2014.
Instead of waiting for the benefits of globalization
to trickle down from large corporations, SMEs and
startups can go global in their own right.
2. DIGITAL PLATFORMS OPEN
THE DOOR TO NEW PARTICIPANTS
44 McKinsey Global Institute 2. Digital platforms open the door to new participants
performance on the platform. Finally, the company can use Alibaba itself to buy supplies and
professional services.49
Similarly, eBay has been helping merchants sell internationally by offering features such as
the ability to be featured on eBay sites in other countries, a global shipping program, and
the option to clear transactions with PayPal. One study found that more than 35 percent of
the top 1,000 eBay sellers have significant cross-border trade, with premium or featured
eBay stores in other countries.50 The company’s own analysis across select emerging and
advanced economies shows that the share of SMEs that export is sharply higher on eBay’s
platform than among offline businesses of comparable size (Exhibit 15). Small businesses
can use platforms to reach a greater number of markets: in China, South Korea, Indonesia,
Thailand, and South Africa, 90 percent or more of eBay sellers export to more than ten
international markets.51
PayPal enables cross-border transactions by acting as an intermediary for SMEs and their
customers. Participants from emerging economies are senders or receivers in 68 percent
of cross-border PayPal transactions. PayPal also helps facilitate small transactions: the
average point-of-sale transaction using a foreign credit card was $169 across four emerging
economies in 2013, while a sample of PayPal data from the same year suggests an average
49 “Alibaba defined,” Alibaba corporate website. See also China’s e-tail revolution: Online shopping as a catalyst
for growth, McKinsey Global Institute, March 2013.
50 Andy Geldman, “The world’s top eBay sellers,” Web Retailer, February 7, 2014.
51 Small online business growth report: Towards an inclusive global economy, eBay Public Policy Lab,
January 2016.
Exhibit 15
Share of eBay commercial sellers1 and offline enterprises that export, 2014
%
eBay enables SMEs to attain global reach that comparable offline businesses have not achieved
SOURCE: eBay; W orld Bank Enterprise Surveys (using latest data available); McKinsey Global Institute analysis
100100100100
95
100
18
14
8
3
10
22
MexicoIndonesiaIndiaChina South
Africa
Brazil
100
9697
20
16
4
South
Korea
GermanyUnited
States
Emerging economies Advanced economies
eBay sellers
All enterprises
1 eBay commercial sellers are defined as sellers with sales of over $10,000 and at least 10 transactions in previous year.
45McKinsey Global Institute Digital globalization: The new era of global flows
transaction sent to emerging economies of just $38. Alipay performs a similar function for
Taobao merchants, providing a critical element of trust needed to facilitate transactions.
SMEs worldwide are joining e-commerce marketplaces to access business resources and
reach new markets. Amazon now hosts two million third-party sellers, while some ten million
small businesses have become merchants on Alibaba platforms.52 Artisans and customers
from around the world find each other on Etsy, a marketplace for handcrafted and vintage
goods; nearly 30 percent of its gross merchandise sales are international.53 More than
20,000 independent designers and artists showcase their work on Pinkoi, a Taiwan-based
online marketplace. The company has connected with customers in more than 47 countries,
using Facebook to expand its reach throughout the Asia-Pacific region.
Cross-border B2B e-commerce sales were approximately a $1.8 trillion to $2 trillion market
in 2014. By 2020, cross-border e-commerce sales to consumers are projected to hit
$1 trillion, accounting for almost 30 percent of total B2C sales.
For businesses, the biggest social media platforms represent a huge base of potential
customers with built-in ways to reach them effectively and directly. Facebook estimates
that more than 50 million SMEs are on its platform, up from 25 million in 2013, and some
30 percent of their fans are cross-border (Exhibit 16). To put this number in perspective,
consider that the World Bank estimated there were 125 million micro, small, and medium-
sized enterprises in the 132 countries in its database in 2010.54 This points to the importance
of social media exposure as a crucial marketing tool, particularly for companies that hope to
raise their global profile.
52 Amazon.com company facts, corporate website; Jack Ma, “America’s online sales opportunity in China,” The
Wall Street Journal, June 8, 2015.
53 2015 third-quarter financial results, Etsy.
54 Khrystyna Kushnir, Melina Laura Mirmulstein, and Rita Ramalho, Micro, small, and medium enterprises around
the world: How many are there, and what affects the count? World Bank/IFC, 2010.
~10M
merchants operate
on Alibaba
Exhibit 16
50
30
25
201520142013
SOURCE: Facebook; McKinsey Global Institute analysis
50 million SMEs use Facebook to find customers, and 30 percent of their fans are from
other countries
Domestic
70
30
Cross-border
Estimated number of SME pages
on Facebook
Million
Share of SME fans that are cross-border
%
+41% p.a.
46 McKinsey Global Institute 2. Digital platforms open the door to new participants
The increasing ability of SMEs to participate directly in globalization is a relatively new
phenomenon, but it is starting to show up in national statistics. It is most clearly seen in the
United States, where the share of exports by large multinational corporations dropped from
84 percent in 1977 to 50 percent in 2013.55 Companies with fewer than 500 employees
accounted for 97.8 percent of all identified US exporters and 97.2 percent of all identified US
importers in 2011. The number of US exporting entities with fewer than 50 employees, in
particular, has grown more rapidly than firms with 50 to 500 employees.56
An analysis of export data for 16 OECD countries shows mixed evidence of increased SME
participation. Between 2005 and 2012, the SME share of total exports grew in ten of the
countries, including the United States and France. But SMEs lost ground in exports in the
remaining six countries (however, in Portugal and elsewhere, this was likely due to tightening
access to credit for small businesses during a prolonged financial crisis).
MANY TECHNOLOGY-BASED STARTUPS ARE NOW “BORN GLOBAL”
The ability to connect globally opens new entrepreneurial possibilities for individuals.
Anyone with a connection and a great idea can launch a business, drawing on the availability
of enterprise software and cheap computing power in the cloud. Academic literature has
highlighted the emergence of a new wave of global startups and SMEs that are making the
most of these types of Web 2.0 tools to innovate. These capabilities are highly relevant for
working with collaborators, customers, and partners in different countries.57
Today’s digitally powered startups can be born global—attracting users, hiring talent,
purchasing inputs, securing funding, and finding mentors across borders from day one.
Consider coModule, an Estonian startup that created technology that brings the Internet
of Things to electric bikes and scooters. Its prototype was unveiled in Barcelona, its seed
funding came from Germany, and its components are sourced from China. The company is
scaling up production and eyeing user markets across Europe and Asia.
MGI surveyed 271 startups worldwide through a partnership with 1776, a global incubator
and venture fund. By working with 1776’s Challenge Cup competition and its Startup
Federation program, we were able to expand the reach of the survey to 19 countries. The
businesses surveyed included members of the Startup Federation, the Global Accelerator
Network, and current and former competitors at 1776’s Challenge Cup events around
the world. (See the technical appendix for more detail on the survey.) While these startups
represent a more globally connected and tech-savvy selection than the typical small
business, the results suggest that even the smallest and youngest enterprises can execute
a global vision if their business model is built on digital technologies. This is a relatively new
development. When many of today’s Internet giants were started, they focused on the US
market alone for a significant period before expanding to other countries. Today, many
digital-based startups market to a global audience from their inception.
55 US trade in goods associated with US multinational corporations from US Bureau of Economic Analysis.
56 “A profile of US importing and exporting companies, 2010–2011,” US Census Bureau press release, April
5, 2013.
57 Jim Bell and Sharon Loane, “‘New-wave’ global firms: Web 2.0 and SME internationalization,” Journal of
Marketing Management, volume 26, issue 3–4, April 2010.
Digitally powered startups can be “born global,”
connecting with international customers, suppliers,
capital, and mentors from day one.
47McKinsey Global Institute Digital globalization: The new era of global flows
A surprising 86 percent of survey respondents pointed to at least one cross-border activity
(Exhibit 17). Almost two-thirds have customers or users in other countries, and almost half
reported sourcing talent from other countries. The rate of cross-border participation varies
widely by company stage. Companies in the growth and scaling phase report more than
twice as many cross-border activities as companies in the concept phase.
Moreover, the surveyed startups from emerging economies were more global across several
dimensions than their counterparts in advanced economies (Exhibit 18). Specifically, they
were more likely to report using foreign inputs, participating in an international accelerator or
incubator program, and having international customers. Those based in South America had
notably more global business activities than those in North America.
This disparity underscores the importance of global digital platforms for startups and small
businesses seeking to overcome limited domestic markets and credit constraints. But the
surveyed company founders reported a number of reasons for seeking out global ties.
MPOWER, a US-based student loan financing company for international students, set out
to solve a problem specific to users from other countries: the inability to get loans from
traditional banks in the United States. Others, such as South Africa–based Anaso Diabitiz,
a diabetes management platform, find that their product is discovered by international
consumers through social media. As the product took off in its home market of South
Africa, patients in Nigeria and Kenya learned about it through Facebook and Twitter—and
today two-thirds of the subscribers are from across the broader African continent and the
Middle East.
Exhibit 17
SOURCE: MGI Global Startup Survey 2015; McKinsey Global Institute analysis
14
36
36
39
47
62
Mentors or advisers
in other countries
Customers, clients, or
users in other countries
Funding or investors
from other countries
Talent (including freelancers)
hired from other countries
Accelerator or incubator
in other countries
Inputs sourced from other
countries (e.g., parts,
cloud services)
86 percent of the tech-enabled startups surveyed by MGI engage in at least one cross-border activity
Share of startup respondents
engaged in cross-border activity
%
Participation by activity type
%
100% = 271 respondents
>1
None
21
14
6
7
13
20
18
Average =
2.4
86%
1
2
3
4
5
6
7
Number of cross-
border activities
48 McKinsey Global Institute 2. Digital platforms open the door to new participants
Whether it is deliberate or serendipitous, international cross-pollination is likely to continue
and accelerate in a more digitally connected global world. But while digital platforms are
enablers for startups, they do not negate the need for local ties and personal connections
based on trust. Our surveyed startups continue to find talent in low-tech ways (with almost
three-quarters relying on personal referrals), and they rated local mentors as more effective
than foreign mentors.
INDIVIDUAL PARTICIPATION HAS GROWING ECONOMIC AND
SOCIAL IMPORTANCE
Thanks to social media and other Internet platforms, individuals are forming cross-border
connections as well. Some 29 percent of Skype call minutes are cross-border, for instance,
compared with less than 2 percent of standard telephone call minutes. Some 92 percent
of online project work on Freelancer.com is cross-border, while only 16 percent of the
workforce in the United States and 7 percent of the workforce in the European Union is
foreign-born. Some 37 percent of Khan Academy students are from foreign countries, while
only 8 percent of all tertiary students in the OECD are foreign-born.
We estimate that nearly one billion individuals around the world are direct participants in
some form of globalization (Exhibit 19). MGI analysis of international ties on Facebook,
Twitter, LinkedIn, and WeChat shows that some 914 million people have at least one
international connection on a social media platform (even adjusting for overlap between
users of these platforms). Each year, some 360 million people participate in cross-border
Exhibit 18
SOURCE: MGI Global Startup Survey 2015; McKinsey Global Institute analysis
1 The difference in participation rates between developed and emerging startups is statistically significant. P-values for inputs (p = 0.005), incubator/accelerator
(p = 0.05), and customers (p = 0.07) are significant at the 0.07 level and below.
2 Emerging economies represented in the survey are Argentina, Armenia, Brazil, China, Colombia, Cyprus, Egypt, Estonia, Hungary, India, Israel, Jordan,
Kenya, Mexico, Moldova, Pakistan, Poland, Russia, South Africa, and Ukraine. Advanced economies represented in the survey are Australia, Canada,
Germany, Ireland, the United Kingdom, and the United States.
Tech-based startups from emerging economies were more likely than other survey respondents
to report seeking out customers and inputs from abroad
Share of startups participating by activity type1
%, 100% = 271 respondents
Customers, clients, or
users in other countries
77
54
Inputs (e.g., cloud services)
sourced from other countries
45
32
Participation in foreign
accelerator or incubator
17 12
Emerging2
Advanced2
49McKinsey Global Institute Digital globalization: The new era of global flows
e-commerce.58 Forty-four million individuals provide services to clients in other countries
on the biggest digital marketplaces for freelance work.59 We estimate that some 12.6 million
people take part in online courses developed and offered in other countries.60
In addition to facilitating cross-border e-commerce, the biggest digital platforms have
built global communities that generate tremendous flows of personal communication,
information, news, and content. Facebook averaged more than a billion daily active users
in December 2015, and Google processes some 3.5 billion searches a day.61 Tencent’s
WeChat instant messaging platform now has 549 million monthly active users—a number
58 Cross-border B2C e-commerce market trends, AliResearch and Accenture, June 2015.
59 Rate observed from site traffic on Freelancer.com applied to total participation in the major platforms
described in Siou Chew Kuek et al., The global opportunity in online outsourcing, World Bank and Dalberg
Global Development Advisors, June 2015.
60 We sum the total membership of leading educational platforms (Coursera, Khan Academy, edX, and Udacity)
and assume the Khan Academy cross-border rate of 37.2 percent for all.
61 Corporate websites and Internet Live Stats.
Exhibit 19
Individuals are participating in globalization, and 914 million have cross-border social media connections
SOURCE: Facebook; AliResearch; US Department of Commerce; OECD; W orld Bank; McKinsey Global Institute analysis
Students
studying abroad
5 million
Cross-border
e-commerce shoppers
361 million
International
travelers
429 million
People living outside
home country
244 million
Cross-border
online students
13 million
Cross-border
online workers
44 million
Social networking users with
at least one foreign connection
914 million
NOTE: Numbers adjusted to account for overlap between platforms and for individuals making multiple international trips in the same year.
DUPLICATE
from ES
50 McKinsey Global Institute 2. Digital platforms open the door to new participants
that is beginning to approach the population of the entire ASEAN region.62 YouTube has
more than one billion users, and 300 hours of video are uploaded every minute.63
Furthermore, social media plays an increasingly important role in connecting people in
developing nations to the rest of the world, opening new avenues for work, learning, and
communication. The share of Facebook users with cross-border friendships is higher in
emerging countries than in developed countries (54 percent vs. 44 percent). It is growing
rapidly, having increased by 3.7 times since 2012 (Exhibit 20). Almost half of those followed
by Twitter users in emerging economies are from other countries, compared with just under
40 percent in advanced economies. This could be because social media users in emerging
economies would already tend to be among that country’s most tech-savvy and globally
minded segment of the population. But it also speaks to the point that people in developing
countries are embracing social media platforms as their link to the rest of the world.
Virtual connections are changing the way people interact with friends, family, and even
strangers across borders—and they sometimes spill over into the physical world. In the
24 hours after the 2015 terrorist attacks in Paris, 4.1 million people in and around the
city marked themselves as safe through Facebook’s Safety Check feature, triggering
notifications to 360 million users around the world that their friends and family were
unharmed.64 Earlier that year, after a major earthquake struck in Nepal, Facebook
implemented a special “donate” button that brought in more $15 million in contributions to
relief efforts from 770,000 people in more than 175 countries.65
62 Tencent corporate financial statement, first quarter 2015.
63 YouTube corporate website.
64 Robinson Meyer, “One small worry about Facebook Safety Check,” The Atlantic, November 18, 2015.
65 Ken Yeung, “Over 770K Facebook users donated $15M to support Nepal earthquake relief,” VentureBeat,
September 28, 2015.
Exhibit 20
The share of Facebook users with at least one international friend tripled in just three years,
with the fastest growth in emerging economies
SOURCE: Facebook; W orld Bank; UNCTAD; McKinsey Global Institute analysis
% of monthly active Facebook users with at least one international friend1
0
10
20
30
40
50
60
+35 p.p.
201514132012
0
10
20
30
40
50
60 +40 p.p.
2012 201513 14
0
10
20
30
40
50
60
+26 p.p.
2012 13 14 2015
World Emerging economies Advanced economies
1 W eighted by number of Internet users to arrive at world average and averages for emerging and advanced economies.
51McKinsey Global Institute Digital globalization: The new era of global flows
Of course, there are limits to the depth and endurance of online connections. Oxford
anthropologist and psychologist Robin Dunbar famously concluded that the maximum
number of casual friendships the human brain can sustain is around 150, while the average
Facebook user has 388. His findings have been borne out by subsequent researchers
applying his theory to the world of social media.66 A study of one million Coursera users
taking massive open online courses offered by the University of Pennsylvania concluded
that only around half of those who register for a course view any lectures, and the average
completion rate was just 4 percent.67 And while social media can be an effective political
organizing tool, it can also be a vehicle for what has been termed “clicktivism,” in which large
numbers of people join a cause by liking, tweeting, or signing an online petition but are not
motivated to take real action.
Despite these caveats, there are in fact ways in which online connections can translate into
tangible economic impact. Digital platforms such as Expedia, TripAdvisor, Yelp, Agoda,
and many more facilitate leisure travel and tourism. Founded in 2008, Airbnb has quickly
built a network of hosts in 34,000 cities in more than 190 countries. Within 40 days of the
United States and Cuba reopening ties, the platform was listing some 2,000 properties on
the island.68 Couchsurfing.com has ten million members around the globe; its network of
free homestays gives travelers on a tight budget the means to see the world and take part in
face-to-face cultural exchanges.69
Digital networks may also facilitate migration. Online talent platforms aggregate data on
candidates and job openings across broader regions and even globally, enabling users to
find international career opportunities.70 As social media use becomes more ingrained into
daily life around the world, expatriates have a way to stay in closer touch with family and
friends in their homeland, reducing the social and emotional hardship of moving to a new
country. Social media can also notify migrants and refugees of changes to government
policies. When a government agency in Nuremburg tweeted in German that it would loosen
immigration protocols for Syrian citizens, the news went viral and led to a rush of 20,000
refugees over the German border within a week.71
Digital platforms offer individuals new ways to learn, collaborate, and develop new skills—
and to showcase their talents to potential employers. Some 44 million people worldwide
find freelance work on Freelancer.com, Upwork, and other digital platforms, while nearly
400 million have posted their professional profiles on LinkedIn. Individuals with creativity and
drive can use digital platforms to propel themselves onto a global stage in ways that would
have been unimaginable in the pre-digital world (see Box 2, “The YouTube economy”).
66 R. I. M. Dunbar, “Coevolution of neocortical size, group size, and language in humans,” Behavioral and Brain
Sciences, volume 16, issue 4, December 1993. See also Nicole Ellison, “Connection strategies: Social capital
implications of Facebook-enabled communication practices,” New Media & Society, volume 13, number
6, September 2011; Bruno Goncalves, Nicola Perra, and Alessandro Vespignani, “Modeling users’ activity
on Twitter networks: Validation of Dunbar’s number,” PLOS One, volume 6, number 8, 2011; and Maria
Konnikova, “The limits of friendship,” The New Yorker, October 7, 2014.
67 Laura Perna et al., “The life cycle of a million MOOC users,” presented at the MOOC Research Initiative
Conference in Austin, Texas, December 5, 2013.
68 Airbnb corporate website. See also Mark Scott, “What Uber can learn from Airbnb’s global expansion,” The
New York Times, July 7, 2015.
69 Couchsurfing.com, About Us.
70 A labor market that works: Connecting talent and opportunity in the digital age, McKinsey Global Institute,
June 2015.
71 Andrea Thomas, Matt Bradley, and Friedrich Geiger, “Obscure German Tweet helped spur migrant march from
Hungary,” The Wall Street Journal, September 10, 2015.
190+
countries with
Airbnb listings
52 McKinsey Global Institute 2. Digital platforms open the door to new participants
Box 2. The YouTube economy
YouTube is proving to be a powerful marketing tool for global brands and
established stars, but it can also launch individuals from obscurity to
opportunity. Consider Havard Rugland, a Norwegian whose dreams of playing
American football once seemed far-fetched. But his mind-boggling YouTube
video of his kicking skills received three million hits, earning him tryouts and an
NFL signing (if not yet stardom).
A number of self-made YouTube stars have created channels that attract
millions of subscribers—and millions in ad revenue. While many adults may
not recognize their names, teenagers certainly do. A recent survey by Variety
asked teens to rank celebrities on various measures of influence, and eight
of the top ten slots went to YouTube creators rather than to more mainstream
media stars.1
Felix Kjellberg, the Swedish Internet sensation known as “PewDiePie,”
became a sensation after posting videos of himself playing video games. He
eventually racked up some 40 million subscribers to his YouTube channel and
millions in annual earnings. Now he and some of his fellow YouTube stars will
be showcased on Revelmode, a digital network backed by Disney-owned
Maker Studios.2 DisneyCollectorBR, one of the most popular channels on
YouTube, is devoted simply to unwrapping and critiquing toys. More than eight
million subscribers follow beauty vlogger Michelle Phan’s channel for makeup
tutorials. She parlayed this influence into deals with Lancôme and L’Oréal—
and eventually into her own subscription cosmetics company, recently valued
at $500 million.3
A number of singers have been discovered after posting videos of themselves
performing on YouTube. They include Cody Simpson, Austin Mahone (who
became an opening act for Taylor Swift), and 5 Seconds of Summer (who
wound up touring with One Direction). Arnel Pineda, who spent part of
his youth homeless in the Philippines, famously posted videos of his band
performing Journey covers—and went on to be hired by Journey as its lead
singer. The Weeknd, spotted on YouTube by Drake, dominated the Billboard
charts in 2015 and recently earned an Oscar nomination for best original song.
1 Susanne Ault, “Digital star popularity grows versus mainstream celebrities,” Variety, July
22, 2015.
2 David Pierson, “YouTube sensation PewDiePie launches his own network,” Los Angeles
Times, January 13, 2016.
3 Natalie Robehmed, “How Michelle Phan built a $500 million company,” Forbes, October 5,
2015; Andrea Chang, “YouTube’s biggest stars are cashing in offline,” Los Angeles Times,
August 7, 2014.
53McKinsey Global Institute Digital globalization: The new era of global flows
Platforms such as Facebook, Instagram, YouTube, and Twitter, along with countless blogs
and comment boards, are changing the nature of media content in fundamental ways. For
decades, movies, music, and news were created by large companies and consumed by
passive audiences. But now individual users anywhere in the world can comment, debate,
share, parody, co-create, and upload their own original content. The Internet has turned a
one-way monologue into a two-way digital conversation in which anyone can make their
voice heard.
•••
Globalization was once driven almost exclusively by the world’s governments, large
multinationals, and major financial institutions. Today artisans, app developers, freelancers,
and all manner of startups can participate. SMEs can scale up rapidly and even go head-
to-head with more established businesses. And individuals from Canada to Cameroon can
forge their own global connections, whether for business, personal ties, entertainment,
creativity, or simple curiosity about the world beyond their own borders. While this capability
is a tremendous social good, it does not mean that all countries share in the benefits of
globalization equally, as Chapter 3 will discuss.
© Getty Images
Now that commerce, data, and people move more fluidly across the world’s borders, most
countries have been drawn into networks of global flows to at least some degree. But their
connectedness varies enormously, and there is room for nations to play different types of
roles within those networks.
Our previous report on global flows introduced the MGI Connectedness Index to provide a
multidimensional picture of how extensively countries around the world are engaging with
the broader global economy.72 But globalization does not stand still. Our latest index offers
an updated snapshot—and by using improved data sources, it also brings our picture of the
world’s connections into sharper focus. This year’s index finds Singapore, a small country
that punches far above its weight in all types of global flows, at the top of the rankings. It is
followed by the Netherlands (one of Europe’s main digital hubs), the United States, Germany,
Ireland, and the United Kingdom.
However, nation-states are not the only unit of analysis for understanding globalization.
Within countries, individual cities and states are pursuing their own opportunities to take
part in global flows. We find that there are only eight truly “global” cities, most of which are
in advanced economies. Some states and provinces, such as California and Guangdong,
have developed such extensive international ties that they should be considered major
players in the global economy in their own right. At a broader level, larger trading blocs and
various regions of the world have different flow dynamics and levels of connectedness. This
view reveals that regions populated by countries that rank lower on our Connectedness
Index, such as those in Africa and Latin America, could start the process of deepening their
global ties by focusing on interactions with their neighbors. Measuring globalization through
each of these lenses provides additional insights into how the world’s web of connections is
constantly evolving.
SINGAPORE TOPS THE LATEST MGI CONNECTEDNESS INDEX
The MGI Connectedness Index offers a comprehensive ranking for 139 countries based
on inflows and outflows of goods, services, finance, people, and data. For each country, it
takes into account the size of each flow relative to GDP or population (flow intensity); it also
considers that country’s share of the global total within each type of flow. The combination
of these two factors results in a “connectedness score” for each type of flow. To gauge the
overall connectedness of a given country, we calculate an average of its score on each of the
five flows. (See the technical appendix for a full discussion of methodology.)
72 Global flows in a digital age: How trade, finance, people, and data connect the world economy, McKinsey
Global Institute, April 2014.
Singapore, the Netherlands, the United States,
Germany, Ireland, the United Kingdom, and China
top the latest MGI Connectedness Index.
3. HOW COUNTRIES, CITIES, AND
REGIONS ARE CONNECTING
56 McKinsey Global Institute 3. How countries, cities, and regions are connecting
Our index is not the only ranking that examines how countries participate in globalization.73
Other studies differ in the metrics included and how each is weighted. Some look at a
combination of flows and other indicators such as trade barriers. Others also include
measures of cultural globalization. The technical appendix of this report provides an
overview of several of these indexes and their key differences.
We believe that our Connectedness Index is the simplest, most transparent measure of how
countries are actually participating in globalization, or their openness to all types of flows.
In addition, our index methodology controls for the size effect of a country when measuring
openness. Some other indexes focus solely on flow intensity, examining each country’s
level of trade or capital flows relative to GDP. However, in dollar terms, this leaves the world’s
largest economies—notably the United States and China—looking comparatively closed;
their domestic markets are so large that only a relatively small share of their economic
activity is cross-border. Conversely, Luxembourg, Singapore, and other smaller countries
inevitably have larger flows of goods, services, and finance relative to the size of their
economies. We therefore combine flow intensity with each country’s share of the global total
to offer a more accurate perspective on its significance in world flows.
Exhibit 21 shows the index rankings for the top 25 countries along with a selection of other
major advanced and emerging economies. The technical appendix contains the complete
results for all 139 countries we analyzed.
Singapore claims the top spot in the rankings this year. Its globalization journey began
decades ago when it emerged as Southeast Asia’s global shipping hub, particularly for oil
and fuels. The nation also adopted four official languages (English, Mandarin, Tamil, and
Malay); this not only reflected its multicultural population but also positioned the country
to do business with the world. In recent decades, Singapore has implemented an explicit
strategy to become a regional hub for services and finance; attracting skilled international
talent and establishing tax policies and incentives to draw FDI were key priorities to make
this a reality. Singapore ranks second overall in the World Economic Forum’s Global
Competitiveness rankings, with particularly high scores for infrastructure, higher education
and training systems, and transparent and efficient institutions. It is also one of the most
digitally wired countries in the world; government statistics show that 87 percent of
households had broadband access in 2014.
Following Singapore are the Netherlands, the United States, Germany, Ireland, and the
United Kingdom. All of these countries have slightly different profiles. The Netherlands is a
major hub for Europe’s data traffic as well as a port and distribution point for traded goods.
It has created tax and regulatory regimes to become a major financial center and to attract
the holding companies and subsidiaries of many major corporations, boosting both financial
flows and services trade. Ireland has taken a similar approach, establishing corporate
tax and regulatory policies to attract high levels of FDI and significant trade in services. A
long list of major US companies, including some of the biggest tech and pharmaceutical
firms, have incorporated or established European operations in Ireland because of its tax
advantages. In fact, foreign-owned enterprises contribute 55 to 60 percent of the gross
value added of all companies located there. As a result, Ireland rose from No. 14 in MGI’s
previous index to No. 5 in this year’s ranking.
73 See, for example, Pankaj Ghemawat and Steven A. Altman, DHL Global Connectedness Index 2014:
Analyzing global flows and their power to increase prosperity, DHL, 2014; and The ICT globalisation index, the
Economist Intelligence Unit, 2014.
57McKinsey Global Institute Digital globalization: The new era of global flows
Exhibit 21
MGI Connectedness Index
Country connectedness index and overall flows data, 2014
Rank of participation by flow as measured by flow intensity and share of world total
1–10 11–25 26–50 >50Connectedness index rank 100+ <7070–99Flow intensity Rank Country Score Connectedness Index rank Flow value1 $ billion Flow intensity2 % of GDPGoods Services Finance People Data 1 Singapore 64.2 1 2 2 12 6 1,392 452 2 Netherlands 54.3 3 3 6 21 1 1,834 211 3 United States 52.7 7 7 3 1 7 6,832 39 4 Germany 51.9 2 4 8 3 2 3,798 99 5 Ireland 45.9 32 1 1 28 9 559 227 6 United Kingdom 40.8 13 5 5 6 3 2,336 79 7 China 34.2 4 16 4 82 38 6,480 63 8 France 30.1 11 8 9 7 4 2,262 80 9 Belgium 28.0 5 6 33 33 8 1,313 246 10 Saudi Arabia 22.6 20 28 27 2 53 790 106 11 United Arab Emirates 22.2 6 23 17 4 46 789 196 12 Switzerland 18.0 12 11 10 17 13 848 115 13 Canada 17.3 16 22 11 11 18 1,403 79 14 Russia 16.1 21 25 18 5 25 1,059 57 15 Spain 14.4 25 13 19 14 16 1,105 79 16 Korea 14.0 8 12 28 50 44 1,510 107 17 Italy 13.4 17 18 24 16 19 1,587 74 18 Sweden 13.0 29 14 22 31 5 572 100 19 Austria 11.7 26 17 31 20 12 470 108 20 Malaysia 11.6 9 19 25 26 43 610 187 21 Mexico 10.7 14 63 34 18 41 1,022 80 22 Thailand 10.7 10 15 36 44 64 605 162 23 Kuwait 10.6 37 46 13 13 75 306 153 24 Japan 10.5 15 20 12 81 20 2,498 54 25 Kazakhstan 10.0 48 73 41 8 57 176 83 26 Ukraine 9.8 38 39 87 10 34 133 101 27 Australia 9.7 30 34 21 15 33 825 57 28 Denmark 8.9 35 9 32 41 11 369 108 29 Jordan 8.8 73 50 75 9 83 50 138 30 India 8.5 24 10 35 58 70 1,316 64 32 Czech Republic 7.5 18 33 57 59 15 397 193 34 Poland 7.0 23 31 47 34 22 585 107 35 Hungary 6.8 22 30 26 62 17 287 209 36 Norway 6.0 36 24 20 46 24 458 92 37 Vietnam 5.7 19 54 45 103 61 350 188 39 Finland 5.5 46 27 23 70 10 390 144 40 Portugal 5.5 47 36 30 23 31 255 111 41 Turkey 5.1 28 40 53 38 29 521 65 43 Israel 4.9 51 32 49 24 56 248 82 44 Brazil 4.5 41 38 14 125 30 869 37 45 Chile 4.1 45 58 16 102 27 239 92 47 Greece 4.1 60 29 54 35 42 160 67 48 New Zealand 3.9 67 48 61 25 51 130 63 51 Indonesia 3.4 31 49 38 106 76 504 57 53 South Africa 3.3 34 57 52 64 80 277 79 54 Philippines 3.2 54 41 44 52 67 230 81 64 Morocco 2.6 58 43 74 56 65 104 97 73 Egypt 2.2 68 42 69 73 71 158 55 83 Nigeria 1.9 55 76 48 128 98 268 47 86 Peru 1.8 62 88 51 104 49 122 60 118 Kenya 1.3 100 84 127 119 91 35 58 SOURCE: McKinsey Global Institute analysis 1 Flows value represents total goods, services, and financial inflows and outflows. 2 Flow intensity represents the total value of goods, services, and financial flows as a share of the country’s GDP. 58 McKinsey Global Institute 3. How countries, cities, and regions are connecting In contrast, the United States and Germany both follow a generalist model with strength across all five flows. The United Kingdom also has broad participation across flows, with a spike in cross-border service and financial flows, a reflection of London’s role as a global financial hub. China’s surge in our rankings is particularly noteworthy. It has climbed from 25th in our previous index to the No. 7 spot. Now the second-largest economy in the world, it accounts for a larger share of all the total world’s flows. Although its participation has risen across all types of flows, it is strongest in trade and FDI and notably lower in data flows and people flows. China ranks only 38th for data flows, reflecting government restrictions on access to foreign Internet sites. China’s short-term people flows are rising: it welcomed 56 million tourists in 2014, and 109 million Chinese traveled abroad, a tenfold increase since 2000.74 Despite those trends, it ranks only 82nd in people flows, since this metric is based heavily on long-term migrants. Japan and South Korea have similar patterns of strong participation in traditional flows of goods, services, and finance but much weaker participation in data and people flows. Japan ranks only 24th in overall connectedness, below Malaysia, Mexico, and Thailand. Like China, it posts a particularly low ranking on people flows, at only 81st in the world. Japan also ranks only 20th for both data flows and service flows. Language and cultural barriers may explain part of this result, as may the residual effects of the country’s history as a closed society. Japan also attracts relatively little foreign investment, and despite its manufacturing prowess, its goods trade has been hindered over the years by an adherence to its own proprietary technologies and standards rather than an embrace of open platforms.75 South Korea ranks higher in our index, although still only No. 16 in the world, due to lower rankings on finance, data, and people flows. Saudi Arabia places 10th in the rankings, followed by the United Arab Emirates at No. 11. Both are major oil exporters that meet most of their demand for manufactured consumer goods through imports. In addition, both have very high rankings in people flows—second and fourth in the world, respectively—reflecting large numbers of both low-skill guest workers and highly skilled professionals. In fact, more than half of Saudi Arabia’s labor force is made up of foreign workers on temporary contracts.76 The UAE has significant service and financial flows in addition to oil exports. Dubai has actively built a role as one of the world’s major ports, transit hubs, and financial centers. 74 United Nations World Tourism Organization statistics. 75 The future of Japan: Reigniting productivity and growth, McKinsey Global Institute, March 2015. 76 Saudi Arabia beyond oil: The investment and productivity transformation, McKinsey Global Institute, December 2015. Singapore’s top ranking reflects its successful strategy to become a regional hub for services and finance. Attracting skilled talent and FDI were among its key priorities. 59McKinsey Global Institute Digital globalization: The new era of global flows ALTHOUGH MORE COUNTRIES ARE BUILDING GLOBAL CONNECTIONS, THERE ARE LARGE GAPS IN PARTICIPATION Today global connections link a larger and more diverse range of countries. For the first time in history, emerging economies are counterparts on more than half of global trade flows, and South-South trade among these countries is the fastest-growing type of connection (Exhibit 22). The value of traded goods and services plus financial flows exceeded 80 percent of GDP for 72 countries in 1990. By 2014, 121 nations were above this threshold. Today even the smallest countries are participating to some degree in globalization. Flows are concentrated in a few leading countries Despite these trends, the world is still far from fully globalized. The MGI Connectedness Index shows that advanced economies in general remain more connected than developing countries—and the leading countries are far ahead of everyone else. The aggregate connectedness score (shown in the third column of Exhibit 21) reveals a very large distance between Singapore, the overall leader with a score of 64, and every other country. Ireland is No. 5 in the ranking, but its overall score is only 46, indicating that it is roughly two-thirds as connected as Singapore.77 Saudi Arabia ranks No. 10 overall, but its score is only 23, meaning it is only one-third as connected as Singapore. Beyond the top 25, all 114 77 See the technical appendix for details on how the scores were calculated and how the index was constructed. Exhibit 22 60 20 80 100 40 0 Other South-North 2014 North-North China-South China-North Other South-South 2000 111002 06 12090403 0801 07 1305 7x 20x 8x 3x 2x Change in value of trade, 2000–14 SOURCE: UNCTAD; McKinsey Global Institute analysis Emerging economies are now involved in more than half of the world’s goods trade Bilateral goods trade by development status, 2000–14 % of total goods trade 60 McKinsey Global Institute 3. How countries, cities, and regions are connecting remaining countries have overall connectedness scores in the single digits. This indicates that all countries have significant room to expand their participation in global flows. This pattern is apparent not only in the aggregate connectedness score but also within each type of flow. All flows are disproportionately concentrated among a small set of countries, with huge gaps between leaders and laggards (Exhibit 23). The top 15 countries in traded goods, for instance, account for 63 percent of the global total, down only slightly since 2005. That share is 62 percent in services, and 79 percent in FDI. At this simple level, only cross- border data flows have a clear reduction in the share of the top 15 countries, falling from 86 percent in 2005 to 77 percent in 2014. Exhibit 23 13 21 66 63 23 14 22 65 13 14 23 62 4 17 79 79 5 16 11 86 2 77 7 16 SOURCE: UNCTAD; IMF; TeleGeography, Global Internet Geography; McKinsey Global Institute analysis Flows remain concentrated among a few leading countries, with little change since 2005 % of world total Global goods flow distribution 2005 2014 Global services flow distribution 2005 2014 Global FDI flow distribution 2005 2014 Global data flow distribution 2005 2014 1 Tbps = terabits per second. NOTE: Numbers may not sum due to rounding. Next 20 countriesTop 15 countries All others $10.6 trillion $19.0 trillion $2.5 trillion $4.9 trillion $1.39 trillion $1.63 trillion 4.8 Tbps1 211 Tbps1 61McKinsey Global Institute Digital globalization: The new era of global flows The individual country scores for each type of flow present an even starker picture, with large drop-offs between the most connected country and the second, third, and following countries. For goods flows, the fifth-ranked country, Belgium, has a score that is only 78 percent of Singapore’s leading score. Global service flows are even more skewed, with the fifth-ranked country (the United Kingdom) at just 35 percent of the leader (Ireland). In data flows, the fifth-ranked country (Sweden) is only 25 percent as connected as the leader (the Netherlands).78 Lagging countries are catching up to leaders only very slowly We use statistical tests of convergence to see if the gaps between country participation in global flows are closing over time. Beta convergence, for instance, measures the extent to which lagging countries are catching up to leaders, while sigma convergence measures whether the variation between countries overall is narrowing. The results show that goods and migration flows have positive beta convergence, indicating that lagging countries are catching up—but only very slowly. If current trends hold, the convergence tests indicate that it will take at least eight years for the gap in goods flows between leaders and laggards to be reduced by half, and 11 years for the same to happen in migration flows. The gap between the leading countries and the rest of the pack is particularly stubborn in foreign investment. The top 15 countries in FDI flows have consistently accounted for approximately 80 percent of the world total over the past 20 years—and we estimate it would take 13 years for the current trajectory to cut the gap between laggards and leaders by half. Data flows do not show beta convergence, indicating that lagging countries are not catching up to the leaders. This may reflect the fact that data flows are still a relatively new phenomenon and continuing to grow quickly for even leading countries. These gaps in participation matter given the economic value of global flows—a topic that will be explored more fully in Chapter 4. Exhibit 24 graphically illustrates the large gaps between leading countries and the rest of the world in the context of their relative prosperity. It shows each country’s overall connectedness score on the vertical axis and its per capita income on the horizontal axis. In general, country connectedness rises with income, but there are interesting exceptions. China stands out for its high level of connectedness relative to its income level. This reflects not only its outsized role in global manufacturing value chains, but also its significant financial flows and its growing flows of services. At the other extreme are several Middle Eastern economies that have high per capita income but much lower global connectedness. 78 See the technical appendix for rankings by type of flow. The MGI Connectedness Index shows that advanced economies are generally more connected than developing countries—and the leaders are far ahead of everyone else. 62 McKinsey Global Institute 3. How countries, cities, and regions are connecting Exhibit 24 A small group of leading countries are much more connected than the rest of the world 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 10 7040 600 20 8050 14030 United States Belgium Australia India Italy China South Korea Russia Canada Japan Germany Mexico Spain United Kingdom Switzerland Connectedness score, 2014 Per capita GDP, 2014 $ thousand, purchasing power parity, current international dollar Ukraine Greece Czech Republic Norway Austria Turkey QatarFinland Denmark Ireland Singapore Indonesia Sweden United Arab EmiratesSaudi Arabia Portugal Kuwait France Netherlands Emerging Developed Correlation coefficient (r) = 0.54 Size of circle represents $ value of flows in 2014 SOURCE: IMF; McKinsey Global Institute analysis Vietnam Brazil Thailand Malaysia 63McKinsey Global Institute Digital globalization: The new era of global flows GLOBAL CONNECTEDNESS CAN ALSO BE VIEWED THROUGH THE LENS OF CITIES, PROVINCES, AND REGIONAL BLOCS The MGI Connectedness Index assesses how countries are participating in global flows— but they are not the only actors in the global economy. Cities, regions within countries, and broader blocs of countries are connecting with the global economy in myriad ways and to varying degrees. Looking through these various lenses provides unique insights into the new patterns of globalization. There are eight truly global cities, but only two are in emerging countries Cities are the real engines of the world economy and serve as major waypoints for global flows. Previous MGI research suggested that just 600 cities will generate some two-thirds of world GDP by 2025.79 Cities with major ports can become important hubs in the global flow of goods. Urban centers with significant Internet infrastructure can play the same kind of role for data and communication flows. Large cities in countries with low barriers to immigration can attract people from around the world. Acting as a waypoint generates significant economic output and high-quality jobs, and it helps a city accumulate knowledge, skills, and talent, with positive spillover effects on its broader economy. Once a city has established itself as a waypoint for a particular flow, other economic activity tends to coalesce or co-locate along with it. A city that establishes itself as a financial hub, for instance, is likely to attract insurance companies and other professional service firms. One that acts as a major waypoint for people and goods traffic through a major airport will become a magnet for logistics and distribution companies. Amsterdam invested in the Amsterdam Internet Exchange (AMS-IX) in the early days of the Internet and continues to be one of the largest data hubs in the world. Since it offers faster connection speeds than any other European city, it has attracted other types of high-tech businesses. Unfortunately, data on global flows are not available at the city level. However, we have obtained data that serve as proxies for each of our five global flows. Container port volumes approximate goods flows; airport passenger volumes serve as a proxy for goods, service, and people flows; the ranking of cities in the Global Financial Centers Index by the Z/Yen Group provides an indication of financial flows; the number of foreign-born residents in a city measures people flows; and Internet bandwidth approximates data flows. Our last report found that the world had only eight truly “global cities,” defined as ranking in the top 25 in at least four of the five major flows: New York, London, Tokyo, Los Angeles, San Francisco, Singapore, Hong Kong, and Dubai. This year, using updated data, we find that Tokyo drops off the list due to a decline in goods trade, while Shanghai takes its place (Exhibit 25). In fact, Chinese cities generally rose in the rankings, while a number of European cities (such as Bremen, Hamburg, Frankfurt, Brussels, Geneva, and Madrid) fell. 79 Urban world: Cities and the rise of the consuming class, McKinsey Global Institute, June 2012. New York, Los Angeles, San Francisco, London, Singapore, Shanghai, Hong Kong, and Dubai are the world’s truly global cities. 64 McKinsey Global Institute 3. How countries, cities, and regions are connecting Exhibit 25 Only eight of the world’s major cities are hubs for at least four of the five major flows SOURCE: Lloyd’s List; Containerisation International; Airports Council International; Global Financial Centers Index; Migration Policy Institute; TeleGeography, Global Internet Geography; McKinsey Global Institute analysis City participation in major flows by rank and change over previous year in each flow1 1 Metropolitan areas with at least 1 million foreign-born residents. Exact foreign-born population of Jeddah not known, so it is included at the bottom of the list. 2 Rankings come from different years: ports (2014), airports (2014), financial centers (2014), migration (2011), and online traffic (2015). Rank2 Goods Goods, services, people Financial People Data and communication 1 Shanghai Atlanta London New York Frankfurt 2 Singapore Beijing New York Los Angeles London 3 Shenzhen London Hong Kong London Amsterdam 4 Hong Kong Tokyo Singapore Hong Kong Paris 5 Ningbo Los Angeles Tokyo Toronto New York 6 Busan Dubai Seoul Paris Los Angeles 7 Guangzhou Chicago Zurich Miami Miami 8 Qingdao Paris Toronto Sydney Stockholm 9 Dubai Dallas/Fort Worth San Francisco Chicago San Francisco 10 Tianjin Hong Kong Washington, DC Singapore Singapore 11 Rotterdam Frankfurt Chicago San Francisco Hong Kong 12 Port Klang Jakarta Boston Melbourne Tokyo 13 Kaohsiung Istanbul Geneva Moscow Moscow 14 Dalian Amsterdam Frankfurt Houston Milan 15 Hamburg Guangzhou Sydney Dubai Vienna 16 Antwerp Singapore Dubai Riyadh Washington, DC 17 Xiamen Denver Montreal Washington, DC Hamburg 18 Tanjung Pelepas New York Vancouver Dallas Beijing 19 Los Angeles Shanghai Luxembourg Jeddah Marseille 20 Long Beach Kuala Lumpur Osaka Copenhagen 21 Laem Chabang San Francisco Shanghai Brussels 22 Tanjung Priok Bangkok Qatar Warsaw 23 Ho Chi Minh City Incheon Shenzhen Shanghai 24 Bremen Charlotte Busan São Paulo 25 New York Las Vegas Tel Aviv Madrid 65McKinsey Global Institute Digital globalization: The new era of global flows It is notable how few cities in the developing world appear in the top 25 across multiple flows, although some do play major roles in one individual flow. Only two of the eight global cities are in emerging economies (Dubai and Shanghai). Five cities in advanced economies (Tokyo, Paris, Chicago, Frankfurt, and Washington, DC), but none in emerging economies, appear in the top 25 for three of the five dimensions. Four cities in emerging countries (Beijing, Guangzhou, Shenzhen, and Moscow) appear in the top 25 for two types of flows. The urban giants of the developing world have room to expand the ways in which they participate in global glows, especially as the world’s economic center of gravity shifts in their direction. Examining the global connectedness of states and provinces reveals large variations within countries Within countries, the global connectedness of different states and provinces may vary dramatically. For several large countries (China, Germany, the United Kingdom, and the United States), we have obtained data on state-level participation in global flows of goods and migration. China has the most striking variation in global goods flows across its provinces (Exhibit 26). Only six of its 34 administrative divisions account for some three-quarters of its foreign trade in goods. 80 These coastal provinces, plus the areas around Beijing and Shanghai, are hubs of cross-border activity. But the interior of China remains largely disconnected from the world. The United States also has significant variation in how states participate in the global goods trade: California and Texas rank highest, reflecting ports in California and the oil sector in Texas. By contrast, industry is spread widely across multiple provinces in Germany and the United Kingdom, and so are their global flows. Goods flows relative to GDP for Germany’s most connected state (North Rhine–Westphalia) were 8.7 times the median, while China’s most connected state (Guangdong) stood at 26 times the median. Evaluating how states and provinces would stack up in the global rankings against nation- states underscores their range of performance and the global role that some play. China’s booming coastal province of Guangdong, for instance, is a major manufacturing and trade hub that would rank sixth globally in terms of goods flows—above the United States. Shanghai and Beijing would top Japan and Italy in goods flows (Exhibit 27). California would rank fourth in the world for people flows, above the United Kingdom, France, and the UAE. At the other end of the spectrum, South Dakota would rank 261st globally in goods flows, below Grenada, Samoa, and Cape Verde. These types of extremes are not apparent in Germany, however. Germany’s most connected states for flows of goods and people rank only 44th and 33rd, respectively, in global terms—but Germany as a whole ranks second in goods flows and third in people flows, reflecting the fact that all of its regions actively participate in global flows. Building global connections does not have to be left to national policy makers. Local business and government leaders everywhere have the ability to forge their own global ties directly—and digital globalization gives them a multitude of new ways to pursue these opportunities. 80 China’s administrative divisions include provinces, municipalities, autonomous regions, and special administrative regions. 66 McKinsey Global Institute 3. How countries, cities, and regions are connecting Exhibit 26 Within countries, regional connectedness varies greatly SOURCE: McKinsey Global Institute analysis Regional goods trade connectivity score, 2014 Higher number indicates greater connectedness NOTE: Countries not to scale. 0–10.0 10.1–20.0 20.1–40.0 40.1–100.0 United States China California Texas United Kingdom West Midlands East of England South East Germany Hamburg Niedersachsen North Rhine- Westphalia Baden- Württemberg Bavaria Guangdong Shanghai Beijing Jiangsu Zhejiang 67McKinsey Global Institute Digital globalization: The new era of global flows Exhibit 27 South Korea United Kingdom Switzerland 44.1 Guangdong (China) Shanghai (China)1 Mexico 25.0 France 28.6 23.7 Beijing (China)1 Canada 24.5 25.8 Thailand 24.4 Japan 24.5 29.6 28.4 Malaysia Italy 30.0 United States 45.9 United Arab Emirates 46.9 90.9 Belgium 100.0Singapore China Netherlands 87.8 51.3 78.4 Germany 92.6 23.6 16.7 Austria Baden–Württemberg (Germany) 16.5 16.3 15.4 Poland 16.7 16.7 Hungary Russia Jiangsu (China) Spain 28.4 14.7 Saudi Arabia 16.2 India 19.7 Vietnam 20.8 18.4 North Rhine–Westphalia (Germany) Texas (United States) 11.2 Bavaria (Germany) 13.8 14.6Slovak Republic Czech Republic If states and provinces are ranked among countries, Guangdong would be sixth globally in goods flows and California would rank fourth in people flows Connectedness index score, goods and people flows, 2014 Goods People SOURCE: McKinsey Global Institute analysis Qatar 21.5 Mexico 22.3 17.0Netherlands 14.8 New York State (United States) 18.3 Austria 21.4 30.6 Italy Australia 24.4 Spain Kuwait Switzerland 31.4 25.6 26.6 Ukraine 36.7 Jordan 33.7 Canada 35.1 Singapore 51.2 100.0 49.9 45.7 France United Kingdom 43.3 Russia Kazakhstan 44.1 Saudi Arabia 81.7 39.9 United Arab Emirates Germany United States California (United States) 40.8 37.0 Malaysia Lebanon 11.5 11.5Croatia 12.0 New Zealand London (United Kingdom)1 Israel Portugal Texas (United States) 10.9 Florida (United States) 12.9 13.6 11.4 12.3 12.6 12.3 12.7 Belarus Ireland North Rhine–Westphalia (Germany) State/region Country 1 Metropolitan areas considered as states/provinces by national authorities. 68 McKinsey Global Institute 3. How countries, cities, and regions are connecting A great deal of global trade circulates among neighbors Connecting with the world starts close to home. As described in Chapter 1, we find a significant share of each flow circulates within geographic regions rather than across long distances. However, this pattern varies around the world. In goods trade, the highest share of intraregional trade is in Europe (61 percent). Countries including Croatia, the Czech Republic, Hungary, Latvia, Norway, Portugal, and Romania conduct most of their trade with their European neighbors (Exhibit 28). Some 45 percent of East Asia’s trade is intraregional, followed by North America (which includes Mexico) at 42 percent. The corresponding share is far lower in Africa, Latin America, and South Asia, where most countries score far lower on the MGI Connectedness Index. This indicates a significant opportunity for developing countries to increase their participation in flows by trading with their neighbors. One way to facilitate greater intraregional trade is to form an economic community or trade bloc. Academic research shows that being part of a formal trade bloc—that is, a region defined by an intergovernmental agreement to reduce or eliminate regional barriers to trade—increases both the level and the impact of global goods trade.81 However, our data show that the intensity of intraregional trade within the world’s biggest trading blocs varies widely. Exhibit 29 shows the share of goods flows that occurs within these established regions. After the North American Free Trade Agreement (NAFTA) was established, the share of goods trade between the United States, Canada, and Mexico peaked at 45 percent of their total in 2000. However, that share had fallen back to 42 percent in 2014 as these countries all began to trade more actively with China. China now accounts for 13 percent of trade with NAFTA countries. Trade between the Association of Southeast Asian Nations group of countries remains far lower than trade within the European Union and NAFTA, at just 24 percent of their value of total trade in 2014, somewhat higher than the 16 percent share in 1990. Recognizing the opportunities at stake, ASEAN’s leaders are working to implement an ambitious ASEAN Economic Community (AEC) integration plan to harmonize many administrative procedures and regulatory standards among member states, creating an open market of 600 million consumers and a more seamless production base.82 The East African Community (EAC) enacted a free trade zone in recent years to eliminate all tariffs among its five member countries (Burundi, Kenya, Rwanda, Tanzania, and Uganda). However, its intraregional trade stands at only 10 percent. Interestingly, EAC trade with China is higher than even ASEAN’s trade with China, at 17 percent compared with 15 percent, despite ASEAN’s proximity. 81 The World Trade Organization has investigated preferential trade agreements and their effects in detail. These bilateral agreements have grown in importance as multilateral agreements have become more difficult to establish and negotiate. See World trade report 2011: The WTO and preferential trade agreements: From co- existence to coherence, World Trade Organization, 2011. 82 Southeast Asia at the crossroads: Three paths to prosperity, McKinsey Global Institute, November 2014. Developing countries have a significant opportunity to increase their participation in global flows by trading with their neighbors. 69McKinsey Global Institute Digital globalization: The new era of global flows Increasing intraregional trade among developing country trade blocs is a clear priority—one that could be a stepping-stone to increasing their overall connectedness and capturing the benefits of scale, specialization, and competition. Although the impact of distance on all forms of global flows is falling as transportation and communication costs decline, economic ties among countries in the same region are still the norm. Bolstering regional ties, particularly in Africa and Latin America, is an opportunity to boost flows—and to raise GDP, as we will discuss in the next chapter. Exhibit 28 Some European countries trade predominantly with their neighbors Intraregional goods trade in Europe (includes imports and exports), 20141 % share of total trade 1 Includes EU-28 and select W estern European countries including Andorra, Iceland, Norway, and Switzerland. SOURCE: UNCTAD; McKinsey Global Institute analysis 40–50 50–60 60–70 70+ Portugal Iceland Lithuania Latvia Estonia Finland Sweden Norway Denmark United Kingdom NetherlandsIreland Germany Poland Belgium Luxembourg Italy France SwitzerlandSpain Greece Malta Cyprus Czech Republic Austria Hungary Bulgaria Romania Croatia Slovak Republic 70 McKinsey Global Institute 3. How countries, cities, and regions are connecting ••• Measuring the connectedness of different countries, cities, and regions around the world illustrates how far globalization has advanced—and how much further it could still progress. Much of the momentum behind globalization is still being driven by a handful of highly connected countries, and much of the world has only just begun to tap into global flows. The next chapter will provide detailed econometric analysis of the value of connectedness, underscoring the importance for emerging economies to pursue catch-up growth. Exhibit 29 24 61 15 42 45 13 SOURCE: UNCTAD; IMF; McKinsey Global Institute analysis The EU is the most integrated of the world’s major trading blocs Trade within and outside of trading blocs, 2014 % share of total goods trade Extra-regional (excluding China) China Intraregional 34 61 6 NAFTAEU-28 73 17 10 ASEAN EAC (East African Community)1 100% = $5.6 trillion 100% = $11.8 trillion 100% = $2.5 trillion 100% = $0.05 trillion 1 Comprises Burundi, Kenya, Rwanda, Tanzania, and Uganda. NOTE: Numbers may not sum due to rounding. 71McKinsey Global Institute Digital globalization: The new era of global flows © Getty Images © Getty Images The economic impact of globalization is growing as the web of connections between countries and companies grows broader and deeper. As vast populations in emerging economies gain a new foothold in the middle class, more of the world’s business involves cross-border flows. Digital flows of data, information, and communication allow ideas and innovation to ricochet around the world; they enable companies to bring together the best inputs and talents to create higher-quality goods and services. Perhaps most important, global flows expose companies to more competition and best practices, spurring them to improve performance and productivity. To measure the economic impact of global flows, we have undertaken extensive econometric analysis, building on our previous report and applying improved data and more sophisticated methodology. We use data on the cross-border inflows and outflows of goods, services, finance, people, and data for 97 countries from 1995 to 2013 to assess the impact of flows on GDP and productivity.83 Our analysis breaks new ground by testing the impact of all types of flows together and by analyzing the impact of gross inflows and outflows, rather than looking more narrowly at trade balances. We find strong evidence that country participation in global flows has both short-term (cyclical) and long-term (structural) impact on GDP. Our results indicate that over the past decade, global flows have raised world GDP by roughly 10 percent over what would have resulted in a world in without any flows. In 2014 alone, this impact was equal to $7.8 trillion in global GDP. Flows of goods and FDI, the two categories most heavily associated with the last era of globalization, account for about half of this. But data flows, worth some $2.8 trillion in 2014, now exert a larger impact on global GDP than the flow of goods. Their role in the global economy has expanded with astonishing speed. After all, the world’s trade networks have developed over the course of centuries while global data flows were negligible as recently as 15 years ago. Chapter 3 showed that each type of flow remains concentrated among a short list of leading countries, and lagging countries are catching up only very slowly. Here, we find that some lagging countries could have boosted their current GDP by 50 percent or even more by accelerating their participation in global flows over the past ten years. They can still realize tremendous opportunities by targeting ways to build deeper ties to the global economy. 83 Our data end in 2013 because that is the most recent year for which a large set of countries report inflows and outflows of migrants. Both inflows and outflows support growth by circulating ideas, research, technologies, talent, and best practices around the world. $7.8T contribution to world GDP in 2014 by all types of global flows combined 4. GLOBAL FLOWS BOOST ECONOMIC GROWTH 74 McKinsey Global Institute 4. Global flows boost economic growth Overall, our analysis provides strong evidence of the economic value of openness—and the benefits are much broader and more nuanced than a simple accounting of net exports. Consumers, for instance, gain purchasing power from trade in goods, which delivers a wider variety of products at lower prices. One study found that middle-class Americans gain more than a quarter of their purchasing power from trade.84 Both inflows and outflows of goods, services, finance, people, and data are important as they expose economies to ideas, research, technologies, talent, and best practices from around the world. Countries that participate in global flows can specialize in what they produce, realize economies of scale, and invite competition, which spurs both efficiency and innovation in local industries. These benefits have been recognized since they were outlined by economist David Ricardo in the early 19th century—but today there is a difference. As they transform traditional flows and enable broader participation, digital technologies are amplifying these effects (Exhibit 30). 84 The economic benefits of US trade, Executive Office of the President, May 2015. Comparative advantage and specialization Broader participation Accelerated information flows Truly global market scale Economies of scale Increased competition boosting efficiency Knowledge diffusion Capital deepening and widening Human capital development HIGHER PRODUCTIVITY MORE INNOVATION INCREASED GDP D IG IT IZ A TI O N How globalization increases GDP Exhibit 30 75McKinsey Global Institute Digital globalization: The new era of global flows GLOBAL FLOWS HAVE INCREASED CURRENT WORLD GDP BY AT LEAST 10 PERCENT, ADDING $7.8 TRILLION IN 2014 Our ongoing research into global flows aims to create a clearer view of a diffuse and evolving phenomenon. A key part of that is providing a better understanding of how flows affect economic growth. Our last report established that global flows contribute to GDP growth and that countries that are more central in the networks of global trade in goods benefit disproportionately, irrespective of the direction of the flows.85 It also found that data flows are an increasingly important driver of economic growth. For this report, we set out to sharpen our picture of global flows by refining our econometric methodology and using updated and improved data. For example, we previously used international telephone calls as a proxy for data flows. But this arguably measured only the communication aspect of data flows rather than the way they transmit information and enable transactions. Now we are able to track used cross-border bandwidth, which is a much closer measure of cross-border data flows. To measure people flows, we use the stock of migrants, both inbound and outbound. Our analysis also separates the impact of foreign direct investment and other types of financial flows. We make this distinction since a large body of academic literature finds that while FDI has a clear positive impact on growth, the evidence for the impact of other types of financial flows (cross-border lending and portfolio investments of equity and debt) is mixed.86 The technical appendix of this report contains a detailed description of our methodology. In short, we use a co-integrated, two-step error-correction model that enables us to examine without bias both the short- and long-term impacts of global flows on growth. To measure the one-way effect from flows on GDP growth, we use lagged covariates as instrumental variables to correct for possibly endogeneity. Our methodology allows us to control for unobserved country-specific effect and for correlation among the variables. We also test the impact of flows on productivity compared with the utilization of capital and labor. Finally, using two different measures of network centrality, we test how a country’s position in the network of global flows contributes to its growth. Our model considers how all types of global flows act in concert to generate growth Our results provide robust evidence that goods, data, and FDI flows have both a positive short-term impact on GDP and a positive long-term structural impact (Exhibit 31). Our results for migration flows are surprising: we find a negligible or even negative impact on GDP growth for developing economies, possibly reflecting the loss of skilled labor (commonly referred to as “brain drain”) or the difficulty that some have in absorbing a large influx of refugees or migrants. We find, however, that migration flows have a positive impact on productivity for advanced economies, which is consistent with other literature.87 85 Global flows in a digital age: How trade, finance, people, and data connect the world economy, McKinsey Global Institute, April 2014. 86 See, for example, Joshua Aizenman, Yothin Jinjarak, and Donghyun Park, “Capital flows and economic growth in the era of financial integration and crisis, 1990–2010,” Open Economies Review, volume 24, issue 3, July 2013. 87 For instance, see Ekrame Boubtane, Jean-Christophe Dumont, and Christophe Rault, Immigration and economic growth in the OECD countries, 1986–2006, IZA discussion paper number 8681, November 2014. Cross-border data flows added some $2.8 trillion to world GDP in 2014, surpassing the impact of the global goods trade. 76 McKinsey Global Institute 4. Global flows boost economic growth The magnitude of the increases in both GDP level and growth generated by global flows is quite striking. In one specification of the model, using the value of flows for each country normalized by their size, we find that over a decade, all types of flows acting together have raised world GDP by 10.1 percent over what would have resulted in a world without any cross-border flows. This amounted to some $7.8 trillion in 2014 alone. Flows of goods trade and FDI account for roughly half of this growth effect (raising world GDP by 3.5 percent and 1.6 percent, respectively; Exhibit 32). Data flows alone directly raised world GDP by 3.0 percent. In addition, however, they also enable trade flows, FDI, and even people flows. If we account for both the direct and indirect impact of data flows, we find they exert an even larger impact on world GDP than physical trade flows (see Box 3, “Valuing cross-border data flows”). People flows have raised the level of world GDP by 2.0 percent.88 Viewed another way, the impact of flows on global GDP may even be larger than 10.1 percent. In a second econometric model specification, we used data on each country’s score in the MGI Connectedness Index for each flow, rather than the value of inflows and outflows. This corrects for the fact that cross-border flows may be lower relative to GDP for a very large economy (such as the United States) simply due to the huge size of its domestic markets. Using connectedness scores instead of actual flows, we find that the contribution 88 As noted above, our econometric results show that migration flows have a negative elasticity with respect to GDP at a global level (despite the positive impact of immigration on productivity in advanced economies). We find an overall positive effect on lifting global GDP in the period we analyzed (2003–13) because global migration flows actually declined slightly over the period relative to world population, thus producing a net positive effect on global growth. Although people flows include growing business travel, tourism, and study abroad in reality, our model considers only long-term migration due to a lack of historical data on other types of people flows for many countries around the world. Exhibit 31 Short-/long-term impact Name of variable Granger causality with real GDP Expected sign of coefficient Estimated sign of coefficient FDI Two-way Positive/positive Positive/positive Goods trade flow Two-way Positive/positive Positive/positive Immigration Two-way Positive/positive Insignificant/negative1 Data flows Two-way Positive/positive Positive/positive Services trade flow Two-way Positive/positive Extended due to correlation with FDI Fixed capital stock n/a Positive/positive Positive/positive Employment n/a Positive/positive Positive/positive Average years of education n/a Positive/positive Insignificant/negative SOURCE: McKinsey Global Institute analysis The coefficients from our econometric model have the expected sign Flow variables Dependent variable (Log) Real GDP Independent variables (Log) 1 Migration flows are negligible or slightly negative at the global level, possibly due to the loss of skilled labor in developing countries or the difficulties of absorbing a large influx of refugees or migrants. However, migration flows have a positive impact on productivity in advanced economies. 77McKinsey Global Institute Digital globalization: The new era of global flows of flows to world GDP in 2014 could be as high as 18.7 percent, or $14.4 trillion—meaning that one in every six dollars of value added in the world comes from cross-border connections.89 89 See technical appendix for the model results. Exhibit 32 Our econometric model shows that global flows account for approximately 10 percent of global GDP output 3.5 3.0 2.0 1.6 10.1All flows FDI Migration2 Data flows Goods trade SOURCE: McKinsey Global Institute analysis 1 Includes inflows and outflows data for 139 countries in MGI Global Flows model; see technical appendix for more details. 2 Global migration flows declined slightly from 2003 to 2013, resulting in a positive impact despite a negative coefficient. Migration flows are negligible or slightly negative at the global level, possibly due to the loss of skilled labor in developing countries or the difficulties of absorbing a large influx of refugees or migrants. However, migration flows have a positive impact on productivity in advanced economies. NOTE: Numbers may not sum due to rounding. Impact on GDP, 20141 $ trillion 2.7 2.3 1.5 1.3 7.8 Long-term impact on level of GDP1 % Box 3. Valuing cross-border data flows Flows of goods, services, and finance are measured in dollar terms that reflects the monetary value of the transactions. Our data on cross-border data flows, in contrast, are measured in volume, or terabits per second. To accurately compare the scale of these flows, we need a common unit of measurement. We have therefore attempted to calculate cross-border data flows in value-added terms. Our methodology is admittedly imprecise by its nature and requires some assumptions. Still, the results are striking even if only directionally correct. We begin with our econometric analysis, which finds that the direct impact of data flows raises world GDP by 3.0 percent annually. This equates to $2.2 trillion in 2014. However, we also know that cross-border data flows enable other types of flows. Consider that cross-border e-commerce now accounts for 12 percent of global trade. In addition, digital communication and platforms likely enable nearly every trade transaction on the planet today. In addition, data flows allow service exports to be delivered digitally, and digital transactions and communication enable foreign direct investment. Even people flows may benefit from digital platforms and the ability to book travel online and research foreign destinations. We make the conservative assumption that data flows account for 12 percent of the value created by other types of flows. This contributes an additional $0.6 trillion to their 2014 impact. Adding these indirect effects to the direct effects found in our model, we find that cross-border data flows may have raised world GDP by roughly $2.8 trillion in 2014. This surpasses the $2.7 trillion impact of the global goods trade. In just a decade, global data flows have generated as much economic value as trade networks formed over the course of centuries. 78 McKinsey Global Institute 4. Global flows boost economic growth Our research is unique in that it considers the impact of all types of flows acting together. Yet our estimates of the impact of flows on GDP are broadly consistent with other academic findings (see the technical appendix for a complete literature review), even if other studies typically focus on one type of flow for a set of countries. This is an important distinction, because we know that flows are correlated with one other. We performed a principal component analysis on how flows move together. The results indicate that one component captures more than 60 percent of all flows. This means that examining the impact of only one type of flow may overestimate its impact on GDP by picking up observed effects from other flows. Nonetheless, our findings for the impact of each individual flow are in line with academic findings. In general, our estimates are smaller since we correct for the effect that one flow may have on another. For example, Wacziarg posits that every increase in trade of ten percentage points relative to GDP adds 0.7 percent to GDP, while our model shows an increase of 0.5 percent.90 Har et al. find that every increase in FDI of ten percentage points relative to GDP raises GDP by as much as 0.5 percent, vs. 0.4 percent in our model.91 Boubtane et al. observe that every ten percentage point increase in immigration raises GDP by one percentage point in OECD countries.92 Our model shows a negligible or even negative impact at the global level, but it similarly finds that migration has a positive impact on productivity in advanced economies. We have not found a study that measures the effect of cross-border data flows on GDP. Due to the difficulties of collecting data on this subject, most of the literature has assessed the GDP impact of increasing Internet penetration. For example, Meijers finds that a ten percentage point increase in Internet penetration leads indirectly to a 0.17 percentage point increase in economic growth.93 Our model shows that a ten percentage point increase in data flows relative to global population raises GDP by 0.2 percent. The economic impact of flows comes mainly from raising productivity, and data flows are central to this effect Global flows can raise GDP growth either by increasing productivity or by increasing the amount of capital and labor used in production. We find both effects at work, although different flows act through different channels, and the productivity effect has by far the largest impact on GDP. Traditional flows of goods and FDI raise productivity in an economy but do not necessarily increase the capital and labor inputs used. This suggests that in the long term, participation in flows promotes more efficiency and innovation, perhaps stemming from increased competition or the faster spread of best practices. We find that data flows support both productivity improvement and increased capital and labor inputs. This indicates that so far, data flows and digitization have raised net employment within countries rather than reducing it, contrary to conventional wisdom. The benefits of data flows are not due to replacing workers (see Box 4, “The impact of global flows on employment”). 90 Romain Wacziarg, “Measuring the dynamics gains from trade,” World Bank Economic Review, volume 15, number 3, October 2001. 91 Wai Mun Har, Kai-Lin Teo, and Yee Kar Man, “FDI and economic growth relationship: An empirical study on Malaysia,” International Business Research, volume 1, number 2, April 2008. 92 Ekrame Boubtane, Jean-Christophe Dumont, and Christophe Rault, Immigration and economic growth in the OECD countries, 1986-2006, IZA discussion paper number 8681, November 2014. 93 Huub Meijers, “Does the Internet generate economic growth, international trade, or both?” International Economics and Economic Policy, volume 11, issue 1, February 2014. 79McKinsey Global Institute Digital globalization: The new era of global flows Box 4. The impact of global flows on employment Our own analysis, like that of academic researchers, shows that trade, foreign direct investment, people flows, and data flows have a net positive effect on growth and prosperity in the long run. As countries specialize in what they do best, global GDP rises and employment growth should follow.1 But globalization also delivers an element of creative destruction as it exposes local industries to international competition and disruptive business models. This is ultimately a healthy dynamic that spurs efficiency and innovation, but it may displace some local industries and workers in the process. In the long run, countries that participate in global flows will find new channels for growth, and workers who lose their jobs in one industry should find opportunities elsewhere. Yet this process does not always play out neatly and quickly. Workers in a shrinking industry must gain new skills to be employed in other sectors—and those opportunities may not be in the same geographies, requiring them to relocate. This can have profound effects on entire communities that lose traditional industries to global trade and competition. Recent research has found, for instance, that increased US goods trade with China since 2000 resulted in the loss of thousands of manufacturing jobs in the United States and that the effects have persisted for more than a decade.2 Providing support to workers who are affected—and to broader local communities that suffer when industries are lost—has too rarely been treated as an urgent policy priority. But the societal cost of neglecting this issue grows over time and spurs negative sentiments toward globalization. Traditional labor market policies and training systems in most countries are not prepared to deal with large-scale, rapid changes.3 Workers displaced by trade (and similarly by automation) will need clearer paths to new roles. This means creating widespread access to accelerated training programs that will help them acquire skills that are in demand. A range of policy options are possible, but all governments must address the issue. One possible policy response is wage insurance.4 Germany may offer a useful model for other countries as well. It is one of the world’s most connected countries, ranking fourth in our global index. But it has avoided widespread unemployment, even at the height of the global recession, by providing income support, paying companies to retain workers, and taking a proactive approach to labor market reforms and reemployment services.5 Interestingly, our econometric analysis suggests that global data flows increase employment in the long run. While the goods trade changes the location of some production activities, data flows enable innovation, remote work, and new types of economic activity that did not exist before. It is too early to say definitively how cross-border data flows will affect employment in the future, particularly given the unpredictable nature of technology innovation, but digital platforms could be a welcome source of opportunity to find work for some individuals. They could also help to deliver some of the educational and training programs that individuals will need to reinvent themselves with new skills. 1 For a discussion of when this may not be the case, see Paul Samuelson, “Where Ricardo and Mill rebut and confirm arguments of mainstream economists supporting globalization,” Journal of Economic Perspectives, volume 18, number 3, summer 2004. 2 David H. Autor, David Dorn, and Gordon H. Hanson, “The China syndrome: Labor market effects of import competition in the United States,” American Economic Review, volume 103, number 6, 2013, and David H. Autor, David Dorn, and Gordon H. Hanson, The China shock: Learning from labor market adjustment to large changes in trade, NBER working paper number 21906, January 2016. 3 For a deeper discussion of labor market inflexibility and possible responses to it, see A labor market that works: Connecting talent with opportunity in the digital age, McKinsey Global Institute, June 2015. 4 See, for example, Lori G. Kletzer, “Why the US needs wage insurance,” Harvard Business Review, January 25, 2016. 5 Marco Caliendo, Income support systems, labor market policies and labor supply: The German experience, IZA discussion paper number 4665, December 2009. 80 McKinsey Global Institute 4. Global flows boost economic growth COUNTRIES ON THE PERIPHERY HAVE THE MOST TO GAIN FROM CROSS-BORDER DATA FLOWS Finally, we assess how a country’s position in the global network of flows changes the economic impact. We do this using two different measures of network centrality: one that measures the number of a country’s bilateral connections compared with the total number possible, and one that measures how connected a country’s trading partners are (a concept known as eigenvector centrality). Our last report tested centrality only within trade flows, given data limitations within other types of flows. We found that the benefits to GDP growth were up to 40 percent higher for countries that were more central in the global network of goods trade—with more bilateral partners and partners that were themselves more connected—than for countries with only a few trading partners.94 This finding showed that it is better to have a broader network of connections and to connect with countries that are more central to the network than to trade solely with a few neighbors. This report extends our centrality analysis to cross-border data flows—and the findings do not show the same effects as in trade flows. Data flows are still in a nascent stage, with links between countries that are less dense and have less reciprocity. In the case of data flows, we find that the benefits to GDP growth for countries at the periphery of the global network are actually higher than for countries at the center of the network. Periphery countries may be using their exposure to data flows to enable broader participation. Moreover, data flows offer access to all the world’s knowledge, information, and innovations. For economies that have been relatively disconnected, the arrival of new digital platforms can have a bigger ripple effect on trade in goods and services than it might have in advanced economies that already have more extensive trade links in place. Our analysis shows that 15 to 30 percent of the GDP impact of data flows comes from consumers, while the remainder comes from B2B linkages within value chains. As policy makers in many countries consider how their countries should participate in a more digital global economy, many are seeking to create the “next Silicon Valley.” Others are erecting barriers to global digital platforms to allow domestic platform providers to grow. But we find that countries benefit from receiving cross-border digital flows as well as producing them. In other words, countries do not need to transform themselves into digital content or platform producers to benefit from data flows. 94 Global flows in a digital age: How trade, finance, people, and data connect the world economy, McKinsey Global Institute, April 2014. Countries at the periphery of the world’s digital networks stand to gain even more than those at the center. 81McKinsey Global Institute Digital globalization: The new era of global flows SOME LAGGING COUNTRIES COULD INCREASE GDP BY 50 PERCENT IN THE LONG TERM BY RAISING PARTICIPATION IN FLOWS With clearer evidence in hand that global flows drive GDP growth, it becomes apparent that there is an opportunity cost associated with limited participation—and closing the gap between leaders in global flows and laggards would create substantial economic value. To determine the size of this unrealized opportunity, we calculate the value that countries realized by increasing participation in each type of flow relative to the size of their economies from 2003 to 2013 (the latest year for which there are global data for all flows). We find that countries in the top quartile increased their flow of goods relative to GDP at an average of 3 percent annually, while goods flows grew at only 1 percent for the bottom quartile. The top-quartile countries increased FDI flows relative to GDP by 5 percent of GDP annually during this period, while those flows shrank by 8 percent annually for countries in the bottom quartile. A similar pattern holds for data flows and people flows. We calculate that if the bottom three quartiles of countries had increased participation in each of the flows at the same rate as the top quartile over the past ten years, global GDP would be some $10 trillion, or 13 percent, higher today. In other words, limited participation in global flows by many countries had a real cost to the world economy. Because few countries are consistently strong across all five flows, there is substantial potential value for many to capture. As discussed in Chapter 3, we find that most countries are relatively connected within one or two flows: for instance, Luxembourg in financial flows and services, or Belgium in goods and service trade. Only a few countries, such as Germany or the United States, rank highly across all flows. In addition, a large set of countries, mostly in the developing world, post low or barely average scores across all flows, and they are catching up to the most connected countries at a very slow pace. Some countries have made explicit moves to open their economies and participate more fully in global flows. For lagging countries, the potential payoff from a well-targeted strategy of opening may be substantial (Exhibit 33). If India, for instance, had accelerated its participation in all types of global flows to match the top-quartile countries over the past ten years, its GDP would have been $1.2 trillion higher (or 58 percent larger) by 2014. Despite its thriving business process offshoring sector, India ranks only 70th in the world for data flows—down from 64th in the previous edition of the index, indicating that other countries have increased data flows faster than India. According to the Internet and Mobile Association of India, the country passed the benchmark of 350 million Internet users in 2015, and penetration is continuing to grow. It will not be simple to create the digital infrastructure and skills development necessary to bring the rest of India online, but the government’s Digital India initiative aims to accelerate the process of connecting the country’s vast rural population to the world economy. Limited participation in global flows by many countries has had a real economic cost. 82 McKinsey Global Institute 4. Global flows boost economic growth Similarly, Brazil could have added some $1.4 trillion to its GDP over the past ten years, making its economy 60 percent bigger by 2014, by accelerating its participation in all types of global flows. Previous MGI research has explored Brazil’s relative lack of global connectedness. Brazil has not capitalized on its proximity to the lucrative US consumer market, for instance. In 2012, even before the current commodity price decline, exports were equivalent to 13 percent of GDP, far below India (24 percent) and Mexico (33 percent). Brazil’s imports are also lower, at 10 percent of GDP, compared with 22 percent for India and 32 percent for Mexico. A number of barriers have constrained trade growth, including poor road and rail infrastructure, cumbersome procedures and inadequate capacity Exhibit 33 Lagging countries could realize enormous growth potential by increasing their participation in global flows SOURCE: McKinsey Global Institute analysis Output gap % of GDP >75
51–75
26–50
1–25
<1 No data 83McKinsey Global Institute Digital globalization: The new era of global flows at its ports, and high import tariffs and taxes. Brazil is particularly low on people flows, ranking only 125th in the world for migrants and travelers. Since 1999, the country has lost 38 percent of its share of South America’s inbound tourism and 30 percent of its share of world inbound tourism.95 These missed opportunities are not lost forever, however—and digitization is creating new ways for countries to participate. As growth in conventional flows of goods and finance flatten, the next wave of growth from globalization will come from data and information flows. Countries can increase their exposure to these flows by expanding Internet penetration and creating thoughtful frameworks that allow data to move both securely and freely across their borders. Conversely, academic studies point out that restrictions on data flows may lower GDP growth by one to two percentage points.96 Chapter 6 contains further discussion of these and other public policy implications. ••• Participating in global flows is not a panacea for the other factors that may dampen a country’s economic growth, such as an uncompetitive business environment, weak rule of law, or corruption. Still, countries that seal themselves off from global flows—and particularly data flows—are forgoing important sources of growth. Continuing to expand the network of global data flows is possible because this new version of globalization is not merely a zero-sum game in which countries compete for low-cost manufacturing: one country’s participation in data flows does not necessarily come at another’s expense; it can actually increase economic growth across the board. 95 Connecting Brazil to the world: A path to inclusive growth, McKinsey Global Institute, May 2014. 96 Matthias Bauer et al., The costs of data localization: Friendly fire on economic recovery, ECIPE occasional paper number 3/2014, May 2014. © Getty Images After expanding across borders in pursuit of new international markets and the advantages of scale, many major corporations now derive more than half of their revenue internationally. But along the way, they may have incurred a “globalization penalty.” Managing across multiple geographies with different cultures, languages, regulations, and tax regimes is no small challenge. It often involves going up against local competitors that may have deeper market insights and better ability to execute on their own turf. The costs of coping with complexity can take a toll on the bottom line as well as organizational health, making it harder to create a cohesive culture and strategy.97 Digitization can tame complexity, however, allowing large companies to manage their global operations in a leaner and more efficient way. Using digital platforms and tools effectively can enable companies to sell in far-flung but fast-growing markets while keeping virtual teams connected in real time. Companies have new ways to identify the best suppliers, inputs, and talent from around the world. The convergence of globalization and digitization means that the world is changing rapidly—and business leaders will need to reassess their organization, strategy, assets, and operations accordingly. The approaches that worked for going global even ten years ago may no longer be relevant. As digitization changes how companies think about what should be global and what should be local, many are reevaluating past decisions about their footprint and international market strategies. The successful global companies of the future will have a very different look than those of the past. A SUCCESSFUL DIGITAL GLOBALIZATION STRATEGY CONSIDERS SEVEN KEY DIMENSIONS Digital innovation has greatly expanded the toolbox companies can use to globalize their market reach and operations (Exhibit 34). Because of these new capabilities, business leaders have an opportunity to rethink organization and strategy—starting with the fundamental question of what kind of global footprint is optimal. For many years, globalization centered on building production facilities around the world to take advantage of low-cost labor, but digitization is changing that equation. Additionally, while in the last era of globalization products were significantly tailored to local markets, some companies are moving toward global products and global product launches. As global value chains shift, decisions about where to base production are becoming more nuanced than simply looking for the lowest labor costs. To compete effectively, companies will have to build new types of assets, including the right arsenal of digital capabilities—and meeting that requirement will require a more global approach to finding and deploying the best talent. Finally, digital globalization calls for building more resilience into organizations. 97 Martin Dewhurst, Jonathan Harris, and Suzanne Heywood, “Understanding your ‘globalization penalty,’” McKinsey Quarterly, July 2011. The successful global companies of the future will have a very different look than those of the past. 5. COMPETING IN A DIGITAL GLOBAL LANDSCAPE 86 McKinsey Global Institute 5. Competing in a digital global landscape Do your footprint and organizational structure make sense in a more digital world? Companies once expanded globally by building a replica of their entire firm in each country where they operated, complete with human resources, finance, sales and marketing, and product development departments. Each country operation mirrored the vertical structure of the parent entity. But now digital technology is expanding the options for how to organize a global company. Companies can establish global hubs for some functions and combine those structures with smaller dispersed sites and sales offices around the world. This option captures the advantages of efficiency and scale for some functions but maintains sales and marketing teams in proximity to local customers. The choices companies make about their footprint and how to manage it have never been wider—and the payoff, in terms of both reach and agility, has never been bigger. Many multinationals are centralizing global functions and back-office operations. In human resources, for example, self-service digital platforms can draw on data flows between countries to handle many overarching issues, while regional managers can retain discretion over decisions such as hiring. Singapore-based Flextronics migrated its fragmented human resources systems for 200,000 workers in 25 countries into one global platform that Exhibit 34 SOURCE: McKinsey Global Institute analysis Cross-border implications of digitization Flow type Data Goods Services Finance FDI Remote monitoring Remote tracking Remote maintenance Supply-chain management Remote inventory management Supplier management Access to global markets Cross-border access to customers Cross-border access to labor Cross-border access to finance Business operations and strategy Centralized back-office operations Cross-border digital payments Real-time communications and collaboration Data sharing and analytics- driven decision making Digitization is transforming business models in ways that enable more cross-border activity 87McKinsey Global Institute Digital globalization: The new era of global flows automatically supports 14 languages.98 The Canada-based Four Seasons chain, which has 42,000 employees in hotels around the world, similarly moved to a globally scaled cloud-based HR system that offers organizational consistency where possible and local customization where necessary.99 Companies can also create virtual teams that span borders, using digital tools for remote collaboration (such as Box and Slack) and customer relationship management (such as Intercom and Zendesk). Unilever, for example, used technology solutions to streamline some 40 global service lines, including financial reporting, internal communications, market research, requisitions and payments, and HR. It created global, virtual delivery organizations with team members who meet via video conference. In the past, Unilever had more than 400 intranets spanning different countries, product groups, and functions—a structure that led to unnecessary expense and misaligned communications. These were replaced with one global intranet that is accessible in more than 20 languages.100 There is no one-size-fits-all solution that will work for every company, however. In R&D and product development, for instance, Apple concentrates its engineering and design talent in Cupertino, California, and brings foreign hires to work there. Google has gone another route, with engineering and design teams in major tech hubs around the world. Sometimes proximity to specialized talent pools can shape these decisions. Boeing maintains a design center in Moscow to take advantage of Russia’s abundance of aerospace scientists and engineers. Cisco established a global development center in Bangalore to tap into the local concentration of engineering talent. This center, Cisco’s largest facility outside the United States, houses R&D operations focused on developing new disruptive technologies and strategies for emerging markets. More than 1,000 of the company’s patents have been filed from India as a result.101 Footprint decisions can also be driven by market demand and expansion opportunities. As emerging economies build their health-care systems, virtually all major Western pharmaceutical companies have established R&D operations in Asia. AstraZeneca, for instance, is creating a new global hub for pharmaceutical development in Shanghai to complement its R&D centers in the United Kingdom and Sweden; it already has 11,000 employees in China.102 The main R&D hub for Novartis is co-located with its global headquarters in Switzerland, but the company also has research facilities in China, India, Japan, Singapore, and the United States. Pharmaceutical companies in emerging markets are similarly expanding their international R&D capabilities. India’s Sun Pharmaceuticals, for example, has R&D centers in Israel and the United States. In a more digitally connected age, some companies are breaking from the traditional model of having one global headquarters location. Lenovo is incorporated in Hong Kong and has headquarters in both Beijing and North Carolina; the company also bases some top executives and research centers in other major hubs across the globe. GM has a China office in Shanghai to serve what is now the world’s largest auto market, plus an international headquarters in Singapore to oversee operations in the rest of Asia, Africa, Australia, and the Middle East. Honda has established multiple independent manufacturing subsidiaries, 98 “Flextronics completes world’s largest deployment of a core HR system in the cloud,” Workday corporate press release, December 9, 2011. 99 Matthew Finnegan, “Four Seasons chooses Workday HCM cloud over ‘costly’ on-premises ERP systems to manage global workforce,” ComputerworldUK, October 20, 2015. 100 Pascal Visée, “The globally effective enterprise,” McKinsey Quarterly, April 2015. 101 Cisco India Overview, corporate fact sheet, 2016. 102 “AstraZeneca continues strategic investment in China to accelerate delivery of innovative biologics and targeted medicines,” corporate press release, December 16, 2015. See also “Innovating in China’s pharma market: An interview with AstraZeneca’s head of R&D in Asia and emerging markets,” McKinsey.com Insights & Publications, February 2012. 88 McKinsey Global Institute 5. Competing in a digital global landscape including Honda China, Honda of North America, and Honda Europe. These divisions are run locally, and each market determines which models to sell. Digitization also enables less capital-intensive business models. Companies that deliver digital goods and services can enter new international markets without establishing a physical presence at all. When Netflix created a subscription model for online streaming of video content in 2007, it gained the ability to add global customers without setting up full-fledged physical operations around the world (after receiving local regulatory approval). By the end of 2015, the company had broadened its international reach to more than 190 countries. As it did so, it closed all of its data centers and moved all of its streaming services onto the public cloud.103 Should you offer one product line around the world or customize for local markets? In many industries, companies that sell into a range of global markets have expanded their product portfolios with tailored offerings that appeal to local consumer preferences and are sensitive to their price points. South Korea’s LG, for example, markets original products in India, including appliances with programming menus in local languages, large washing machines for big families, and microwaves with one-touch “Indian menu” functions.104 Mondelez found that the Oreos beloved by Americans did not have the same appeal elsewhere, so it tweaked the cookie’s formulation, size, shape, packaging, and price points in China and India to appeal to local palates and preferences.105 In some industries, product tailoring is driven by local regulatory requirements or language differences. Food companies, for example, must meet one set of requirements for their products to be certified as halal in Malaysia and a different set of standards in Indonesia (and neither country’s requirements are in line with Saudi Arabia’s standards).106 But some companies eschew this strategy and offer truly global products that are the same everywhere in the world. Apple, for instance, offers just three models of its iPhone and iPad, all with consistent design and branding no matter where they are sold; settings can be reconfigured to change the language. Even its retail stores have the same design aesthetics the world over. Luxury brands typically take the same strategy. Brands such as Gucci, Burberry, and Prada position themselves as aspirational for newly affluent consumers in emerging markets and provide the same products and customer experience everywhere. Creating a single global product is not just for companies that target upscale customers. Now that social media allows everyone to see the latest celebrity trends, many consumers around the world want the same styles. Some mass market retailers, including H&M and Ikea, offer a consistent brand, store format, and core product selection worldwide but tweak a small share of the merchandise mix to reflect local differences. Digital products often follow the one-world model. Google, for example, has search, mapping, and e-mail products that are not designed with any particular set of regional customers in mind (although different languages are available). Facebook, Uber, Airbnb, and various e-commerce marketplaces have simply scaled up their digital platforms in country after country with limited customization. The success of this model has encouraged digital entrepreneurs to think about designing globally scalable products and expanding 103 Caroline Donnelly, “Netflix shuts down final data centre to go all-in on public cloud,” Computer Weekly, August 17, 2015. 104 Winning the $30 trillion decathlon: Going for gold in emerging markets, McKinsey & Company, August 2012. 105 See Jeff Beer, “Marketing to China: Oreo’s Chinese twist,” Canadian Business, November 22, 2012; and “Smart cookie,” case study from the Aditya Birla India Centre of London Business School, Business Today (India), March 31, 2013. 106 Southeast Asia at the crossroads: Three paths to prosperity, McKinsey Global Institute, November 2014. 89McKinsey Global Institute Digital globalization: The new era of global flows internationally much earlier in their life cycle than earlier generations of companies (a topic discussed in Chapter 2). Digitization has also created a shift toward launching products globally as opposed to staggering releases country by country. The entertainment industry is a case in point. For many years, Hollywood studios waited until after a movie’s US run to release it overseas, where the foreign box office could still compensate for domestic disappointments. The highest-grossing movie of 1995, Die Hard: With a Vengeance, was screening in only three countries within ten days of its US release date; they represented only 10 percent of the total markets where it would eventually be released. In 2015, Star Wars: The Force Awakens launched in every major market (80 countries) except China in the same week, and more than half of its record-breaking debut weekend box office came from international ticket sales (Exhibit 35). The information transparency offered by the Internet means that consumers around the world can see reviews immediately. There is no longer an opportunity to tweak and remarket; if a movie bombs in one place, it will be a global bomb. 80 countries where the latest Star Wars was released in its first week Exhibit 35 Hollywood releases illustrate the growing trend toward simultaneous global launches SOURCE: IMDB.com; boxofficemojo.com; McKinsey Global Institute analysis Simultaneous release of highest-grossing Hollywood movies in multiple countries1 1 W ithin 10 days of release data; excludes movie premieres. Highest-grossing movies in their respective years. 2 Rounded. 1990 1 1995 3 2000 5 2005 43 2012 2015 80 Total number of countries in which movie was released2 30 30 60 65 80 80 70 90 McKinsey Global Institute 5. Competing in a digital global landscape Music, games, and books are following the same trend toward global releases. As social media exposes consumers from around the world to what is available, products have a chance to go viral. In 2015, Adele’s song “Hello” racked up 50 million views on YouTube in its first 48 hours, and during its first week of release, her smash album 25 was No. 1 on the download list of iTunes stores in 110 countries.107 An additional incentive for global product releases is to protect intellectual property and capitalize on the initial wave of demand before piracy can strike. This is not to say that tailoring for local markets is dead. The need to create offerings at a lower price point for emerging economies is still valid. Microsoft, for instance, recently announced a new Nokia smartphone with extended battery life that will retail for around $20, with an eye toward selling it in emerging markets.108 GE Healthcare has produced low-cost products such as a handheld ultrasound device and a CT scanner with emerging markets specifically in mind; it recently announced the formation of a Sustainable Healthcare Solutions business line to continue this focus on “frugal” product development.109 In fact, some of the low-cost offerings developed for emerging markets can now boomerang back to advanced economies, where there is also demand for value-engineered products. Many global automakers are attempting to balance the need for localization against the need for global scale in a complex manufacturing process by using a platform approach. They rely on a set of common underlying designs that can be customized by swapping certain components to create differentiated models. Today most major carmakers are whittling down the number of platforms across their international manufacturing operations—a streamlining effort that could produce billions in savings. Some companies are finding artful ways of combining the global and the local in their marketing initiatives. The right approach can put a more intimate or relatable face on a global brand or take a fundamentally local experience and imbue it with a broader, more aspirational message. Starbucks, for example, launched a “Meet Me at Starbucks” global campaign through a mini-documentary shot in 59 different stores in 29 countries. The project, which shows a day in the life of a Starbucks store to emphasize its role as a community gathering place around the world, spans New York, Rio, Bogotá, Singapore, Beijing, Mumbai, Toronto, Paris, and Berlin.110 The long-running “keep walking” campaign for Diageo’s Johnnie Walker scotch was adapted for the Chinese market by associating it with the traditional Confucian saying that “the journey of a thousand miles begins with a single step.” As global value chains shift, do your suppliers and customers channels still make sense? Multinationals and their long, intricate supply chains are the driving force behind the world’s flows of goods. Digital tools can orchestrate a multitude of vendors stretching around the globe with greater precision and efficiency, opening up new possibilities for procurement. Companies such as Cisco and P&G have built “control towers” that offer up-to-the-minute visibility across complex global supply chains. These hubs synthesize information from sensors, actuators, RFID tags, GPS tracking, and more into dynamic models that can help managers evaluate alternatives instantly if risks or bottlenecks arise.111 107 Clarisse Loughrey, “Adele’s new album 25 is No. 1 on iTunes in almost every country in the world,” The Independent, November 26, 2015. 108 “The new Nokia 105 helps give people a voice,” Microsoft corporate blog, June 3, 2015. 109 “GE Healthcare announces $300 million commitment to support emerging market health,” corporate press release, September 23, 2015. 110 Maureen Morrison, “Starbucks launches first brand campaign, ‘Meet me at Starbucks,’” Ad Age, September 29, 2014. 111 See, for example, the interviews with supply chain executives in Bob Trebilcock, “What does it take to remain a supply chain leader?” Supply Chain Management Review, January 2, 2015; and Steve Banker, “Procter & Gamble’s futuristic control tower environment,” Forbes.com, July 1, 2015. 91McKinsey Global Institute Digital globalization: The new era of global flows But even as technology enables more complex procurement and collaboration, the importance of different factor costs is shifting. After years of choosing production locations largely on the basis of where low-cost labor is available, many manufacturers are beginning to reassess those decisions. Factors such as logistics costs, lead time, productivity, consumer preferences, and proximity to other company operations are being given greater weight. For some products, low-cost labor will continue to be the decisive factor. As China’s wages rise and the country makes a push to move up the value chain into more innovative, higher- value-added industries, more of its manufacturing business is up for grabs. Japan, for instance, has been shifting FDI from China to Southeast Asia. Now even some Chinese companies are shifting textile and apparel production to Africa. However, more than two-thirds of global manufacturing activity takes place in industries that tend to locate close to demand—and companies have to consider emerging markets as sources of that demand, not just supply. As incomes rise in these regions, demand is fragmenting as customers expect greater variation and more after-sales service.112 According to a recent UPS survey, approximately one-third of high-tech companies are moving manufacturing or assembly closer to end-user markets—and this number is up by 25 percentage points from 2010.113 In the future, 3D printing could change the very definition of what constitutes an intermediate good, disrupting logistics companies. UPS, for example, recently launched a pilot program to experiment with offering industrial-grade 3D printing services.114 Do you have the right assets to compete digitally and globally? Companies will need new types of assets to succeed in this new landscape. Building digital platforms and data centers may be critical for a growing range of companies, not just the Internet giants. Advanced digital capabilities are a major source of competitive advantage, and even traditional industries that lagged behind in the first wave of digitization are beginning to transform rapidly. Consider how big data analytics are transforming manufacturing. The manufacturers at the leading edge are digitizing and connecting their equipment, facilities, fleets, and other physical assets with the Internet of Things.115 They are developing expertise in big data analytics to generate insights from the flood of data being collected. GE, for example, is boosting investment to position itself as a leader in the industrial Internet. The company hopes to improve productivity, innovation, and customer retention in its own manufacturing operations and to become a provider of related services, applications, and platforms to other industrial firms. Other companies are taking a similar path. Rio Tinto, for example, transmits data continuously from its mines, processing plants, and vehicle fleet to “excellence centres” located in Brisbane, Australia. Analysts interact with this data on some of the largest touch screens in the world, creating models to head off potential production delays before they occur and making decisions about operational efficiency almost in real time. The company plans to open a third excellence centre for analytics in India in early 2016.116 112 Katy George, Sree Ramaswamy, and Lou Rassey, “Next-shoring: A CEO’s guide,” McKinsey Quarterly, January 2014. See also Manufacturing the future: The next era of global growth and innovation, McKinsey Global Institute, November 2012. 113 Change in the (supply) chain, United Parcel Service, 2015. 114 Lindsay Ellis and Laura Stevens, “UPS tests a 3D printing service,” The Wall Street Journal, September 18, 2015. 115 Digital America: A tale of the haves and have-mores, McKinsey Global Institute, December 2015. 116 Carly Leonida, “Rio Tinto to open Mining Excellence Centre,” Mining Magazine, February 25, 2015. 1 OF 3 high-tech companies report plans to move production closer to end-user markets 92 McKinsey Global Institute 5. Competing in a digital global landscape Businesses in all industries need to take a fresh look at the assets they hold, including customer relationships and market data, and consider whether there are new ways to monetize them given the emergence of new markets and new technologies. Alibaba, for instance, is at the center of China’s e-commerce ecosystem and has a vast pool of transactional data on the vendors that operate on its platform. Building on these advantages, the companies has moved into new areas such as mobile payments and small business financing. The insurance industry could similarly harness its sophisticated data pools on different forms of risk to create new products and services. Another key asset that many companies undervalue is an effective online presence. A passive corporate website is no longer sufficient. Companies also need a responsive social media voice and perhaps even their own proprietary platform. In consumer- facing industries, customer reviews on social media are increasingly important in driving business—and conversely, poor reviews can cause harm. Airline complaints on Twitter can go viral, and bad reviews on TripAdvisor or Yelp can cost hotels and restaurants dearly as growing numbers of international travelers rely on these sites to shape spending decisions. Many e-commerce sites now invite customers to post product reviews, and integrating their feedback quickly into the next product development cycle can pay real dividends. Creating a 24/7 team to monitor social media, handle customer complaints, and maintain a reputation is critical. Are you ready for a new world of digitally accelerated global competition? Corporate competition has intensified dramatically as emerging-market giants and digital disruptors go global. Both sets of competitors are lean, agile, and aggressive; both have cost advantages that enable them to take on established industry leaders. They are also demonstrating the ability to operate fluidly across geographic and sector boundaries.117 As digital technologies reduce the time, capital, and minimum scale needed for startups to compete globally, these dynamics are accelerating. From 1965 to 2012, the “topple rate” at which companies lose market-leading positions increased by almost 40 percent—and the world is only speeding up from here.118 Emerging-market companies are breaking into the top ranks of their industries globally. Between 1980 and 2000, the share of Fortune Global 500 companies based outside developed regions stayed relatively flat, at 5 percent. By 2010, this share was up to 17 percent of the total, and it climbed further still to reach 26 percent in 2013. Based on projected growth by region, MGI has forecast that emerging economies will account for more than 45 percent of the Fortune Global 500 by 2025 (Exhibit 36).119 Unlike publicly listed companies in the United States and Europe, many of the new emerging-market competitors are state- or family-owned, which can give them the flexibility to pursue longer-term strategies. Having grown up in difficult operating environments, they have a natural advantage in other fast-growing emerging markets. The Chinese telecom firm Huawei, for example, has become the third-largest smartphone vendor in the world, with a strong presence in markets from Africa and India to Myanmar. Indonesia’s Indofood has successfully introduced its Indomie noodles across Africa, becoming the most popular brand in the huge Nigerian market. As the global playing field becomes more crowded with international companies, the war for talent is taking on another dimension. A recent survey of US executives found that almost 40 percent of companies had missed business opportunities in the past five years due to lack of international competencies. More than a quarter of companies indicate that it is 117 Playing to win: The new global competition for corporate profits, McKinsey Global Institute, September 2015. 118 John Hagel III et al., 2013 Shift Index metrics: The burdens of the past, Deloitte, 2013. 119 Urban world: The shifting global business landscape, McKinsey Global Institute, October 2013. 93McKinsey Global Institute Digital globalization: The new era of global flows difficult to find US talent with the international knowledge, expertise, and language skills needed to manage global operations.120 Online talent platforms are creating a more global labor market—and this development gives workers more mobility and gives companies new ways to poach their competitors’ top performers. Attracting and retaining valued employees is a growing issue for Western multinationals operating in emerging economies. These companies were once considered the most prestigious employers in these countries, but now local firms are becoming global players themselves, and they can offer competitive compensation and career paths. One survey found that 34 foreign firms were listed among the 50 most attractive employers in China in 2004, but only 15 made the list in 2014.121 120 Shirley J. Daniel, Fujiao Xie, and Ben L. Kedia, 2014 US business needs for employees with international expertise, presented at the Internationalization of US Education in the 21st Century research conference in Williamsburg, Virginia, April 11–13, 2014. 121 Universum global survey of most attractive employers, 2014. Exhibit 36 Number of Fortune Global 500 companies1 By 2025, emerging regions are expected to be home to almost 230 companies in the Fortune Global 500, up from 85 in 2010 SOURCE: MGI CompanyScope; McKinsey Global Institute analysis 1 The Fortune Global 500 is an annual ranking of the top 500 companies worldwide by gross revenue in US dollars. 2 All emerging regions with the exceptions of China and Latin America combined until 2000. 3 Fortune Global 500 share in 2025 projected from revenue share of countries in 2025. NOTE: Numbers may not sum due to rounding. 12 54 120 10 34 13 15 8 26 78 11 8 12 26 South Asia 2 Southeast Asia Africa Latin America Other emerging2 2025E3 271 476 2000 6 1990 477 6 1 1980 477 2010 415 41 Developed regions China Total in emerging regions Emerging regions 23 23 24 85 229 94 McKinsey Global Institute 5. Competing in a digital global landscape Technology firms represent another huge source of competition. Some of the truly disruptive players are siphoning value out of industries and giving it away for free to consumers as a way to build their positions. Skype, for instance, shifted some $37 billion to consumers in 2013 alone by offering free international calls.122 Many technology-enabled firms are blurring traditional industry boundaries as they add new business lines. Alphabet has expanded into areas well beyond Google’s original search and advertising businesses, including longevity and biotech research, smart home products, venture capital investing, and high-speed Internet fiber services. The largest Internet platform operators are giving rise to yet another competitive threat: huge pools of SMEs that can now reach customers around the world. Thousands of Chinese manufacturers operating on Alibaba now have global reach, as do thousands of SMEs using eBay. The smaller firms operating within global e-commerce marketplaces now have the resources and reach to cherry-pick customers from industry incumbents. The Internet is also creating global pricing pressures. Consumers can comparison shop across multiple channels and markets, making it more difficult for companies to implement tiered pricing strategies. Apparel brands that could maintain a luxury image and command higher prices in certain markets now have difficult decisions to make. To get around this issue, some supermarkets and other retailers have introduced private-label brands. This trend is even spreading to major B2B distributors, with companies such as Grainger and Sysco increasingly emphasizing their own private labels. The Internet has also cut into the window of exclusivity companies once enjoyed on new products and services. Similar versions can be launched in new markets before the originator has time to scale up. Just months after the launch of Uber, there were similar companies operating in China and India. Rocket Internet, a German startup incubator, specializes in financing online businesses that bring successful business models from one country to new international markets.123 How do you manage new types of risk in a more digital and interconnected world? The impact of external shocks is magnified in a more interconnected world—and ripple effects spread even faster in a more digital world. The 2008 financial crisis showed how rapidly the linkages between the world’s capital markets can allow contagion to spread. Just a few years later, a series of natural disasters underscored the vulnerability of long global supply chains. Toyota’s production, for example, took a major hit in the aftermath of the 2011 Japanese earthquake and tsunami—because of damage not to its own factories but to the operations of its suppliers. In fact, the ramifications were felt around the world, with component shortages causing the temporary shutdown of GM and Ford plants in the United States.124 Heavy monsoons in Thailand that same year produced flooding in a region that produced nearly half of the world’s supply of hard drive disks, sending global prices soaring.125 Global executives polled in McKinsey’s latest survey on economic conditions cited geopolitical instability as the greatest risk to growth.126 122 Playing to win: The new global competition for corporate profits, McKinsey Global Institute, September 2015. 123 Sarah Gordon and Dan McCrum, “Rocket Internet: Waiting for lift-off,” Financial Times, October 19, 2015. 124 Bill Canis, The motor vehicle supply chain: Effects of the Japanese earthquake and tsunami, Congressional Research Service report for Congress, May 2011. 125 Rich Miller, “How floods in Thailand made AWS rethink its supply chain,” Data Center Frontier, October 8, 2015. 126 McKinsey survey of more than 2,000 executives from companies representing a range of geographies, industries, and sizes. Economic conditions snapshot, December 2015: McKinsey Global Survey results, McKinsey & Company, December 2015. 95McKinsey Global Institute Digital globalization: The new era of global flows Risk management has to be near the top of every corporation’s agenda. Some companies have created multidisciplinary risk teams and implemented more flexible procurement contracts and manufacturing systems. Regionalizing production near large end markets can both reduce complexity in supply chains and minimize exposure to disruptions in transit. Manufacturers need to periodically reevaluate the right balance between the use of global suppliers and the resilience of operations. One of the most important precautions is to diversify the supply chain, avoiding overreliance on any single supplier. In the pharmaceuticals manufacturing industry, up to 30 percent of company revenue can be traced to a single production site; up to three-quarters of revenue for some blockbuster drugs is at risk due to single-sourcing somewhere along the supply chain.127 As the world grows more dependent on information systems, new types of risk—such as the failure of power grids or damaging information leaks—enter the equation. The private sector is also becoming more vulnerable to cyberattacks by disgruntled employees, criminals, political activists, and even other nations. High-profile hacks and breaches have hit many of the world’s largest companies. One study has estimated that cybercrime costs the global economy some $400 billion in annual losses; these can include consumer data breaches, financial crimes, market manipulation, and theft of intellectual property. This is line with an estimate from Lloyd’s of London.128 A recent joint study by McKinsey and the World Economic Forum found that nearly 80 percent of technology executives said that they cannot keep up with attackers’ increasing sophistication and that protective measures (such as avoiding public cloud services or limiting the degree to which employees share information) are already having a negative business impact. Companies can prioritize information assets based on business risks, test continuously to improve incident response, and work with frontline employees to emphasize basic protective measures. If a breach does occur, a quick, decisive, and forthright response from marketing, public affairs, and customer service functions can be critical to restoring customer trust.129 ••• As digital technologies and globalization continue to reshape industries, the challenges for companies are mounting—but so are the opportunities. This new world similarly poses more complex questions for policy makers and regulators as economies around the world race to carve out new roles in global value chains. Chapter 6 will explore some of the implications for governments seeking to capture benefits of participating in global flows. 127 Manufacturing the future: The next era of global growth and innovation, McKinsey Global Institute, November 2012. 128 Net losses: Estimating the global cost of cybercrime, Center for Strategic and International Studies and McAfee, June 2014. See also Stephen Gandel, “Lloyd’s CEO: Cyber attacks cost companies $400 billion every year,” Fortune, January 23, 2015. 129 David Chinn, James Kaplan, and Allen Weinberg, Risk and responsibility in a hyperconnected world: Implications for enterprises, McKinsey & Company and the World Economic Forum, January 2014. © Getty Images Countries cannot afford to shut themselves off from global flows. Given their role in substantially raising GDP and boosting productivity growth, there is too much value at stake. But the goal is much broader than simply running a trade surplus. Our analysis finds that inflows and outflows alike contribute to growth. Narrow export strategies often ignore the real value of globalization: the flow of ideas, talent, and inputs that allow companies to innovate in new ways and raise productivity in the economy. Pursuing this value has never been a straightforward proposition, and today’s more digital form of globalization makes the calculus even more complex. Trade negotiations will need to include new dimensions to address issues surrounding cross-border data flows and the exchange of information and communication technology (ICT) goods. National policy makers increasingly need a global mindset to avoid erecting barriers that can lead to competitive disadvantages. The current wave of churn and transition creates openings for countries to carve out profitable roles in the global economy. Those opportunities will favor locations that build the infrastructure, institutions, and business environments that their companies and citizens need to participate fully. Building these enablers can have the double benefit of boosting domestic productivity—and without them, the economic impact of flows will be muted. Realizing the full economic potential of global marketplaces, platforms, and communities will require a deeper level of international cooperation. It will also depend on whether policy makers can successful manage the volatility associated with an interconnected and rapidly evolving digital economy. While it is impossible to anticipate all of the issues that will come into play, this chapter offers a framework. POLICY MAKERS NEED A CLEAR AGENDA TO CAPTURE THE FULL POTENTIAL OF GLOBAL FLOWS Even as governments try to create the right enabling environments for technology to fuel growth, digitization is handing them a host of entirely new policy challenges. Many digital firms have innovative business models that existing regulatory structures never considered. The digital economy evolves so rapidly that regulators have to take a test-and-learn approach to keep up with the pace of innovation. Many of the challenges associated with digitizing economic activity are now playing out on a global scale. 6. THE NEW WORLD OF POLICY CHALLENGES 98 McKinsey Global Institute 6. The new world of policy challenges Think strategically about the role your country can play in global value chains Officials building a national agenda to compete successfully in this new era could start by taking a step back and thinking strategically about how their country can participate in global flows based on assets they already have or could build. As global value chains shift, countries can redefine the roles they play within them. The United States, for example, has long been a major engine of consumer demand for imported goods, but it now plays an equally important role as the world’s leading producer of digital platforms and content. As new digital hubs form, the network of global flows may be redrawn in the years ahead. Automation is narrowing the window of opportunity for developing countries to become the world’s low-cost manufacturers, and 3D printing could transform how—and where—many categories of goods are produced. But other types of opportunities exist. Some countries could build on their geographic proximity to major consumer markets, as Mexico and countries in Eastern Europe have done. Others may develop successful niches as global transit hubs, although it is crucial to find ways to add value in addition to serving as a waypoint. Singapore, for instance, has become central to flows of goods and services, while Dubai has become a hub of transportation, trade, and finance. Other countries have used a selective approach, targeting a particular flow or industry to cultivate: China long ago transformed itself into the world’s manufacturing powerhouse, for instance, and is now pursuing an active strategy to move up the value chain into more innovative industries. The Philippines, Morocco, and South Africa have built on the advantages of language to become global providers of business process outsourcing services. Countries may also build on pools of talent within their borders, as Italy has done with high-end fashion design and textiles and as India has done with IT engineering. Recognizing the value of data flows, many locations are trying to create the “next Silicon Valley.” But innovation is notoriously hard to orchestrate—and that is not the only way to participate in the digital global economy. Our research finds that countries benefit from receiving cross-border digital flows as well as producing them. In other words, countries do not need to transform themselves into digital content or platform producers to benefit from data flows. Address policy and administrative barriers that hinder global flows For national economies, opening up to all types of global inflows and outflows is crucial for sustaining growth. Previous MGI research on global productivity trends has underscored this effect in Brazil, where some sectors have been more exposed to global market forces and some remain heavily protected. Embraer, the country’s flagship aerospace company, for example, was privatized and now successfully goes head-to-head with global competitors. Because Brazil lifted import tariffs on aircraft components, the company is able to source from global suppliers. By contrast, import tariffs on vehicles have encouraged foreign carmakers to establish production within Brazil to serve its large consumer market, but the Brazilian automotive industry has not integrated effectively into global value chains. Its productivity lags well behind peer economies such as Mexico, which has developed world-class assembly plants and rapidly gained global market share. Within Mexico, too, sectors that have privatized, embraced free trade, and welcomed foreign investment and technology have pulled far ahead of traditional industries in terms of productivity performance.130 Pursuing bilateral and multilateral trade partnerships is the cornerstone of a more open approach. Another important step is removing import tariffs, quotas, and subsidies for national industries, all of which can introduce distortions. Other types of legal and 130 See previous MGI studies: Connecting Brazil to the world: A path to inclusive growth, May 2014, and A tale of two Mexicos: Growth and prosperity in a two-speed economy, March 2014. 99McKinsey Global Institute Digital globalization: The new era of global flows administrative barriers can constrain the impact of global flows; these may include limitations on foreign business ownership and investment, import licensing, regulatory requirements that deviate from international norms, and limits on immigration. ASEAN, for instance, has largely eliminated import tariffs among its ten member states, but its ongoing effort to build a seamless trading bloc involves painstaking multilateral efforts to harmonize many types of product standards, certification procedures, customs requirements, and cross-border regulations covering traded services and the movement of labor.131 Removing these types of barriers can enable large multinational companies, SMEs, entrepreneurs, and individuals alike to take advantage of opportunities beyond their own borders. Address the dislocations Although the overall economic benefits of opening to global flows are clear, they can also disrupt local industries by exposing them to international competition and new business models. Some jobs and businesses may be lost even as new opportunities for growth are created. (See Chapter 4 for more on this topic.) Governments have to consider these trade-offs and open themselves to global flows at a pace their economies and societies can absorb. Labor markets and training systems in most countries have not proven flexible enough to deal with rapid change on this scale. But providing support to affected workers and creating a clearer path for them to find new roles deserves greater priority. Wage insurance is one policy option.132 Another is ensuring that adults who are already in the workforce have access to short, concentrated training programs for acquiring new skills. Germany may offer a useful model for policy makers to consider. One of the world’s most connected countries, ranking fourth in our global index, it has avoided large-scale unemployment by providing income support and taking a proactive approach to labor market reforms.133 Invest in human capital The Internet can promote inclusiveness, as long as education and training systems provide language fluency, basic digital literacy, and other skills so that individuals can take advantage of the opportunities. But educational systems in most countries are not keeping up with the demands of a digital world; few mandate computer programming classes in primary or secondary school. Even as the new digital era is raising the importance of education, technology also offers new possibilities for increasing its quality and reaching more people of all ages, whether through online educational platforms with open access, learning programs that adapt to a student’s performance, or classroom tools that allow teachers to tailor instruction. A more digital economy places a new premium on skills, innovation, and adaptability. The countries that are reaping disproportionate benefits are able to cultivate and attract pools of highly educated and specialized technical talent. Investment in human capital development will be a critical determinant of which nations come out on top. Build the necessary physical infrastructure and close the digital divide Even in a more digital world, physical infrastructure remains vital for tapping into global flows of all types. Roads, ports, airports, and rail are the conduits of trade and mobility; investment that modernizes and maintains these systems can propel economic growth. Many countries—emerging and advanced economies alike—have paid insufficient attention to those assets, creating economic inefficiencies and allowing foundational systems to erode. 131 Southeast Asia at the crossroads: Three paths to prosperity, McKinsey Global Institute, November 2014. 132 See, for example, Lori G. Kletzer, “Why the US needs wage insurance,” Harvard Business Review, January 25, 2016. 133 Marco Caliendo, Income support systems, labor market policies and labor supply: The German experience, IZA discussion paper number 4665, December 2009. 100 McKinsey Global Institute 6. The new world of policy challenges Today any list of infrastructure priorities also has to include universal, affordable Internet access. The number of worldwide Internet users now exceeds 3.2 billion, but growth is slowing. In 2011, the UN Broadband Commission set targets of reaching 60 percent worldwide Internet penetration by 2015, with 40 percent household penetration in developing nations. Those goals remain unmet, however: at the end of 2015, 57 percent of the world’s population, or four billion people, remained offline.134 The enormous digital divide in the world’s poorest countries and along gender lines remains stubbornly hard to bridge. As the flow of ideas, information, and innovation becomes more central to participating in the global economy, access to digital platforms and communication becomes an urgent development issue. The value of connecting the offline population to the Internet is significant. The World Bank has calculated that a 10 percent increase in broadband access is associated with a 1.38 percentage point increase in GDP growth in developing countries and a 1.21 percentage point increase in advanced economies.135 Our econometric analysis shows that countries with higher Internet penetration reap up to 25 percent more benefit from cross-border data flows than those with limited Internet penetration. Tremendous value can be created organically and unexpectedly when companies and citizens consume data and information—and then combine it with their own ingenuity. Create a strong business and institutional environment for the digital economy to thrive Just as purchasing IT systems offers no guarantee that a company will be a digital leader, building Internet infrastructure is not sufficient for countries to capture the full potential benefits of digital globalization. Their business sectors and consumer populations need to be able to engage and innovate online. A recent World Bank report finds that digital technologies have not improved productivity and reduced inequality to the degree once hoped in countries that lack strong fundamentals such as education, good governance, and a supportive business environment.136 These attributes have always been important for attracting foreign investment, and they are even more critical today. The benefits of digital globalization are heavily concentrated among countries with those ingredients in place, and lagging countries that fail to make broader reforms in these areas risk falling even further behind. The Internet can accelerate development and promote efficiency in emerging economies, but it is not a shortcut around building good governance. Countries still need healthy business environments that nurture startups, allow inefficient firms to exit, support research, and provide a solid legal framework for intellectual property and property rights. Without these elements, local companies will not be able to use global flows to raise their game, and foreign investors and companies will be deterred. India, for example, is attempting to tackle these issues and build a stronger foundation through initiatives such as Startup India and Digital India. A recent McKinsey survey found that business executives around the world believe that government agencies can provide more transparency and information on opportunities for domestic companies in foreign markets and opportunities for foreign companies 134 The state of broadband 2015: Broadband as a foundation for sustainable development, International Telecommunication Union and UNESCO Broadband Commission for Digital Development, September 2015. 135 Christine Zhen-Wei Qiang and Carlo M. Rossotto with Kaoru Kimura, “Economic impacts of broadband,” in Information and communications for development 2009: Extending reach and increasing impact, World Bank, 2009. 136 World development report 2016: Digital dividends, World Bank, January 2016. 101McKinsey Global Institute Digital globalization: The new era of global flows within the country. Having a one-stop shop to obtain such information, identify potential business partners, and understand the regulatory and approval process is becoming essential. Now that small firms have new avenues for participating in digitally facilitated global trade, governments can raise awareness of these growth opportunities. Expanding the information, mentoring, and financing available to small businesses can help them take advantage of this new shift in cross-border commerce. Governments can provide another enabler by opening their enormous data sets to encourage private-sector innovation. Making data more open and widely available in shareable formats can create substantial economic value, estimated at more than $3 trillion by MGI. Governments from India, Mexico, New Zealand, Singapore, the United Kingdom, and the United States are among those that have launched open data initiatives.137 Maintain an open Internet while protecting privacy Taking an open approach to cross-border data flows can accelerate growth. Yet many countries are considering limitations on what kind of data can be transmitted across borders and where data must be stored. Some are moving toward regulations that would require companies to use servers physically located within their borders to process and store data generated there. Variations on this type of law exist in Indonesia, Nigeria, Russia, Vietnam, and elsewhere. Other countries limit certain types of personal data transfers or have unique consent requirements.138 In 2014, Brazil passed a sweeping “Internet bill of rights”; some technologists questioned whether its privacy provisions, restrictions on data collection, and requirements that Brazilians’ data must remain stored on servers within the country could limit the use of large-scale analytics.139 Privacy has been a major issue in Europe. A 2014 ruling by the European Court of Justice upheld the “right to be forgotten”—that is, requiring search engines to honor requests from individual users to remove links to personal, inaccurate, or outdated information. As we went to press, the future of the “safe harbor” agreement governing data transfers between the European Union and the United States remained uncertain. Requirements that data must be sequestered locally raise a host of issues for companies, including cloud data storage and even personnel data for multinationals with operations and employees in restricted nations. In particular, some of the business models surrounding the Internet of Things are predicated on transmitting data to intermediaries, some of which may be in other countries. Some companies are proceeding with building their own data centers in locations around the world to cope with these types of requirements. But the compliance burdens of sequestering data and operating across multiple countries with varying regulations could limit the economic benefits of cross-border data flows. Beyond those governments that are acting out of concern for the privacy of their citizens, others regard the freewheeling nature of the Internet as a challenge to their authority and have moved to censor content, block websites, or place users under surveillance. 137 Open data: Unlocking innovation and performance with liquid information, McKinsey Global Institute, October 2013. 138 Data localization: A challenge to global commerce and the free flow of information, Albright-Stonebridge Group, September 2015. 139 Maria Medrano, “Brazil’s Internet Bill of Rights,” Americas Quarterly, April 2015. 102 McKinsey Global Institute 6. The new world of policy challenges While legitimate privacy concerns do need to be addressed through the development of universal standards, the movement toward data localization raises the danger of balkanizing the Internet. The economic benefits of cross-border data flows could be limited if the Internet becomes governed by a web of varying country-specific regulations. A study by the European Centre for International Political Economy examined the impact of recently proposed or enacted data localization and security regulations in seven economies. It found that these rules would lower GDP in all seven cases, with Vietnam (-1.7 percent), China (-1.1 percent), and Indonesia (-0.5 percent) poised for the largest losses.140 Another persistent issue is the tendency of parts of the digital economy to develop natural monopolies. The biggest platforms have enormous network effects and low marginal costs precisely because of their enormous size. Some governments are wary of this type of market power being held by a foreign company. But restricting the biggest global platforms denies that country’s citizens and small businesses the opportunity to participate. Our analysis described in Chapter 4 finds that countries benefit significantly from data consumption, not merely from being home to Internet companies and platform providers. Governments may want to consider whether their countries can produce their own robust digital platforms to compete (as China has done). But walling a country off from global platforms while failing to cultivate its own is a harmful combination. Make cybersecurity a top priority A world that runs on data flows is also vulnerable to cyberattacks. Private-sector companies and government agencies alike have suffered serious data breaches at the hands of hackers. One study has estimated that cybercrime costs the global economy some $400 billion in annual losses; these can include consumer data breaches, financial crimes, market manipulation, and theft of intellectual property.141 Aside from the substantial business costs, hackers may pose public safety and even national security risks. Governments will need to work closely with the business community to stay on top of new threats and share information and new technology solutions. Regulators may need to mandate standards for securing consumer data, and public agencies need to take additional steps to safeguard their own assets. Data privacy and security are thorny issues in almost every area of digital use. They are central to realizing the full economic value of big data analytics and the Internet of Things, which are predicated on collecting and sharing data. Governments have to make choices about data collection, access, usage, and consent, especially for data generated in public spaces. The dangers that hackers could create in physical settings have to be carefully considered and guarded against; policy makers can help to address security issues by creating frameworks for liability.142 Beyond the threat of breaches, governments need to be aware of what is lurking on the so-called Darknet. The public Internet is vast, but it is dwarfed in size by the “Deep Web” of non-indexed websites, as we explain in Chapter 1. Much of the Deep Web is simply private information held by companies, enormous government databases, pay-to-use databases, or private message boards. It is also used by activists, dissidents, journalists, whistleblowers, and others who have legitimate needs for maintaining anonymity. But hidden on the Darknet portion of the Deep Web is an entire online world of criminal activity, including money laundering, drug trafficking, human trafficking, child pornography, hackers for hire, and terrorist networks. Some criminal rings have been broken up and their members 140 See Matthias Bauer et al., The costs of data localization: Friendly fire on economic recovery, ECIPE occasional paper number 3/2014, May 2014. The study also found that investment could drop by 4.2 percent in Brazil, 3.9 percent in the European Union, and 3.1 percent in Vietnam. 141 Net losses: Estimating the global cost of cybercrime, Center for Strategic and International Studies and McAfee, June 2014. 142 The Internet of Things: Mapping the value beyond the hype, McKinsey Global Institute, June 2015. 103McKinsey Global Institute Digital globalization: The new era of global flows prosecuted, including the founder of the notorious Silk Road black market. But reining in the illicit global trade being conducted in cyberspace will require deeper international coordination.143 ••• Policy makers have to strike the right balance between capturing the benefits of openness while mitigating the risks—and in a digital world, both opportunities and challenges are appearing with unprecedented speed. This new version of globalization is creating a faster- moving and vastly more complex global economy, but it offers new ways to realize the value of connectedness. This wave of change can accelerate growth for the countries that approach it with optimism and vision. 143 Daniel Sui, James Caverlee, and Dakota Rudesill, The Deep Web and the Darknet: A look inside the Internet’s massive black box, Woodrow Wilson International Center for Scholars, October 2015. © Getty Images This appendix outlines key points on the methodology in the following sections: 1. Data sources and definitions 2. Econometric model and statistical analyses 3. Academic literature on the relationship between global flows and GDP 4. Methodology for the MGI Connectedness Index 5. Methodology for global connectedness of regions within countries 6. Global survey of startups 1. DATA SOURCES AND DEFINITIONS Global cross-border flows We compiled a data set covering five categories of flows for 139 countries from 1980 to 2014 or the latest data available (Exhibit A1). The data set draws on multiple sources that we describe in more detail later in this section. For each type of flow, we assembled inflows and outflows for each country individually, and wherever possible bilaterally. The coverage of bilateral data over time and across countries varies by flow. Figures for total flows used in this report refer to the broadest coverage available. We used subcategories of overall goods, services, and financial flows for specific analyses. For example, our analysis of knowledge-intensive flows includes only the knowledge- intensive subcategories of each aggregate flow. The mapping to the goods categories has been performed based on the United Nations’ six-digit harmonized coding system, HS 2002. We assigned service categories using the 11 chapters in the Comtrade database of global commodities trade statistics maintained by the United Nations Statistics Division. We categorized financial flows by the nature of their investment (i.e., FDI, equity, bonds, and loans) and based them on data from several sources. People flows are not composed of any single aggregate flow. Instead, we analyzed several components such as international student flows, long-term migrants, refugees, and overnight visitors. For data flows, we examined used cross-border bandwidth from TeleGeography. TECHNICAL APPENDIX 106 McKinsey Global Institute Technical appendix Exhibit A1 UNCTAD UNCTAD Capital-intensive goods 160 UNCTAD UNCTAD Migrants Total financial stock Portfolio investment IMF Travelers 194 EWN1 Bonds IMF 112 World Bank 185 Reserves 214 260 273 Loans 125IMF Equity Remittance 113 World Bank 251 IMF IMF 265IMFFDI flow UNCTAD Cultural and social services Knowledge-intensive services Labor-intensive services 146 UNCTAD IMF 181 138 104 Services total UNCTAD Government services Primary resources 160 UNCTAD 146 135 Capital-intensive services UNCTAD 160 Total financial flows UNCTAD UNCTAD Labor-intensive goods 218 160R&D-intensive goods Goods total Maximum number of countriesMajor flows 171TeleGeographyUsed bandwidth 201International students OECD Source 221UNWTO The compiled data set contains hundreds of countries and provides comprehensive coverage of the past decade G oods S ervices Finance P eople D ata SOURCE: McKinsey Global Institute analysis 1 External W ealth of Nations database. 88 90 95 1505 1000 107McKinsey Global Institute Digital globalization: The new era of global flows Goods flows This report analyzes the historical growth in the global flow of goods, its dispersion across countries and regions, and its transformation due to digitization and knowledge intensity. The primary source is the UNCTAD database, which provides non-bilateral data from 1980 onward. We analyzed more than 5,200 product codes between 2002 and 2014, dividing them into four categories: capital-intensive manufacturing, labor-intensive manufacturing, primary resources, and R&D-intensive manufacturing. For bilateral trade, we used data from the World Bank’s World Integrated Trade Solution database, which is available from 2000 to 2014. Each of the four categories of goods trade mentioned above has a number of subgroups. � Capital-intensive manufacturing — Food, beverages, and tobacco. Includes the production, processing, and preservation of meat, fish, fruit, vegetables, oils, and fats; the manufacture of dairy products, grain mill products, and starches and starch products; and the production of other food products and beverages including spirits, wines, malt liquors, soft drinks, and mineral waters. Also includes items related to the manufacture of tobacco products. — Paper products and publishing. Includes the manufacture of pulp, paper, and paperboard; the manufacture of corrugated paper and paperboard and containers made out of those materials; and the manufacture of specialty paper products, including carbon paper, toilet paper, envelopes, and postcards. — Manufacturing of petroleum, rubber, plastic, mineral, and metal products. Includes the production of products related to the commodities of metals, mining, and oil and gas. � Labor-intensive manufacturing — Textiles. Includes the spinning, weaving, and finishing of textiles; the manufacture of carpets, rugs, rope, twine, and netting; and the manufacture of knitted and crocheted fabrics and articles. — Leather, fur products, and apparel. Includes the manufacture of fur and non-fur apparel, and the dressing and dyeing of fur. Also includes the production of footwear, the tanning and dressing of leather, and the manufacture of leather products. — Wood products and furniture. Includes the sawmilling and planing of wood; the manufacture of wood, cork, and straw; and the manufacture of wood products, including furniture, musical instruments, sporting goods, and toys. � Primary resources — Agriculture, hunting, fishing, and related activities. Includes the growing of crops, the farming of animals, the hunting and trapping of animals, fishing, and the operation of fish hatcheries and farms. — Forestry and logging. Includes goods produced through forestry and logging and related service activities. 108 McKinsey Global Institute Technical appendix — Metals. Includes the manufacture of basic iron and steel, basic precious and non- ferrous metals, and the casting of metals. Also includes other fabricated metal products, such as tanks, reservoirs, and construction materials. — Mining. Includes the mining and agglomeration of hard coal and lignite, uranium, and thorium ores; the mining of ferrous and non-ferrous metal ores; the mining and quarrying of stone, sand, and clay; and the extraction of crude petroleum and natural gas. — Oil and gas. Includes the manufacture of coke oven products and refined petroleum products as well as the processing of nuclear fuel. � R&D-intensive manufacturing — Chemicals and chemical products. Includes the manufacture of basic chemicals, fertilizers and nitrogen compounds, plastics and rubbers, pesticides and other agro-chemical products, paints, pharmaceuticals, soaps and detergents, artificial or synthetic fibers, yarn, and filaments. — Electrical, telecommunication, and computing machinery. Includes the manufacture of office, accounting, and computing machinery; electrical machinery such as motors, generators, transformers, wires and cables, and electrical equipment; television and radio transmitters and receivers; and sound and video recording equipment. — Motor vehicles and other transport equipment. Includes the manufacture of motor vehicles, trailers and semitrailers, and parts and accessories for motor vehicles and their engines; and other transport equipment such as ships, railway and tramway locomotives, aircraft and spacecraft, motorcycles, and bicycles. — Medical, precision and optical instruments. Includes the manufacture of medical appliances and instruments, optical instruments, photographic equipment, and watches and clocks. — Other machinery and equipment. Includes the manufacture of general-purpose machinery such as engines, turbines, pumps, compressors, ovens, and lifting and handling equipment; special-purpose machinery such as agricultural machinery, weapons and ammunition; machinery for the production of mining and metal products, food, beverage, and tobacco products; and domestic appliances. We separately subdivided goods into three categories: finished goods, intermediate goods, and raw materials. The broad economic categories (BEC) classification system specified by the United States has three designations: consumer goods, capital goods, and intermediate goods. We consider both consumer and capital goods as finished goods. We further subdivide intermediate goods into raw materials and intermediate goods. � Finished goods. Includes finished capital and consumer goods such as industrial machines, ships and aircrafts, refined petroleum, sugar, and apparel. � Intermediate goods. Includes parts used as inputs for making finished products such as pharmaceutical inputs, vehicle parts, and steel products. � Raw materials. Includes commodities and processed commodities. Examples of commodities are coal, corn, cotton, and crude petroleum. Processed commodities include materials such as animal fats and oils, coffee, and processed metal. 109McKinsey Global Institute Digital globalization: The new era of global flows Services flows All non-bilateral flows of services draw on UNCTAD and UN Comtrade data. We subdivided services flows into five categories: knowledge-intensive, labor-intensive, capital-intensive, cultural and social, and government. However, the coverage of services data is most effective at the aggregate level rather than broken into these five categories. Each of these also has a number of subgroups:144 � Knowledge-intensive services — Insurance services. The provision of insurance to non-residents by resident insurance enterprises and vice versa; services provided for freight insurance on goods exported and imported; services provided for other types of direct insurance including life and non-life; and services provided for re-insurance. — Financial services. Financial intermediation services and auxiliary services conducted between residents and non-residents other than those related to insurance enterprises and pension funds. — Computer and information services. Resident and non-resident transactions related to hardware consultancy, software implementation, information services (i.e., data processing, data base, news agency), and maintenance and repair of computers and related equipment. — Royalties and license fees. Includes receipts (exports) and payments (imports) of residents and non-residents for the authorized use of intangible non-produced, non-financial assets and proprietary rights such as trademarks, copyrights, patents, processes, techniques, designs, manufacturing rights, and franchises; and the use, through licensing agreements, of produced originals or prototypes such as manuscripts and films. — Other business services. Covers merchanting and other trade-related services as well as operational leasing services; and miscellaneous business, professional, and technical services.145 � Labor-intensive services — Travel. Goods and services, including those related to health and education, acquired by travelers during visits of less than one year. The goods and services are purchased by, or on behalf of, the traveler or provided, without a quid pro quo, for the traveler to use or give away. — Construction services. Construction and installation project work performed on a temporary basis in the compiling economy or in extraterritorial enclaves by resident and non-resident enterprises and associated personnel, excluding foreign affiliates. 144 The definitions used for the subgroups are closely based on IMF Balance of Payments Manual definitions. 145 Purchase of a good by a resident of the compiling economy from a non-resident and the subsequent resale of the good to another non-resident. Value created between purchase and resale is recoded as value of merchanting service. 110 McKinsey Global Institute Technical appendix � Capital-intensive services — Communications services. Communications transactions between residents and non-residents (i.e., postal, courier, and telecommunications services). — Transportation. Transportation services performed by residents of one economy for those of another and vice versa, and that involve the carriage of passengers, the movement of goods (freight), rentals (charters) of carriers with crew, and related supporting and auxiliary services. � Cultural and social services. Audiovisual and related services and other cultural services provided by residents to non-residents and vice versa. � Government services. All services (e.g., spending by embassies and consulates) associated with government sectors or international and regional organizations and not classified under other items. Financial flows Aggregate financial flows comprise the following asset classes: � FDI. Investments that establish at least a 10 percent stake in a foreign entity. Any subsequent lending between the direct investor and the financial recipient is also captured in this category. � Equity. Any equity or share purchased by an investor in another country that gives the investor no more than a 10 percent stake. � Bonds. Any tradable debt security that is purchased by a foreign investor, including public and corporate (both financial and non-financial) bonds, mortgage-backed securities, other asset-backed securities, and collateralized debt obligations. � Loans. Any other assets not classified in the above three categories, primarily loans, currency, and deposits, and a small share of trade credit. In addition to these four classes, data on outward investments capture a fifth category of reserve assets: assets acquired or held by monetary authorities in a foreign currency. Reserve assets are distinguished from the other four classes to avoid double-counting. We take all these data from balance of payments statistics from the International Monetary Fund. Further, we also look at data on flows of remittances from the World Bank. We do not include these flows in our core analysis of major financial flows because they either overlap with other financial flows, such as loans, or are the reverse of goods and services flows. We also analyze a bilateral data set for FDI, which gives us an indication of the origin and destination of this flow. In addition to financial flows data, we take countries’ foreign assets and foreign liabilities (financial stock data) into consideration. This is sourced from Philip R. Lane’s “External Wealth of Nations” data set. 111McKinsey Global Institute Digital globalization: The new era of global flows People flows Unlike for flows of goods, services, and finance, we do not have additive data sets on people flows. Instead, we look at overlapping categories and describe people flows from different angles. All data we collected for people flows are bilateral, indicating both departure and arrival countries. We use three direct measures: � Migrants. While all other measures capture or approximate flows of people, our migration data are stock data, indicating foreign-born residents by country. Data are from the World Bank and are available for 1980 to 2013. � Travelers. Arrivals of non-resident visitors or tourists at national borders, drawn from the UN World Tourism Organization, available from 1995 to 2013. � Students. International flows of mobile students at the tertiary level (ISCED 5 and 6) from the UNESCO Institute for Statistics, available from 1999 to 2013, Data flows For data flows, we look at cross-border used bandwidth data from TeleGeography, which provides data by region, country, and key routes from 2005 to 2014. Used capacity is the sum of all capacity deployed for Internet backbones, private networks, and switched voice networks. It does not include capacity that is used for restoration and redundancy purposes. Knowledge-intensive flows We define knowledge-intensive flows as cross-border goods, services, finance, people, and data and communications flows that are rich with ideas, knowledge, and information. The aim is to approximate a value of global flows that are linked to today’s knowledge economy. We define the following subcategories of global flows as knowledge-intensive: � R&D-intensive manufacturing (goods flows). Of all goods flows, those classified as R&D-intensive manufacturing are considered to have the highest portion of knowledge involved in production or development. When these goods cross borders, the knowledge embodied in these products or their development is at least partially transferred across those borders. � Knowledge-intensive services (services flows). Knowledge-intensive services are those requiring the highest skill level of the parties providing the service (e.g., financial services) or that directly represent the realized value of knowledge or content creation (e.g., the payment of royalties and license fees). � FDI (financial flows). Of all financial flows, FDI is most clearly linked to the transfer of knowledge across borders as companies conducting greenfield FDI transfer knowledge to the new location. We also consider brownfield FDI as a transfer of knowledge because companies that acquire other companies either use their own knowledge and management techniques to improve the business of the acquisition or use the knowledge embedded in their investment to improve their own business. � Cross-border telecom revenue (data and communication flows). In our services database, telecommunications is a financial-intensive service. However, it is an important proxy for data and communication flows and represents a minimum value of those flows. Therefore, we have included this revenue in an attempt to capture a significant portion of the value of data and communication flows. We have elected to include only business, as opposed to residential, revenue because we believe these flows have the most substantial knowledge component. 112 McKinsey Global Institute Technical appendix � Business traveler spending (people flows). In our services database, business travel is a labor-intensive service. However, it can be used as a proxy for knowledge-intensive people flows and, at the very least, represents a minimum value of such flows. When business travelers move across borders, they carry knowledge with them; in fact, these travelers have likely traveled to a different country to either impart that knowledge or acquire knowledge that they will carry back to their home country. Country classifications For some analyses, we classify each of the 139 countries in our sample as either a developing or a developed economy. For developed economies, we use the term “advanced economies” interchangeably. We refer to developing economies as “emerging markets” or “emerging economies.” We also assigned each country to one of ten regions, six of which we define as emerging and four as developed. This classification of countries and their assignment to regions follows the approach used in previous McKinsey Global Institute reports.146 We define Singapore, Hong Kong, Taiwan, and Macao as developed economies despite the fact that they are located in emerging-market regions (Exhibit A2). 146 See, for example, Financial globalization: Retreat or reset? McKinsey Global Institute, March 2013. Exhibit A2 Classification of countries into regions and development level SOURCE: McKinsey Global Institute Financial Assets database; McKinsey Global Institute analysis 1 Combined to form “Other Asia” in analysis regarding interregional vs. intraregional trade. 2 Classified as developed despite being located in a region classified as emerging. Europe, Middle East, and Africa Western Europe ▪ Austria ▪ Belgium ▪ Denmark ▪ Finland ▪ France ▪ Germany ▪ Greece ▪ Iceland ▪ Ireland ▪ Italy ▪ Malta ▪ Netherlands ▪ Norway ▪ Portugal ▪ Spain ▪ Sweden ▪ Switzerland ▪ United Kingdom Eastern Europe and Central Asia ▪ Bulgaria ▪ Czech Republic ▪ Hungary ▪ Kazakhstan ▪ Lithuania ▪ Latvia ▪ Poland ▪ Russia ▪ Slovakia ▪ Turkey Plus 19 other countries Africa and Middle East ▪ Algeria ▪ Angola ▪ Botswana ▪ Cameroon ▪ Egypt ▪ Ghana ▪ Iran ▪ Israel ▪ Jordan ▪ Kenya ▪ Kuwait ▪ Lebanon ▪ Morocco ▪ Nigeria ▪ Saudi Arabia ▪ South Africa ▪ Tunisia ▪ United Arab Emirates Plus 26 other countries D eveloped regions E m erging regions Americas North America ▪ Canada ▪ United States Latin America ▪ Argentina ▪ Bolivia ▪ Brazil ▪ Chile ▪ Colombia ▪ Costa Rica ▪ Dominican Republic ▪ Ecuador ▪ Guatemala ▪ Jamaica ▪ Mexico ▪ Panama ▪ Uruguay ▪ Venezuela Plus 10 other countries Asia Northeast Asia ▪ Japan ▪ South Korea Australasia ▪ Australia ▪ New Zealand China region ▪ China South Asia1 ▪ India ▪ Pakistan ▪ Bangladesh ▪ Sri Lanka ▪ Maldives Southeast Asia1 ▪ Philippines ▪ Malaysia ▪ Cambodia ▪ Indonesia ▪ Thailand ▪ Vietnam ▪ Singapore2 Plus 5 other countries 113McKinsey Global Institute Digital globalization: The new era of global flows 2. ECONOMETRIC MODEL AND STATISTICAL ANALYSES In Chapter 4 of this report, we discuss why openness to global flows matters for economic performance. According to the classical Solow growth model, the main factors determining GDP growth are the physical capital stock of a nation and its human capital.147 Newer growth-theory models also include a role for technological progress or innovation. To test for the additional effect of cross-border flows on GDP growth, which may raise total factor productivity or increase the utilization of capital and labor, we employ a two-step error regression model. Our econometric model Our model is based on a classic Cobb-Douglas production function of the following form: where Y represents GDP, K is the fixed capital stock in an economy, L is the labor stock in an economy, and A is total factor productivity, which includes technological progress and innovation. We used a two-step error correction model (ECM). The ECM allows us to incorporate two aspects into our analysis: first is the inter-temporal relation in GDP growth (i.e., this year’s growth affects next year’s growth). We can thus estimate the change in GDP growth as a function of changes in independent variables such as flows. This inter-temporal relationship also allows us to differentiate between the short-term and long-term effects of the independent variables of GDP growth. We use data from 97 countries, from 1995 to 2013 (the latest year for which those data are available for all flows and all countries). To account for country-specific fixed effects that may also contribute to GDP growth, we format the data as a panel data set, pooling cross- sectional and time-series data, and run the regression on it using a fixed-effect model. This controls for time-invariant effects, such as the legal system or colonial history of the country. Given that we are using long time series in our analysis, we also pay attention to testing for cointegration in the data series. This is especially troublesome in building a dynamic econometric model, as cointegrated data series share a common trend and thus may lead to a false finding of correlation among the variables. We first used sophisticated methods to test for and correct cointegration and then used a two-step error-correction model for the estimation. To begin, we test the long-term relationship between the data series. We run a unit root test on the residual from this step and select a model whose residual is cointegrated. In the second step, we estimate the short-term relationship with the same set of covariates from the long-run relationship and its residual. Lastly, we control our potential endogeneity in the data series using instrumental variables. As instruments, we use the difference between the three-year lag of flows and four-year lags.148 We tested different time lags to see which instruments are stronger and determined that the three- to four-year lag is most suitable. 147 For an overview of different growth models, see, for example, Robert J. Barro and Xavier Sala-i-Martin, Economic growth, MIT Press, 2003. 148 Given that our specification includes the one-year lag of GDP, our lagged instrumental variables are at least t–2 or t–3. Given our panel data, we could have used the GMM method, which increases the number of lagged instruments. The gain in efficiency was small, so we report the traditional instrumental techniques, with Hausman test passed for their validity. Exhibit A15 Formulas Y = A × Kα × Lβ (1) Cobb-Douglas production function logGDPi,t = α + SkβklogCVi,k,t + SiβilogFlowi,j,t–1 + εi,t Econometric step 1 of 2-step ECM ∆logGDPi,t = α + Skβk∆logCVi,k,t + Sjβj∆logFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t Econometric step 2 of 2-step ECM flow’s GDP share in leveli,j,t = (flowi,j,t) βj real GDPi,t Flow’s share in GDP flow’s GDP share in growthi,j,t = (log(flowi,j,t) – log(flowi,j,t–1)) × βj (log(real GDPi,t) – log(real GDPi,t–1)) Flow’s share in growth log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 + θjcentralityi,tlogDataFlowi,j,t + εi,t Centrality Step 1 of 2-step ECM log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 + θjcentralityi,tlogDataFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t Centrality Step 2 of 2-step ECM 114 McKinsey Global Institute Technical appendix Using the ECM in this way allows us to make a more credible claim regarding the direction of causality. We thus examine the impact of the growth of flows in one period on the growth of GDP for the subsequent period, controlling for unobserved country-specific effect, the noise from the covariates that are not integrated, and endogeneity among the covariates and GDP. The transformation of the Cobb-Douglas function to logarithmic scale allows us to estimate the elasticities of each variable (i.e., by how much does GDP growth change if the explanatory variable changes by 1 percent). The estimated model takes the following form: Step 1 of two-step ECM where i represents country i, t the current year , t–1 the past year, S is the symbol of sum, j is the index of flow, and k is the index of control variables. The definition of each variable is as follows: � logGDPit is the marginal change in real GDP for country i between year t and t–1 in natural log � α is the constant term � logCVi,k,t is the k-th control variable and is the marginal change in the lagged level of respective control variables for country i in time t. Their coefficient estimates are interpreted as the long-term elasticity of real GDP with respect to a change in the control variable. � logFlowi,j,t–1 is the j-th flow and is the change in the lagged level of the respective flow. It determines the long-term elasticity. � εi,t is the residual or error that we will save for the second step, which has to be cointegrated (i.e., it does not suffer from having a unit root). Step 2 of two-step ECM: � ∆logGDPi,t is the change in growth of real GDP for country i between year t and t–1 in natural log � ∆logCVi,k,t–1 comprises the set of k control variables and is the change in growth of the respective control variable from year t–1 to t in natural log. Its coefficient estimates are interpreted as the short-term elasticity of real GDP growth with respect to the control variables. � ∆logFlowi,j,t–1 comprises the set of j flows and is the change in growth of the flows in natural log. Its coefficient estimates determine the short-term elasticity of real GDP growth with respect to flows. Exhibit A15 Formulas Y = A × Kα × Lβ (1) Cobb-Douglas production function logGDPi,t = α + SkβklogCVi,k,t + SiβilogFlowi,j,t–1 + εi,t Econometric step 1 of 2-step ECM ∆logGDPi,t = α + Skβk∆logCVi,k,t + Sjβj∆logFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t Econometric step 2 of 2-step ECM flow’s GDP share in leveli,j,t = (flowi,j,t) βj real GDPi,t Flow’s share in GDP flow’s GDP share in growthi,j,t = (log(flowi,j,t) – log(flowi,j,t–1)) × βj (log(real GDPi,t) – log(real GDPi,t–1)) Flow’s share in growth log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 + θjcentralityi,tlogDataFlowi,j,t + εi,t Centrality Step 1 of 2-step ECM log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 + θjcentralityi,tlogDataFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t Centrality Step 2 of 2-step ECM Exhibit A15 Formulas Y = A × Kα × Lβ (1) Cobb-Douglas production function logGDPi,t = α + SkβklogCVi,k,t + SiβilogFlowi,j,t–1 + εi,t Econometric step 1 of 2-step ECM ∆logGDPi,t = α + Skβk∆logCVi,k,t + Sjβj∆logFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t Econometric step 2 of 2-step ECM flow’s GDP share in leveli,j,t = (flowi,j,t) βj real GDPi,t Flow’s share in GDP flow’s GDP share in growthi,j,t = (log(flowi,j,t) – log(flowi,j,t–1)) × βj (log(real GDPi,t) – log(real GDPi,t–1)) Flow’s share in growth log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 + θjcentralityi,tlogDataFlowi,j,t + εi,t Centrality Step 1 of 2-step ECM log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 + θjcentralityi,tlogDataFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t Centrality Step 2 of 2-step ECM 115McKinsey Global Institute Digital globalization: The new era of global flows � logGDP per capitai,t–2 is the two-year lagged value of real GDP per capita in natural log, capturing the catch-up effect of the developing countries on the real GDP growth. Since this effect does not apply to countries that already have high GDP per capita (i.e., advanced countries), its coefficient γ is expected to be negative. � εi,t–1 is the lagged residual from the step one. Its coefficient δ needs to be negative and statistically significant for the second step to be statistically valid. Conceptually, this coefficient captures the rate of the short-term model converging into the long- term model. � ui,t is the error term from step two of ECM. Each of the control variables enters the regression as a stand-alone term. The control variables CV in the estimation are as follows: � Human capital, measured by average years of schooling in the adult population � Real fixed capital stock, derived from the accumulated real fixed investment in the country after depreciation, to capture capital inputs in the economy � Employment, to capture labor inputs in the economy. We test the relationship between flows and GDP by measuring flows in different ways. First, we use the sum of inflows and outflows for the country, normalized by nominal GDP (for goods trade and FDI) or population (for migration flows and data flows). Second, we use each country’s score in the MGI Connectedness Index. The normalized flows used in the estimation are goods trade, migration, FDI flows, and cross-border data usage. The connectedness scores used are for goods trade, labor-intensive services trade, travelers, FDI flow, and used cross-border bandwidth. These variables are selected by assessing their correlation with each other, Granger causality with GDP, and using a backward/forward stepwise model selection. We obtain the signs we expected on each of the flow variables (Exhibit A3), with the exception of migration flows. The coefficient on migration flows is negative for long-term elasticity, while we would expect it to be positive. We believe this result is due to loss of skilled labor in developing countries or their difficulty in absorbing large migrant or refugee flows. Exhibits A4 and A5 show the short-term and long-term elasticities from the two measures of global flows described above. 116 McKinsey Global Institute Technical appendix Exhibit A3 Short-/long-term impact Name of variable Granger causality with real GDP Expected sign of coefficient Estimated sign of coefficient FDI Two-way Positive/positive Positive/positive Goods trade flow Two-way Positive/positive Positive/positive Immigration Two-way Positive/positive Insignificant/negative1 Data flows Two-way Positive/positive Positive/positive Services trade flow Two-way Positive/positive Extended due to correlation with FDI Fixed capital stock n/a Positive/positive Positive/positive Employment n/a Positive/positive Positive/positive Average years of education n/a Positive/positive Insignificant/negative SOURCE: McKinsey Global Institute analysis The coefficients from our econometric model have the expected sign Flow variables Dependent variable (Log) Real GDP Independent variables (Log) 1 Migration flows are negligible or slightly negative at the global level, possibly due to the loss of skilled labor in developing countries or the difficulties of absorbing a large influx of refugees or migrants. However, migration flows have a positive impact on productivity in advanced economies. Exhibit A4 GDP impact of global flows, using normalized flow values SOURCE: McKinsey Global Institute analysis Elasticities Long term Short term Coefficients P-values Coefficients P-values Flow variables Flow variables Goods trade 0.05 0.0129 Goods trade 0.0817 0.0002 FDI 0.04 0 FDI 0.0039 0.0761 Migration -0.05 0.0036 Immigration Insignificant n/a Data 0.02 0 Data usage 0.025 0.0154 Macroeconomic variables Macroeconomic variables Fixed capital stock 0.48 0 Fixed capital stock 0.76 0 Employment 0.39 0 Employment 0.49 0 Average years of education Insignificant n/a Average years of education Not available n/a Dependent variable: Real GDP 97 countries, 1995–2013 117McKinsey Global Institute Digital globalization: The new era of global flows Calculating the impact of flows on GDP output and GDP growth After we estimate the two-step ECM, we use the short-term and long-term elasticities (shown in Exhibit A4) to calculate the contribution of flows to both the level of GDP and the growth rate of GDP for a country i for flow j at time t as follows: Where βj is the flow j’s GDP long-term elasticity from step one of two-step ECM Where βj is the flow j’s GDP short-term elasticity from step two of two-step ECM. These expressions are derived from the two-step ECM because it is a combination of the change in GDP in level (the long-term model) and the change in GDP growth (the short-term model). We calculate the above for both normalized flows and for a model specification using connectedness index scores. The results indicate that flows accounted for 10.1 percent of global GDP over the past ten years (Exhibit A6). Exhibit A5 GDP impact of global flows, using connectedness scores for each flow SOURCE: McKinsey Global Institute analysis Elasticities Long term Short term Coefficients P-values Coefficients P-values Connectedness scores Internet traffic 0.0180 0.1202 Insignificant n/a Goods trade 0.0706 0.0006 0.4264 0.0022 Service trade1 0.0249 0.0498 Insignificant n/a Travelers 0.0399 0.0407 Insignificant n/a FDI flow 0.0204 0.0444 Insignificant n/a Macroeconomic variables Fixed capital stock 0.4718 0 0.70 0 Employment 0.4685 0 0.48 0 Average years of education Insignificant n/a n/a n/a Dependent variable: Real GDP 97 countries, 1995–2013 1 W e use data on labor-intensive services trade only, not knowledge-intensive services or capital-intensive services, because the latter are highly correlated with FDI. Exhibit A15 Formulas Y = A × Kα × Lβ (1) Cobb-Douglas production function logGDPi,t = α + SkβklogCVi,k,t + SiβilogFlowi,j,t–1 + εi,t Econometric step 1 of 2-step ECM ∆logGDPi,t = α + Skβk∆logCVi,k,t + Sjβj∆logFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t Econometric step 2 of 2-step ECM flow’s GDP share in leveli,j,t = (flowi,j,t) βj real GDPi,t Flow’s share in GDP flow’s GDP share in growthi,j,t = (log(flowi,j,t) – log(flowi,j,t–1)) × βj (log(real GDPi,t) – log(real GDPi,t–1)) Flow’s share in growth log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 + θjcentralityi,tlogDataFlowi,j,t + εi,t Centrality Step 1 of 2-step ECM log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 + θjcentralityi,tlogDataFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t Centrality Step 2 of 2-step ECM Exhibit A15 Formulas Y = A × Kα × Lβ (1) Cobb-Douglas production function logGDPi,t = α + SkβklogCVi,k,t + SiβilogFlowi,j,t–1 + εi,t Econometric step 1 of 2-step ECM ∆logGDPi,t = α + Skβk∆logCVi,k,t + Sjβj∆logFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t Econometric step 2 of 2-step ECM flow’s GDP share in leveli,j,t = (flowi,j,t) βj real GDPi,t Flow’s share in GDP flow’s GDP share in growthi,j,t = (log(flowi,j,t) – log(flowi,j,t–1)) × βj (log(real GDPi,t) – log(real GDPi,t–1)) Flow’s share in growth log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 + θjcentralityi,tlogDataFlowi,j,t + εi,t Centrality Step 1 of 2-step ECM log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 + θjcentralityi,tlogDataFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t Centrality Step 2 of 2-step ECM 118 McKinsey Global Institute Technical appendix Impact on productivity Flows may increase GDP by raising either the usage of capital and labor, or by raising productivity. To calculate this impact, we run a separate model specification that includes GDP/employment, or labor productivity.149 The difference between the coefficient on this variable and the coefficient of flows on GDP is the residual, which is the impact on productivity. The results are shown in Exhibit A7. We find that all flows affect GDP mainly through productivity. Goods flows, FDI, and data flows all positively increase productivity. Migration flows have a negative impact on productivity for emerging economies. As noted above, this may reflect the impact of “brain drain” (the loss of skilled labor) or the difficulties developing countries encounter when absorbing large migrant or refugee flows. For advanced economies, we find a positive impact of migration on productivity. Exhibit A7 also shows that data flows have a positive impact on an economy’s utilization of capital and labor. Thus far, fears about digital flows reducing employment appear unfounded. 149 We run a similar regression using capital productivity and find similar results. Exhibit A6 All flows combined contributed 10.1 percent of GDP from 2003 to 2013, with goods and data having largest impact SOURCE: McKinsey Global Institute analysis Flows contribution to GDP FLOWS MODEL—ALL COUNTRIES Shares of output % Midpoint 5th percentile 95th percentile 2008–13 2003–13 1998–2013 2008–13 2003–13 1998–2013 2008–13 2003–13 1998–2013 All flows 9.36 10.05 11.03 8.97 9.62 10.55 10.06 10.82 11.90 Goods trade 3.20 3.51 3.97 3.28 3.56 4.03 3.21 3.46 3.91 FDI flow 1.68 1.64 1.52 1.74 1.70 1.58 1.62 1.58 1.46 Immigration 1.74 1.95 2.24 1.16 1.30 1.50 2.61 2.92 3.36 Data flow 2.70 2.95 3.29 2.79 3.06 3.43 2.62 2.85 3.17 Exhibit A7 All flows contribute to raising productivity, but only data flows contribute to increasing labor and capital inputs SOURCE: McKinsey Global Institute analysis Flows model Flow variables Long-term elasticity for real GDP Impact on productivity Impact on increased inputs Goods trade 0.05 0.05 0 FDI 0.04 0.05 -0.0023 Migration -0.05 -0.04 -0.0101 Data usage 0.02 0.02 0.0029 119McKinsey Global Institute Digital globalization: The new era of global flows Network centrality In a separate analysis, we investigated the importance of a country’s position within the network of trade flows and within the network of data flows for its GDP growth. More connections with a greater number of neighbors should reflect a more diverse portfolio of imports and data. A broader portfolio benefits a country by enriching the consumption basket, by enabling companies to source ideas and inputs from all over the world, and by taking advantage of the global competitive landscape to diversify and reduce dependence on any single partner. Broader network coverage for exporters reflects their competitiveness and ability to sell in many markets. More routes and a more central position in the network therefore indicate the presence of highly competitive firms that can participate in global trade and markets and thereby have a positive impact on their home country’s GDP. To estimate the impact of centrality, we use two measures: the eigenvector centrality of data flow and the number of routes. The eigenvector centrality of data flow is the position in a network from calculating the eigenvector based on a country’s neighbors’ positions. It is the data-trading partners’ network rather than direct links that affects a country’s eigenvector centrality. In contrast, the number of routes directly measures the connections of a given country with other countries. We use the same model specification in terms of control variables as we do for the combination of flows. Centrality is introduced as an interaction term with log of data flow for the long run and with the difference in log of data flow: Step 1 of two-step ECM Step 2 of two-step ECM Correlation between the data flow and its interaction with either measure of centrality is found to be negligible. When the centrality has a dampening effect on GDP through data flow, we expect θj to be less βj than for the data flow. We observe that countries on the periphery of the data flows actually benefit more than those at the center of the data flows. 3. ACADEMIC LITERATURE ON THE RELATIONSHIP BETWEEN GLOBAL FLOWS AND GDP A large body of academic literature has examined the impact of different types of global flows on GDP, productivity, and innovation. Generally, these studies are based on one type of flow (i.e., trade, financial flows, or immigration) and often on one country. Some have created their own indexes of globalization including trade and other flows, as well as other metrics such as trade restrictions, enhanced technology transmission, and improvements in macroeconomic policy. From this literature, we find general consensus that trade, FDI, and immigration support higher levels and growth rates of GDP and also of productivity growth. However, the size estimates of this impact vary across studies. These estimates are shown in Exhibit A8, along with our own, which are in a similar range. Exhibit A15 Formulas Y = A × Kα × Lβ (1) Cobb-Douglas production function logGDPi,t = α + SkβklogCVi,k,t + SiβilogFlowi,j,t–1 + εi,t Econometric step 1 of 2-step ECM ∆logGDPi,t = α + Skβk∆logCVi,k,t + Sjβj∆logFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t Econometric step 2 of 2-step ECM flow’s GDP share in leveli,j,t = (flowi,j,t) βj real GDPi,t Flow’s share in GDP flow’s GDP share in growthi,j,t = (log(flowi,j,t) – log(flowi,j,t–1)) × βj (log(real GDPi,t) – log(real GDPi,t–1)) Flow’s share in growth log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 + θjcentralityi,tlogDataFlowi,j,t + εi,t Centrality Step 1 of 2-step ECM log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 + θjcentralityi,tlogDataFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t Centrality Step 2 of 2-step ECM Exhibit A15 Formulas Y = A × Kα × Lβ (1) Cobb-Douglas production function logGDPi,t = α + SkβklogCVi,k,t + SiβilogFlowi,j,t–1 + εi,t Econometric step 1 of 2-step ECM ∆logGDPi,t = α + Skβk∆logCVi,k,t + Sjβj∆logFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t Econometric step 2 of 2-step ECM flow’s GDP share in leveli,j,t = (flowi,j,t) βj real GDPi,t Flow’s share in GDP flow’s GDP share in growthi,j,t = (log(flowi,j,t) – log(flowi,j,t–1)) × βj (log(real GDPi,t) – log(real GDPi,t–1)) Flow’s share in growth log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 + θjcentralityi,tlogDataFlowi,j,t + εi,t Centrality Step 1 of 2-step ECM log GDPi,t = α + SkβklogCVi,k,t + SjβjlogFlowi,j,t–1 + θjcentralityi,tlogDataFlowi,j,t + γlogGDP per capitai,t–2 + δεi,t–1 + ui,t Centrality Step 2 of 2-step ECM 120 McKinsey Global Institute Technical appendix Exhibit A8 Academic literature on flows and GDP growth SOURCE: McKinsey Global Institute analysis 10% increase in flow results in the following % increase in GDP Flow Literature review MGI model Reference literature Goods trade 0.2–1.5 0.5 Frankel and Romer, “Does trade cause growth?” American Economic Review, 1999 Bianjing, Zuoshi, and Jingkui, Endogenous international trade and economic growth: Empirical study based on 120 Chinese cities, 2011 Wacziarg, “Measuring the dynamics gains from trade,” World Bank Economic Review, 2001 US Executive Office of the President, The economic benefits of US trade, 2015 De Loecker and Goldberg, “Firm performance in a global market,” Annual Review of Economics, 2013 FDI flow 0.04–0.5 0.4 Aizenmann, Jinjarak, and Park, “Capital flows and economic growth in the era of financial integration and crisis,” Open Economies Review, 2013 Bordo, Meissner, and Stuckler, “Foreign currency debt, financial crises and economic growth: A long-run view,” Journal of International Money and Finance, 2010 Reinhart and Reinhart, “Capital flow bonanzas: An encompassing view of the past and present,” in NBER International Seminar on Macroeconomics 2008, 2009 Kose, Prasad, and Terrones, Does openness to international financial flows raise productivity growth? NBER, 2008 Migration1 1.0–5.0 -0.51 Peri, The effect of immigration on productivity: Evidence from US states, 2009 Boubtane, Dumont, and Rault, Immigration and economic growth in the OECD countries, 1986–2006, IZA discussion paper, 2014 US Executive Office of the President, The economic benefits of fixing our broken immigration system, 2013 Internet penetration 0.2–1.4 0.2 Qiang and Rossotto with Kimura, “Economic impacts of broadband,” in Information and communications for development 2009: Extending reach and increasing impact, World Bank, 2009 Freund and Weinhold, “The effect of the Internet on international trade,” Journal of International Economics, 2004 Meijers, “Does the Internet generate economic growth, international trade, or both?” International Economics and Economic Policy, 2014 Falk and Hagsten, E-commerce trends and impacts across Europe, UNCTAD discussion paper, 2015 USITC, Digital trade in the US and global economies, part 2, 2014 Onyejiwu, Inter-country variations in digital technology in Africa, WIDER discussion paper, 2002 1 W e find that migration flows are negligible or slightly negative at the global level, possibly due to the loss of skilled labor in developing countries or the difficulties of absorbing a large influx of refugees or migrants. However, we find that migration flows have a positive impact on productivity in advanced economies, consistent with other academic literature. 121McKinsey Global Institute Digital globalization: The new era of global flows Flows of goods and services The academic literature provides a wide range of estimates on the impact of trade flows on GDP growth. Frankel and Romer use bilateral date for 63 countries for 1985 and estimate that a one percentage point increase in the trade-to-GDP ratio causes almost a 1.5 percentage point increase in per capita income growth.150 At the other end of the range, Bianjing, Zuoshi, and Jingkui, based on international trade of 120 cities, suggest that a 1 percent increase in international trade produces only a 0.19 to 0.22 percentage point increase in income.151 Wacziarg analyzed data for 57 countries from 1970 to 1989, measuring the impact of trade policy openness on economic growth, where openness is a function of investment, enhanced technology transmission, and improvements in macroeconomic policy. The analysis suggests that an 8.5 percentage point increase in the trade policy measure, corresponding roughly to one standard deviation, is associated with a 0.6 percentage point increase in annual GDP growth. Investment is the most important channel, accounting for almost two-thirds of this effect.152 The US Council of Economic Advisors reviewed the effects of trade and found that the reduction of trade barriers since World War II has raised US GDP by 7.3 percent, or approximately $1.3 trillion in 2014. It also found that a 10 percent increase in an industry’s exports is associated with a 0.2 percent increase in that industry’s labor productivity. Based on the average industry’s increase in exports, international trade may have been responsible for about one-quarter of total US productivity growth over the 1990s and 2000s. The paper also emphasizes that trade spurs both labor productivity and innovation, as measured by total factor productivity. A 10 percentage point decrease in tariffs corresponds to about a 0.4 percentage point increase in labor productivity growth and about a half percentage point increase in TFP growth over the two decades.153 Other studies have also examined the impact of trade on productivity growth. De Loecker and Goldberg review the literature and conclude that there is strong evidence that globalization raises firm-level productivity.154 The econometric analysis in this report is consistent with the low end of the range. We find that a 1 percent increase in goods trade to GDP results in a 0.05 percent growth in GDP. This may reflect the fact that in our analysis, we control for other types of flows, such as data flows and financial flows, that may accompany trade flows. 150 Jeffrey A. Frankel and David Romer, “Does trade cause growth?” American Economic Review, volume 89, issue 3, June 1999. 151 Ma Bianjing, Xie Zuoshi, and Li Jingkui, Endogenous international trade and economic growth: An empirical study based on 120 Chinese cities, available at SSRN, August 2011. 152 Romain Wacziarg, “Measuring the dynamics gains from trade,” World Bank Economic Review, volume 15, number 3, October 2001. 153 The economic benefits of US trade, Executive Office of the President, May 2015. 154 Jan De Loecker and Pinelopi Koujianou Goldberg, “Firm performance in a global market,” The Annual Review of Economics, October 2013. 122 McKinsey Global Institute Technical appendix FDI and other financial flows The literature on the impact of financial flows on GDP growth is mixed. In general, academic studies find that FDI has a positive impact on GDP growth, while broader financial flows and capital account openness have a mixed effect. For instance, Mun et al. find that a 1 percent increase in FDI flows relative to GDP result in a 0.05 percent increase in GDP growth.155 This is based on analyzing data from Malaysia spanning 1970 to 2005. Similarly, Aizenman, Jinjarak, and Park examine a sample of 100 countries using data from 1990 to 2010 and find that a one standard deviation increase in FDI inflow has increases the growth of GDP per capita by 0.90 to 0.94 percent.156 The impact of broader financial flows (including cross-border lending and portfolio purchases of equities and bonds) on GDP growth is less clear. Bordo et al. find that flows of foreign currency debt increase the risk of financial crises, based on their study of foreign currency debt, financial crises, and short- and long-term output effects from 1880 to 1913 and from 1973 to 2003 for 45 countries.157 Similarly, Reinhart and Reinhart examine a large group of countries over nearly 50 years and find that episodes of heavy capital inflows are associated with a higher incidence of banking, currency, and inflation crises, particularly in developing countries.158 While Kose et al. find that a 10 percentage point increase in ratio of FDI and equity liabilities to GDP would be associated with a 0.4 percentage point increase in TFP, a similar increase in ratio of debt liabilities to GDP would lead to a decrease in TFP growth of 0.2 percentage points.159 Migration flows The economic literature typically finds a positive impact of immigration on GDP growth for the destination country receiving migrants. Peri analyzes all 50 US states plus Washington, DC, in census years between 1960 and 2006; he finds that when immigrants produce a 1 percent increase in employment in a US state, income per worker rises by 0.5 percent.160 Boubtane and Dumont observe that a one percentage point increase in immigration results in 0.1 percentage point increase in economic growth for 22 OECD countries.161 The US Council of Economic Advisors finds that enhancing immigration reforms that increase immigration would increase real GDP relative to current projections by 3.3 percent by 2023 and 5.4 percent by 2033.162 Our research finds a negative, albeit small, impact of immigration on GDP growth. This may be the result of developing countries losing skilled talent or poorer countries encountering difficulties in absorbing a large influx of refugees or unskilled immigrants. We find, however, that migration flows have a positive impact on total factor productivity for advanced economies, which is consistent with other literature.163 155 Har Wei Mun, Teo Kai Lin, and Yee Kar Man, “FDI and economic growth relationship: An empirical study on Malaysia,” International Business Research, volume 1, number 2, April 2008. 156 Joshua Aizenman, Yothin Jinjarak, and Donghyun Park, “Capital flows and economic growth in the era of financial integration and crisis, 1990–2010,” Open Economies Review, volume 24, issue 3, July 2013. 157 Michael D. Bordo, Christopher M. Meissner, and David Stuckler, “Foreign currency debt, financial crises and economic growth: A long-run view,” Journal of International Money and Finance, volume 29, 2010. 158 Carmen Reinhart and Vincent Reinhart, “Capital flow bonanzas: An encompassing view of the past and present,” in NBER International Seminar on Macroeconomics 2008, Jeffrey Frankel and Christopher Pissarides, eds., University of Chicago Press, 2009. 159 M. Ayhan Kose, Eswar S. Prasad, and Marco E. Terrones, Does openness to international financial flows raise productivity growth? NBER working paper number 14558, December 2008. 160 Giovanni Peri, The effect of immigration on productivity: Evidence from US states, NBER working paper number 15507, November 2009. 161 Ekrame Boubtane, Jean-Christophe Dumont, and Christophe Rault, Immigration and economic growth in the OECD countries, 1986–2006, IZA discussion paper number 8681, November 2014. 162 The economic benefits of fixing our broken immigration system, Executive Office of the President, July 2013. 163 For instance, see Ekrame Boubtane, Jean-Christophe Dumont, and Christophe Rault, Immigration and economic growth in the OECD countries, 1986–2006, IZA discussion paper number 8681, November 2014. 123McKinsey Global Institute Digital globalization: The new era of global flows Data flows Most of the literature available on the impact of data flows on economic growth relates to increase in Internet penetration. For example, Choi and Yi use data from 207 countries from 1991 to 2000 and find that that a one percentage point increase in the Internet user ratio leads to a 0.057 percentage point increase in GDP.164 Meijers finds that a 10 percentage point increase of Internet penetration leads to a 0.17 percentage point increase of economic growth and an increase in international trade.165 A World Bank study for 120 countries from 1980 to 2006 reports that a 10 percent increase in broadband penetration resulted in a 1.38 percent point increase in GDP growth in developing countries and a 1.21 percent point increase in growth in developed countries.166 Increases in Internet usage can also promote more trade. One study concludes that a 10 percent increase in Internet access leads to a 0.2 percent increase in exports.167 A recent study of EU firms also found that engaging in e-commerce increases labor productivity and that e-commerce accounted for 17 percent of EU labor productivity growth between 2003 and 2010.168 A 2014 study by the US International Trade Commission (ITC) calculated the productivity gains from the Internet by surveying US businesses and converting the results into an economic model. The ITC found that the productivity gains from the Internet have increased US real GDP by 3.4 to 3.5 percent.169 4. METHODOLOGY FOR THE MGI CONNECTEDNESS INDEX The MGI Connectedness Index ranks countries on the extent of their engagement with the global economy through inflows and outflows of goods, services, finance, people, and data. To obtain a more granular picture, the index ranks countries on their connectedness to each individual flow as well as compiling an aggregate score. We consider each country’s inflows and outflows of goods, services, finance, people, and data. Financial flows include FDI, equity, debt, and other flows (mainly cross-border lending). For people flows, we consider countries’ connectedness in terms of the stock of foreign- born migrants resident in a given country and that country’s citizens living abroad. After assigning each country a score for each type of flow, we weight those scores equally to obtain that country’s overall connectedness score. Our methodology differs from other globalization indexes in both the weighting and the data used (Exhibit A9). Our 2014 index assesses 139 countries that provide data for 2014 for each of these flows. For people flows, however, data are available only through 2013. 164 Changkyu Choi and Myung Hoon Yi, “The effects of the Internet on economic growth: Evidence from cross- country panel data,” Economic Letters, volume 105, issue 1, October 2009. 165 Huub Meijers, “Does the Internet generate economic growth, international trade, or both?” International Economics and Economic Policy, volume 11, issue 1, February 2014. 166 Christine Zhen-Wei Qiang and Carlo M. Rossotto with Kaoru Kimura, “Economic impacts of broadband,” in Information and communications for development 2009: Extending reach and increasing impact, World Bank, 2009. 167 Caroline L. Freund and Diana Weinhold, “The effect of the Internet on international trade,” Journal of International Economics, volume 62, 2004. 168 Martin Falk and Eva Hagsten, E-commerce trends and impacts across Europe, UNCTAD discussion paper number 220, March 2015. 169 United States International Trade Commission, Digital trade in the US and global economies, part 2, August 2014. 124 McKinsey Global Institute Technical appendix Exhibit A9 The MGI Connectedness Index measures five types of inflows and outflows, unlike other studies SOURCE: McKinsey Global Institute analysis MGI DHL/Ghemawat E&Y/EIU KDF Overview 139 countries in 2014 1980–2014 5 dimensions: goods, services, finance, people, and data 140 countries in 2014 2012–14 5 dimensions: goods, services, finance, people, and data and communications 60 countries in 2012 2009–12 6 dimensions: goods, services, finance, people, data and communications, and culture 187 countries in 2015 1970–2015 7 dimensions: goods, services, finance, people, data and communications, culture, and political globalization D im ensions Goods 20 Total goods flows (100%) 35 Total goods flows (75%) 22 Total goods and services flows (40%) Trade openness1 barriers (10%) Tariff and non-tariff1 barriers (10%) Ease of trading1 (10%) Current account restrictions (10%) Share of main trading partners in total trade (20%) 36 Total goods and services flows (11%) Hidden trade barriers (12%) Mean tariff (14%) Taxes on international trade (13%) Services 20 Total goods flows (100%) Total service trade (25%) Financial 20 FDI flows (40%) Portfolio investment flows (10%) Bank and other flows (10%) Foreign investment assets and liabilities (40%) FDI stocks (25%) FDI flow (25%) Portfolio equity stock (25%) Portfolio equity flows (25%) FDI stock (50%) Portfolio capital flows (8%) Policy toward FDI1 (8%) Domestic favoritism1 (8%) Expropriation risk (8%) State control (8%) FDI stocks (13%) Portfolio investment stocks (12%) Income payments to foreign nationals/GDP flows (13%) Restrictions on capital account (11%) People 20 Immigrant stock (80%) Travelers flow (20%) 15 Immigrant stock (33%) Travelers flow (33%) International student flow (33%) 19 Net immigration rate (40%) Travelers flow (40%) Hiring of foreign nationals (20%) 36 Immigrant stock (21%) Travelers flow (26%) International calls flow (25%) International letters flow (25%) Data and communi- cation 20 Cross-border used Internet bandwidth (100%) 15 International bandwidth stock (40%) International calls flow (40%) Traded publications (20%) 21 ICT goods flows (30%) Creative goods flows (30%) Broadband subs stock (20%) Internet subs stock (20%) Internet users stock (36%) Television stock (38%) Trade in newspapers flows (26%) Cultural/ political 17 Cultural integration Travelers flow (33%) International fixed telephone call (33%) Openness to foreign culture influence1 (33%) Cultural globalization McDonald’s restaurants (44%) IKEA stores (44%) Trade in books (11%) 26 Political globalization Embassies, member- ships, UN Security Council missions, international treaties 1 Elasticity for TFP with respect to flows is calculated by subtracting the elasticity for labor (or capital) productivity from that for GDP. Dimension variables (variable weight) Percentage weight in overall indexXX 125McKinsey Global Institute Digital globalization: The new era of global flows Normalization and ranking We assess each country’s connectedness within each type of flow by looking at two dimensions: flow intensity and flow share. Flow intensity measures the size of a given flow as a share of a given country’s GDP or population. As an illustration, Germany’s flow intensity in the goods trade is 71 percent— that is, the value of all goods imported to and exported from Germany in 2014 was equivalent to 71 percent of Germany’s GDP. It is important to note that flows in goods and services are measured in nominal values, while national accounts capture GDP as value added. This is why the ratio of trade to GDP can easily exceed 100 percent. The intensity of goods, services, and financial flows is calculated relative to GDP. The intensity of people and data flows is measured relative to the size of a country’s population. Relying on flow intensity alone would artificially boost small countries over those with large, diversified domestic economies in the rankings. To correct for this, we include a measure of each country’s share of the global total of each flow, which we call flow share. Germany’s flow share for goods is 7 percent—that is, Germany accounted for 7 percent of all global inflows and outflows of goods in 2014. To combine these two measures of connectedness into an index and calculate composite trade intensity, we use the following methodology: To smooth the distribution between countries, we use the resulting figure to assign countries a normalized score relative to other countries on a scale of 1 to 100. This normalized score can be used to rank all the countries in each of the five types of flows. Exhibit A16 Moar ormulas Compound flow intensity score Compound flow intensity score = (n–1) × (inflow + outflow)2 GDP × global flow Normalized compound flow intensity score *** SAME for regions Normalized compound flow intensity score = 99 × (Compound flow intensity score – min) (max – min) + 1 Regional compound score Compound flow intensity score = (nregions –1) × (inflowregion + outflowregion)2 GDPregion × country flow Regions vs. countries Compound flow intensity score = (ncountries –1) × (inflowregion + outflowregion)2 GDPregion × world flow Exhibit A16 Moar ormulas Compound flow intensity score Compound flow intensity score = (n–1) × (inflow + outflow)2 GDP × global flow Normalized compound flow intensity score *** SAME for regions Normalized compound flow intensity score = 99 × (Compound flow intensity score – min) (max – min) + 1 Regional compound score Compound flow intensity score = (nregions –1) × (inflowregion + outflowregion)2 GDPregion × country flow Regions vs. countries Compound flow intensity score = (ncountries –1) × (inflowregion + outflowregion)2 GDPregion × world flow 126 McKinsey Global Institute Technical appendix Subcomponents of the scores on people and financial flows The scores for people flows consider two components: migrants and travelers. The migrant and traveler compound intensity scores are calculated separately following the methodology described above. The overall score for people flows is a weighted average of the migrant score and the traveler score. We assign the migrant score a weighting of 80 percent, while travelers account for 20 percent of the overall score. Financial flows have four components: FDI, portfolio investment flows (equity and bonds), other financial flows (loans and deposits), and the stock of foreign financial assets and liabilities. The overall financial flows score is a weighted average of these four flows. FDI and foreign assets and liabilities stock are given weights of 40 percent each, while portfolio investment flows and other flows have weights of 10 percent. A country’s overall connectedness score is calculated by weighing the score on each of the five types of flows equally and calculating a simple average. This score then determines each country’s position in the overall rankings. The MGI Connectedness Index: Full rankings The abbreviated version of the MGI Connectedness Index that appears in Chapter 3 contains the top 25 countries plus a selection of other major economies. Exhibits A10 through A12, however, list the full rankings for all 139 countries. As noted in Chapter 3, more countries are participating in global flows today, but we also observe that flows remain concentrated among a small set of highly connected countries. We can see this from the connectedness score of each country on each flow. Exhibit A13 shows the top 15 countries in each flow. The score for the top country is normalized to 100, so the scores for the other countries can be interpreted in relation to the most connected country. We see that data flows are the most heavily concentrated, with the steepest drop- offs from the leaders. Service flows are the second most concentrated. Goods flows are the least concentrated, with country scores declining less rapidly after the leading country. Financial flows and people flows are similar, and people flows in particular have higher scores for a larger set of countries. 127McKinsey Global Institute Digital globalization: The new era of global flows Exhibit A10 MGI Connectedness Index (1/3) Country connectedness index and overall flows data, 2014 Rank of participation by flow as measured by flow intensity and share of world total 1–10 11–25 26–50 >50Connectivity index rank 100+ <7070–99Flow intensity Rank Country Score Connectedness Index rank Flow value1 $ billion Flow intensity2 % of GDPGoods Services Finance People Data 1 Singapore 64.2 1 2 2 12 6 1,392 452 2 Netherlands 54.3 3 3 6 21 1 1,834 211 3 United States 52.7 7 7 3 1 7 6,832 39 4 Germany 51.9 2 4 8 3 2 3,798 99 5 Ireland 45.9 32 1 1 28 9 559 227 6 United Kingdom 40.8 13 5 5 6 3 2,336 79 7 China 34.2 4 16 4 82 38 6,480 63 8 France 30.1 11 8 9 7 4 2,262 80 9 Belgium 28.0 5 6 33 33 8 1,313 246 10 Saudi Arabia 22.6 20 28 27 2 53 790 106 11 United Arab Emirates 22.2 6 23 17 4 46 789 196 12 Switzerland 18.0 12 11 10 17 13 848 115 13 Canada 17.3 16 22 11 11 18 1,403 79 14 Russia 16.1 21 25 18 5 25 1,059 57 15 Spain 14.4 25 13 19 14 16 1,105 79 16 Korea 14.0 8 12 28 50 44 1,510 107 17 Italy 13.4 17 18 24 16 19 1,587 74 18 Sweden 13.0 29 14 22 31 5 572 100 19 Austria 11.7 26 17 31 20 12 470 108 20 Malaysia 11.6 9 19 25 26 43 610 187 21 Mexico 10.7 14 63 34 18 41 1,022 80 22 Thailand 10.7 10 15 36 44 64 605 162 23 Kuwait 10.6 37 46 13 13 75 306 153 24 Japan 10.5 15 20 12 81 20 2,498 54 25 Kazakhstan 10.0 48 73 41 8 57 176 83 26 Ukraine 9.8 38 39 87 10 34 133 101 27 Australia 9.7 30 34 21 15 33 825 57 28 Denmark 8.9 35 9 32 41 11 369 108 29 Jordan 8.8 73 50 75 9 83 50 138 30 India 8.5 24 10 35 58 70 1,316 64 31 Qatar 7.8 33 35 29 19 59 300 141 32 Czech Republic 7.5 18 33 57 59 15 397 193 33 Malta 7.4 97 26 7 90 50 31 308 34 Poland 7.0 23 31 47 34 22 585 107 35 Hungary 6.8 22 30 26 62 17 287 209 36 Norway 6.0 36 24 20 46 24 458 92 37 Vietnam 5.7 19 54 45 103 61 350 188 38 Lebanon 5.6 82 21 46 22 103 69 151 39 Finland 5.5 46 27 23 70 10 390 144 40 Portugal 5.5 47 36 30 23 31 255 111 41 Turkey 5.1 28 40 53 38 29 521 65 42 Slovak Republic 5.0 27 60 68 67 14 205 205 43 Israel 4.9 51 32 49 24 56 248 82 44 Brazil 4.5 41 38 14 125 30 869 37 45 Chile 4.1 45 58 16 102 27 239 92 46 Belarus 4.1 40 66 101 29 47 92 121 47 Greece 4.1 60 29 54 35 42 160 67 SOURCE: McKinsey Global Institute analysis 1 Flows value represents total goods, services, and financial inflows and outflows. 2 Flow intensity represents the total value of goods, services, and financial flows as a share of the country’s GDP. 128 McKinsey Global Institute Technical appendix Exhibit A11 MGI Connectedness Index (2/3) Country connectedness index and overall flows data, 2014 Rank of participation by flow as measured by flow intensity and share of world total 1–10 11–25 26–50 >50Connectivity index rank 100+ <7070–99Flow intensity Rank Country Score Connectedness Index rank Flow value1 $ billion Flow intensity2 % of GDPGoods Services Finance People Data 48 New Zealand 3.9 67 48 61 25 51 130 63 49 Romania 3.9 39 51 83 36 28 194 97 50 Croatia 3.7 76 45 104 27 37 57 100 51 Indonesia 3.4 31 49 38 106 76 504 57 52 Mozambique 3.3 95 70 15 117 110 40 246 53 South Africa 3.3 34 57 52 64 80 277 79 54 Philippines 3.2 54 41 44 52 67 230 81 55 Bulgaria 3.1 49 53 67 48 23 92 165 56 Albania 3.1 114 72 79 30 73 16 117 57 Oman 3.1 44 65 55 54 66 121 148 58 Bosnia and Herzegovina 3.0 86 123 113 32 62 21 117 59 Lithuania 2.8 43 55 112 68 35 87 181 60 Cote d'Ivoire 2.7 80 104 136 37 114 28 82 61 Slovenia 2.7 42 56 64 75 36 105 212 62 Pakistan 2.7 78 91 84 39 88 116 47 63 Azerbaijan 2.6 75 62 37 57 69 92 122 64 Morocco 2.6 58 43 74 56 65 104 97 65 Estonia 2.6 56 47 60 72 21 54 209 66 Bangladesh 2.6 71 99 62 43 113 109 62 67 Serbia 2.5 74 61 103 45 45 52 118 68 Bahrain 2.4 65 118 56 49 58 28 82 69 Moldova 2.4 105 102 102 40 52 12 154 70 Cyprus 2.3 122 37 43 76 55 18 79 71 Jamaica 2.3 115 69 100 42 72 17 113 72 Argentina 2.3 64 68 63 60 32 198 37 73 Egypt 2.2 68 42 69 73 71 158 55 74 Colombia 2.2 61 89 40 83 54 197 52 75 Latvia 2.2 66 67 76 66 26 51 158 76 Armenia 2.2 121 97 99 47 81 12 113 77 Libya 2.2 53 78 59 84 108 65 159 78 Panama 2.1 69 44 42 129 39 74 161 79 Dominican Republic 2.1 94 77 93 53 82 41 64 80 El Salvador 2.1 98 110 94 51 89 26 104 81 Algeria 2.0 52 82 91 91 85 152 71 82 Angola 1.9 50 64 86 134 111 100 76 83 Nigeria 1.9 55 76 48 128 98 268 47 84 Burkina Faso 1.9 123 117 139 55 134 8 67 85 Venezuela 1.9 57 85 71 86 60 172 34 86 Peru 1.8 62 88 51 104 49 122 60 87 Macedonia, FYR 1.8 93 101 124 63 48 18 156 88 Georgia 1.8 106 79 77 65 63 20 123 89 Sri Lanka 1.8 91 81 81 69 93 56 75 90 Guyana 1.7 118 128 116 61 125 4 133 91 Brunei 1.7 89 138 58 74 95 25 146 92 Cambodia 1.7 70 71 72 94 112 35 210 93 Ecuador 1.7 72 112 111 80 79 67 66 SOURCE: McKinsey Global Institute analysis 1 Flows value represents total goods, services, and financial inflows and outflows. 2 Flow intensity represents the total value of goods, services, and financial flows as a share of the country’s GDP. 129McKinsey Global Institute Digital globalization: The new era of global flows Exhibit A12 MGI Connectedness Index (3/3) Country connectedness index and overall flows data, 2014 Rank of participation by flow as measured by flow intensity and share of world total 1–10 11–25 26–50 >50Connectivity index rank 100+ <7070–99Flow intensity Rank Country Score Connectedness Index rank Flow value1 $ billion Flow intensity2 % of GDPGoods Services Finance People Data 94 Tunisia 1.7 63 74 108 107 78 53 110 95 Mongolia 1.6 99 98 39 136 86 23 194 96 Kyrgyz Republic 1.6 102 87 88 78 118 13 173 97 Paraguay 1.6 83 133 121 77 94 26 84 98 Iran, Islamic Rep. 1.6 59 86 106 127 96 185 45 99 Costa Rica 1.5 81 75 65 101 74 43 87 100 Lao PDR 1.5 125 131 122 71 120 8 68 101 Ghana 1.5 77 80 126 109 100 32 83 102 Suriname 1.4 116 134 133 79 119 5 83 103 Liberia 1.4 134 121 50 100 139 7 333 104 Bolivia 1.4 84 108 85 93 84 33 98 105 Honduras 1.4 79 100 80 112 99 29 148 106 Yemen 1.4 90 116 114 88 116 26 70 107 Iceland 1.4 107 59 98 123 40 20 120 108 Guatemala 1.4 88 105 92 99 92 44 75 109 Montenegro 1.4 131 111 89 85 77 5 105 110 Uruguay 1.4 103 95 70 95 68 35 60 111 Maldives 1.4 132 52 109 121 126 7 225 112 Nicaragua 1.4 92 113 90 97 90 17 144 113 Gabon 1.4 101 126 130 87 105 13 73 114 Tajikistan 1.4 119 114 78 89 127 11 120 115 Barbados 1.3 136 93 73 108 87 4 97 116 Fm Sudan 1.3 124 129 105 92 97 20 27 117 Mali 1.3 120 106 110 98 131 11 95 118 Kenya 1.3 100 84 127 119 91 35 58 119 Fiji 1.3 113 90 119 111 121 7 163 120 Congo, Dem. Rep. 1.3 110 120 66 118 138 15 46 121 Cape Verde 1.2 137 107 125 105 123 2 114 122 Lesotho 1.2 111 139 129 110 133 3 167 123 Samoa 1.2 138 135 132 96 129 1 121 124 Zambia 1.2 87 124 82 138 115 29 105 125 Senegal 1.2 109 103 117 120 106 15 93 126 Botswana 1.2 85 137 120 133 107 19 121 127 Namibia 1.2 96 122 96 132 101 16 120 128 Uganda 1.2 126 83 95 131 117 18 67 129 Guinea 1.2 130 132 107 113 136 5 82 130 Tanzania 1.2 108 94 97 135 104 25 51 131 Cameroon 1.2 112 96 137 126 124 17 54 132 Benin 1.2 127 125 128 115 132 7 78 133 Rwanda 1.2 135 127 135 114 130 5 64 134 Swaziland 1.1 117 115 138 124 128 5 143 135 Papua New Guinea 1.1 104 92 118 139 137 10 55 136 Belize 1.1 133 119 123 122 102 3 153 137 Grenada 1.1 139 136 134 116 109 1 96 138 Sierra Leone 1.1 128 130 115 130 135 5 96 139 Seychelles 1.1 129 109 131 137 122 3 179 SOURCE: McKinsey Global Institute analysis 1 Flows value represents total goods, services, and financial inflows and outflows. 2 Flow intensity represents the total value of goods, services, and financial flows as a share of the country’s GDP. 130 McKinsey Global Institute Technical appendix Exhibit A13 25.8 South Korea 45.9 29.6 China 28.6 46.9 Belgium Switzerland United Kingdom 87.8 France 24.5 24.5 28.2 Malaysia 30.0 Japan Mexico Thailand 44.1 United States United Arab Emirates 78.4 90.9 Germany 92.6 100.0 Netherlands Singapore People Data SOURCE: McKinsey Global Institute analysis Flows are concentrated among a few leading countries The 15 most connected countries within each type of flow, 2014 Connectedness score Goods Services Finance 100.0Ireland 38.5 Singapore 83.7 12.7 Spain 12.9 20.0 Thailand Sweden 14.5 South Korea 14.7 India Switzerland 15.5 16.8 Denmark France United States United Kingdom 29.9 33.5 27.5 34.7 Germany Netherlands Belgium 36.2 24.2 66.5 United Kingdom Switzerland Germany Netherlands 68.9 16.2 Malta 57.3 France 14.2 China Canada United States Ireland 18.8 18.3 25.0 25.2 Singapore 83.2 100.0 10.8 Brazil 10.8 Kuwait 11.4 Japan Mozambique Jordan 31.4 Singapore United Kingdom 33.7 Kazakhstan 40.8 Australia 26.6 Spain 30.6 Kuwait Canada Ukraine 35.1 36.7 37.0 39.9 France 100.0 51.2 81.7 Russia 44.1 Saudi Arabia United States 45.7 Germany 43.3 United Arab Emirates 3.2 3.4 3.5 5.8 6.7 6.8 Czech Republic Slovak Republic Switzerland Belgium Ireland Denmark Austria 10.3 14.9 Finland United States 18.8 34.8 Germany 20.6 France Sweden 100.0 United Kingdom 24.5 42.7 Netherlands 55.5 Singapore 131McKinsey Global Institute Digital globalization: The new era of global flows 5. METHODOLOGY FOR GLOBAL CONNECTEDNESS OF REGIONS WITHIN COUNTRIES In addition to measuring the global connectedness of countries, we have looked at variation in the global connectedness of states and provinces within five large countries: the United States, the United Kingdom, Germany, Brazil, and China. We were able to obtain data for goods flows (imports and exports) for specific regions within these five countries as well as immigration data. For the United States, we were also able to obtain FDI inflows and outflows and service exports by state. We did the regional analysis using two approaches. The first compared the global connectedness of different regions within a given country. The second treated each region as a country to see how it would compare if inserted into the MGI Connectedness Index for 2014. Regional comparison within each country For regional comparison within a country, we used a formula for compound intensity similar to that used in the MGI Connectedness Index: The key differences are that here regions are treated as the entities. Hence n becomes the total number of regions in the country, GDP is the GDP of the region, and inflows and outflows are the global flows for the region. We use the same method to normalize the flow intensity scores, i.e.: In this case, we use the maximum and minimum compound intensity scores among the regions within the country. For the United Kingdom, for example, we analyze 12 regions. The southeast has the highest compound intensity score, whereas Northern Ireland has the lowest compound intensity score. The results show considerable variation in the connectedness of regions within countries, particularly for China and the United States. Germany, in contrast, has the least variation among regions in international goods trade. Exhibit A16 Moar ormulas Compound flow intensity score Compound flow intensity score = (n–1) × (inflow + outflow)2 GDP × global flow Normalized compound flow intensity score *** SAME for regions Normalized compound flow intensity score = 99 × (Compound flow intensity score – min) (max – min) + 1 Regional compound score Compound flow intensity score = (nregions –1) × (inflowregion + outflowregion)2 GDPregion × country flow Regions vs. countries Compound flow intensity score = (ncountries –1) × (inflowregion + outflowregion)2 GDPregion × world flow Exhibit A16 Moar ormulas Compound flow intensity score Compound flow intensity score = (n–1) × (inflow + outflow)2 GDP × global flow Normalized compound flow intensity score *** SAME for regions Normalized compound flow intensity score = 99 × (Compound flow intensity score – min) (max – min) + 1 Regional compound score Compound flow intensity score = (nregions –1) × (inflowregion + outflowregion)2 GDPregion × country flow Regions vs. countries Compound flow intensity score = (ncountries –1) × (inflowregion + outflowregion)2 GDPregion × world flow 132 McKinsey Global Institute Technical appendix Comparison of regions with countries To compare regions with countries, the relevant formula for compound intensity score becomes: World flow is used instead of country flow to normalize the regional flow, since the region is treated as a separate country. The n continues to be the overall number of countries. Since the region is already included as a part of the country, the total world flow remains the same, and we keep n constant as the number of countries. To calculate the normalized compound intensity score, we use a similar technique as for countries, using the maximum and minimum compound intensity scores among all countries for normalizing. The results show that if states and provinces were countries, many would be among the most connected in the world. For instance, California would rank fourth in the world in people flows, while Guangdong would rank sixth in the world in goods flows. 6. GLOBAL SURVEY OF STARTUPS Background and methodology In collaboration with 1776, a global incubator and venture fund, and its Challenge Cup competition and Startup Federation programs, MGI undertook a survey and set of interviews to understand the extent to which it is now possible for startups to form global connections from their inception. The survey included: � Demographic questions (industry, company stage, home country, company age) � One core question regarding the types of global activities in which respondents participate (i.e., users in other countries, talent hired from other countries, funding from other countries, inputs from other countries, mentors/advisers in other countries, incubators in other countries) � Several targeted follow-up questions based on responses to the core question (e.g., for those who responded that they had users in other countries, the survey asked the number of countries with users, the share of user base that was international, and what that share would likely be in five years) � A final question asking respondents to rank the barriers to additional global activities. Exhibit A16 Moar ormulas Compound flow intensity score Compound flow intensity score = (n–1) × (inflow + outflow)2 GDP × global flow Normalized compound flow intensity score *** SAME for regions Normalized compound flow intensity score = 99 × (Compound flow intensity score – min) (max – min) + 1 Regional compound score Compound flow intensity score = (nregions –1) × (inflowregion + outflowregion)2 GDPregion × country flow Regions vs. countries Compound flow intensity score = (ncountries –1) × (inflowregion + outflowregion)2 GDPregion × world flow 133McKinsey Global Institute Digital globalization: The new era of global flows The research was conducted in October and November 2015 through two primary methods. The first involved e-mail outreach to 1776’s global community of startups and Startup Federation partners, Global Accelerator Network, and current and former participants in 1776’s Challenge Cup competition to encourage them to participate in the survey online. This resulted in 168 responses (62 percent of all responses). The second method involved in-person interviews of startup founders attending Challenge Cup events in Montreal (October 22, 2015), Washington, DC (October 29), Bogotá (November 5), Beijing (November 10), Budapest (November 12), Pretoria (November 18), and Mumbai (November 21). These discussions resulted in 103 responses (38 percent of all responses). � The sectors represented include education, health, cities and transportation, energy and sustainability, technology and communication, enterprise software, e-commerce, entertainment, and financial technology, among others. � Company stages include startups in concept phase (16 percent), product launch (24 percent), customer validation (21 percent), and growth/scaling phase (39 percent). � A majority of respondents (59 percent) said their startups were founded in 2014 or 2015, with the remainder founded between 2006 and 2013. � US startups accounted for 56 percent of respondents. Also represented were Asia and Australia (13 percent), Europe (11 percent), Africa and the Middle East (11 percent), and South America (5 percent). Despite this diversity, it is important to note that respondents are not typical of the average startup. Participation in cross-border networks such as the Startup Federation and pitch competitions like 1776’s global Challenge Cup suggest that respondents are likely more sophisticated, internationally minded, and technologically savvy than entrepreneurs or small businesses focused on delivering local services. Even with that caveat, however, the data suggest that there is a significant and influential subset of digitally empowered startups that are now global from inception. The 271 survey participants are a diverse group across a range of metrics (Exhibit A14). 134 McKinsey Global Institute Technical appendix Exhibit A14 MGI Global Startup Survey 2015: Respondent demographics SOURCE: McKinsey Global Institute Global Startup Survey 2015 100% = 271 respondents NOTE: Numbers may not sum due to rounding. Health 6 6 14 E-commerce Energy and sustainability 6 Enterprise software and security Technology and communication 12 Cities and transportation 7 21 Education 18 Other Financial technology 4 6Entertainment Concept 44 (16%) Customer validation 66 (24%) Growth/ scaling105 (39%) 56 (21%) Product launch 36 (13%) 2013 2006–11 2012 31 (11%) 45 (17%) 2014 80 (30%) 2015 79 (29%) Eastern Europe Other United States 4 7 India South America 2 Israel 4 3 5 China 5 Canada Western Europe 6 Africa 7 56 In-person interview 103 (38%) 168 (62%) Online survey Sector % Stage Number (%) Company age Number (%) Geography % Survey method Number (%) 135McKinsey Global Institute Digital globalization: The new era of global flows © Getty Images © Getty Images A Abeliansky, Ana L., Imnaculada Martínez-Zarzoso, and Klaus Prettner, The impact of 3D printing on trade and FDI, Center for European, Governance and Economic Development Research discussion paper number 262, October 2015. Aghion, Philippe, and Jeffrey G. Williamson, Growth, inequality, and globalization: Theory, history, and policy, Cambridge University Press, 1998. Ahmed, Shaghil and Andrei Zlate, “Capital flows to emerging market economies: A brave new world?” Journal of International Money and Finance, volume 48, November 2014. Aizenman, Joshua, Yothin Jinjarak, and Donghyun Park, “Capital flows and economic growth in the era of financial integration and crisis, 1990–2010,” Open Economies Review, volume 24, issue 3, July 2013. Akamai, State of the Internet Q1 2015 report, volume 8, number 1, 2015. Albright-Stonebridge Group, Data localization: A challenge to global commerce and the free flow of information, September 2015. AliResearch and Accenture, Cross-border B2C e-commerce market trends, June 2015. Autor, David H., David Dorn, and Gordon H. Hanson, The China shock: Learning from labor market adjustment to large changes in trade, NBER working paper number 21906, January 2016. Autor, David H., David Dorn, and Gordon H. Hanson, “The China syndrome: Labor market effects of import competition in the United States,” American Economic Review, volume 103, number 6, 2013. B Bahar, Dany, Ricardo Hausmann, and César A. Hidalgo, “Neighbors and the evolution of comparative advantage of nations: Evidence of international knowledge diffusion?” Journal of International Economics, volume 92, number 1, 2014. Barnett, George A., and Han Woo Park, “Examining the international Internet using multiple measures: New methods for measuring the communication base of globalized cyberspace,” Quality and Quantity, volume 48, issue 1, January 2014. Barro, Robert J., and Xavier Sala-i-Martin, Economic growth, MIT Press, 2003. Bauer, Matthias, Hosuk Lee-Makiyama, Erik van der Marel, and Bert Verschelde, The costs of data localisation: Friendly fire on economic recovery, European Centre for International Political Economy (ECIPE) occasional paper number 3/2014, May 2014. Bednárová, Pavla, “Global competitiveness of developed market economies and its relation to dimensions of globalization,” in Proceedings of the 11th International Conference Liberec Economic Forum, 2013. Bednárová, Pavla, Sárka Laboutková, and Ales Kocourek, “On the relationship between globalization and human development,” in Proceedings of the 10th International Conference Liberec Economic Forum, 2011. Bell, Jim, and Sharon Loane, “‘New-wave’ global firms: Web 2.0 and SME internationalisation,” Journal of Marketing Management, volume 26, issue 3–4, April 2010. BIBLIOGRAPHY 138 McKinsey Global Institute Bibliography Bergh, Andreas, and Therese Nilsson, “Do liberalization and globalization increase income inequality?” European Journal of Political Economy, volume 26, issue 4, December 2010. Bernard, Andrew B., and J. Bradford Jensen, “Exceptional exporter performance: Cause, effect, or both?” Journal of International Economics, volume 47, issue 1, February 1999. Bernard, Andrew B., J. Bradford Jensen, and Peter K. Schott, “Trade costs, firms, and productivity,” Journal of Monetary Economics, volume 53, issue 5, July 2006. Bianjing, Ma, Xie Zuoshi, and Li Jingkui, Endogenous international trade and economic growth: An empirical study based on 120 Chinese cities, available at SSRN, August 2011. Boubtane, Ekrame, Jean-Christophe Dumont, and Christophe Rault, Immigration and economic growth in the OECD countries, 1986–2006, Institute for the Study of Labor (IZA) discussion paper number 8681, November 2014. Borchert, Ingo, Batshur Gootiiz, and Aaditya Mattoo, Policy barriers to international trade in services: New empirical evidence, World Bank policy research working paper number 6109, June 2012. Busse, Matthias, and Jens Königer, Trade and economic growth: A re-examination of the empirical evidence, Hamburg Institute of International Economics research paper number 123, 2012. C Castro, Daniel, and Alan McQuinn, Cross-border data flows enable growth in all industries, Information Technology and Innovation Foundation, February 2015. Center for Strategic and International Studies and McAfee, Net losses: Estimating the global cost of cybercrime, June 2014. Centre for Entrepreneurship, SMEs and Local Development, Fostering SMEs’ participation in global markets: Final report, July 2013. Chakravorti, Bhaskar, Christopher Tunnard, and Ravi Shankar Chaturvedi, Digital planet: Readying for the rise of the e-consumer, The Fletcher School, Tufts University, September 2014. Chang, Chun-Ping, Chien-Chiang Lee, and Meng-Chi Hsieh, “Does globalization promote real output? Evidence from quantile cointegration regression,” Economic Modelling, volume 44, 2015. Chinn, David, James Kaplan, and Allen Weinberg, Risk and responsibility in a hyperconnected world: Implications for enterprises, McKinsey & Company and the World Economic Forum, January 2014. Choi, Changkyu, and Myung Hoon Yi, “The effects of the Internet on economic growth: Evidence from cross-country panel data,” Economic Letters, volume 105, issue 1, October 2009. Cisco, Cisco Visual Networking Index: Forecast and methodology, 2014–2019, May 2015. Cisco, Cisco Visual Networking Index: Global mobile data traffic forecast update, 2014– 2019, February 2015. D De Loecker, Jan, and Pinelopi Koujianou Goldberg, “Firm performance in a global market,” Annual Review of Economics, October 2013. Dewhurst, Martin, Jonathan Harris, and Suzanne Heywood, “Understanding your ‘globalization penalty,’” McKinsey Quarterly, July 2011. 139McKinsey Global Institute Digital globalization: The new era of global flows Dutta, Soumitra, Bruno Lanvin, and Sacha Wunsch-Vincent, eds., The global innovation index 2015: Effective innovation policies for development, Cornell University, Insead, and World Intellectual Property Organization, 2015. E eBay, Small online business growth report: Towards an inclusive global economy, January 2016. Economist Intelligence Unit, The ICT globalisation index, 2014. Economist Intelligence Unit, Networked manufacturing: The digital future, April 2014. Ellison, Nicole, “Connection strategies: Social capital implications of Facebook-enabled communication practices,” New Media & Society, volume 13, number 6, September 2011. Evenett, Simon J., and Johannes Fritz, The tide turns? Trade, protectionism, and slowing global growth, Centre for Economic Policy Research, 18th Global Trade Alert Report, 2015. Executive Office of the President, “The economic benefits of US trade,” May 2015. F Falk, Martin, and Eva Hagsten, E-commerce trends and impacts across Europe, UNCTAD discussion paper number 220, March 2015. Francois, Joseph, Bertin Martens, and Fan Yang, The macro-economic impact of cross- border ecommerce in the EU, European Commission Joint Research Centre, 2014. Frankel, Jeffrey A., and David Romer, “Does trade cause growth?” American Economic Review, volume 89, issue 3, June 1999. Friedman, Thomas L., The world is flat 3.0: A brief history of the twenty-first century, Picador, 2007. G George, Katy, Sree Ramaswamy, and Lou Rassey, “Next-shoring: A CEO’s guide,” McKinsey Quarterly, January 2014. Ghemawat, Pankaj, and Steven A. Altman, Depth Index of Globalization 2013: And the big shift to emerging economies, IESE Business School, University of Navarra, 2013. Ghemawat, Pankaj, and Steven A. Altman, DHL Global Connectedness Index 2014: Analyzing global flows and their power to increase prosperity, DHL, 2014. Google, Google economic impact report, United States 2014, August 2014. Grossman, Gene M., and Elhanan Helpman, “Globalization and growth,” American Economic Review, volume 105, number 5, May 2015. H Hatton, Timothy J., “The economics of international migration: A short history of the debate,” Labour Economics, volume 30, 2014. Hausmann, Ricardo, César A. Hidalgo, Sebastian Bustos, Michele Coscia, Alexander Simoes, and Muhammed A. Yildirim, The atlas of economic complexity: Mapping paths to prosperity (new edition), Center for International Development, Harvard University, 2013. Hofheinz, Paul, and Michael Mandel, Uncovering the hidden value of digital trade: Towards a 21st century agenda of transatlantic prosperity, Lisbon Council and Progressive Policy Institute interactive policy brief issue 19/2015, July 2015. 140 McKinsey Global Institute Bibliography I IAB Europe, Consumers driving the digital uptake: The economic value of online advertising- based services for consumers, September 2010. International Post Corporation, IPC cross-border e-commerce report, 2010. International Telecommunication Union and UNESCO, The state of broadband 2015, September 2015. J Jensen, J. Bradford, Global trade in services: Fear, facts, and offshoring, Peterson Institute for International Economics, September 2011. K Kaplan, James M., and Kayvaun Rowshankish, Addressing the impact of data location regulation in financial services, Global Commission on Internet Governance paper series number 14, Centre for International Governance Innovation and Chatham House, May 2015. Kelly, Sanja, Madeline Earp, Laura Reed, Adrian Shahbaz, and Mai Truong, Freedom on the Net 2014: Tightening the Net: Governments expand online controls, Freedom House, October 2014. Kilic, Cuneyt, Effects of globalization on economic growth: Panel data analysis for developing countries, University of Ploiesti Bulletin, volume 67, 2015. Kletzer, Lori G., “Why the US needs wage insurance,” Harvard Business Review, January 25, 2016. Koepke, Robin, Capital flows to emerging markets, Institute of International Finance, January 2015. Kose, M. Ayhan, Eswar S. Prasad, and Marco E. Terrones, Does openness to international financial flows raise productivity growth? NBER working paper number 14558, December 2008. Kumar, Nirmalya, and Phanish Puranam, “Emerging markets: Have you restructured for global success?” Harvard Business Review, October 2011. M Mandel, Michael, Beyond goods and services: The (unmeasured) rise of the data-driven economy, Progressive Policy Institute, October 2012. Mandel, Michael, Data, trade, and growth, Progressive Policy Institute, August 2013. Martens, Pim, Marco Caselli, Philippe De Lombaerde, Lukas Figge, and Jan Aart Scholte, “New directions in globalization indices,” Globalization, volume 12, issue 2, 2015. McAnaney, Patrick, Innovation that matters: How city networks drive civic entrepreneurship, 1776, US Chamber of Commerce Foundation, and Free Enterprise, May 2015. McKinsey Global Institute, Digital America: A tale of the haves and have-mores, December 2015. McKinsey Global Institute, Financial globalization: Retreat or reset? March 2013. McKinsey Global Institute, Global flows in a digital age: How trade, finance, people, and data connect the world economy, April 2014. McKinsey Global Institute, The Internet of Things: Mapping the value beyond the hype, June 2015. 141McKinsey Global Institute Digital globalization: The new era of global flows McKinsey Global Institute, A labor market that works: Connecting talent with opportunity in the digital age, June 2015. McKinsey Global Institute, Manufacturing the future: The next era of global growth and innovation, November 2012. McKinsey Global Institute, Playing to win: The new global competition for corporate profits, September 2015. McKinsey Global Institute, Urban world: The shifting global business landscape, October 2013. McKinsey Technology, Media and Telecom Practice, Offline and falling behind: Barriers to Internet adoption, September 2014. Meijers, Huub, “Does the Internet generate economic growth, international trade, or both?” International Economics and Economic Policy, volume 11, issue 1, February 2014. Melitz, Marc J., and Gianmarco I. P. Ottaviano, “Market size, trade, and productivity,” Review of Economic Studies, volume 75, issue 1, 2008. Meltzer, Joshua P., The importance of the Internet and transatlantic data flows for US and EU trade and investment, Global Economy and Development working paper number 79, Brookings Institution, October 2014. Meltzer, Joshua P., Maximizing the opportunities of the Internet for international trade, The E15 Initiative policy options paper, January 2016. Meltzer, Joshua P., Using the Internet to promote services exports by small-and medium- sized enterprises, Global Economy and Development working paper number 83, Brookings Institution, February 2015. Mettler, Ann, and Anthony D. Williams, The rise of the micro-multinational: How freelancers and technology-savvy start-ups are driving growth, jobs and innovation, Lisbon Council policy brief, June 2012. Miller, Ben, and Robert D. Atkinson, Digital drag: Ranking 125 nations by taxes and tariffs on ICT goods and services, Information Technology and Innovation Foundation, October 2014. Mun, Har Wei, Teo Kai Lin, and Yee Kar Man, “FDI and economic growth relationship: An empirical study on Malaysia,” International Business Research, volume 1, number 2, April 2008. N Neto, Delfim Gomes, and Francisco José Veiga, “Financial globalization, convergence and growth: The role of foreign direct investment,” Journal of International Money and Finance, volume 37, October 2013. Nicholson, Jessica R., and Ryan Noonan, Digital economy and cross-border trade: The value of digitally-deliverable services, US Department of Commerce, ESA issue brief number 01-14, January 2014. O OECD, Interconnected economies: Benefiting from global value chains, May 2013. OECD, OECD digital economy outlook 2015, 2015. OECD, Measuring the digital economy: A new perspective, December 2014. OECD, Supporting investment in knowledge capital, growth and innovation, October 2013. Olarreaga, Marcelo, and Sidley Austin LLP, Enabling traders to enter and grow on the global stage, eBay, 2012. 142 McKinsey Global Institute Bibliography Osterhammel, Jürgen, and Niels P. Petersson, Globalization: A short history, Princeton University Press, 2009. P Peri, Giovanni, The effect of immigration on productivity: Evidence from US states, NBER working paper number 15507, November 2009. Potrafke, Niklas, “The evidence on globalisation,” World Economy, volume 38, number 3, March 2015. Q Qiang, Christine Zhen-Wei and Carlo M. Rossotto with Kaoru Kimura, “Economic impacts of broadband,” in Information and communications for development 2009: Extending reach and increasing impact, World Bank, 2009. R Rao, B. Bhaskara, Artur Tamazian, and Krishna Chaitanya Vadlamannati, “Growth effects of a comprehensive measure of globalization with country-specific time series data,” Applied Economics, volume 43, issue 5, 2011. Reinhart, Carmen, and Vincent Reinhart, “Capital flow bonanzas: An encompassing view of the past and present,” in NBER International Seminar on Macroeconomics 2008, Jeffrey Frankel and Christopher Pissarides, eds., University of Chicago Press, 2009. S Samuelson, Paul A., “Where Ricardo and Mill rebut and confirm arguments of mainstream economists supporting globalization,” Journal of Economic Perspectives, volume 18, number 3, summer 2004. Sarac, Aysegul, Nabil Absi, and Stéphane Dauzère-Pérès, “A literature review on the impact of RFID technologies on supply chain management,” International Journal of Production Economics, volume 128, number 1, November 2010. Shahbaz, Muhammad, “Does trade openness affect long run growth? Cointegration, causality and forecast error variance decomposition tests for Pakistan,” Economic Modelling, volume 29, 2012. Smick, David, “Could globalization crack up?” International Economy, fall 2012. Spence, Michael, The next convergence: The future of economic growth in a multispeed world, Farrar, Straus and Giroux, 2011. Spence, Michael, and Sandile Hlatshwayo, The evolving structure of the American economy and the employment challenge, Council on Foreign Relations working paper, March 2011. Sui, Daniel, James Caverlee, and Dakota Rudesill, The Deep Web and the Darknet: A look inside the Internet’s massive black box, Woodrow Wilson International Center for Scholars, October 2015. T Tang, Kam Ki, “Correcting the size bias in trade openness and globalization measures,” Global Economy Journal, volume 11, issue 3, 2011. TeleGeography, Global Bandwidth Research Service; Global Bandwidth Forecast Service; Global Internet Geography, www.telegeography.com. Terziaa, Nuray, “The impact of e-commerce on international trade and employment,” Procedia Social and Behavioral Sciences, volume 24, 2011. http://www.telegeography.com 143McKinsey Global Institute Digital globalization: The new era of global flows Timmis, Jonathan, The Internet and international trade in goods, University of Nottingham School of Economics discussion paper, August 2012. U UNCTAD, Global value chains and development: Investment and value added trade in the global economy, February 2013. UNCTAD, Information economy report 2015: Unlocking the potential of ecommerce for developing countries, March 2015. UN High Commissioner on Refugees, UNHCR global trends report: World at war, June 2015. United Parcel Service, Change in the (supply) chain, 2015. United States International Trade Commission, Digital trade in the US and global economies, part 2, August 2014. V Villaverde, José, and Adolfo Maza, “Globalisation, growth and convergence,” World Economy, volume 34, issue 6, June 2011. W Wacziarg, Romain, “Measuring the dynamic gains from trade,” World Bank Economic Review, volume 15, number 3, October 2001. Westmore, Ben, International migration: The relationship with economic and policy factors in the home and destination country, OECD economics department working paper, July 2014. World Bank, “Potential macroeconomic implications of the Trans-Pacific Partnership,” in Global economic prospects: Spillovers amid weak growth, World Bank Group Flagship Report, January 2016. World Bank, World development report 2016: Digital dividends, January 2016. World Bank, World development report 2016: Internet for development, concept note, report number 91877, December 2014. World Trade Organization, World trade report 2011: The WTO and preferential trade agreements: From co-existence to coherence, 2011. RELATED MGI AND MCKINSEY RESEARCH Digital America: A tale of the haves and have-mores (December 2015) While the most advanced sectors, companies, and individuals push the boundaries of technology use, the US economy as a whole is realizing only 18 percent of its digital potential. Playing to win: The new global competition for corporate profits (September 2015) A 30-year period of unprecedented corporate- profit growth could be drawing to a close. Competition is intensifying as emerging-market companies go global and technology-enabled firms make rapid moves into new sectors. The Internet of Things: Mapping the value beyond the hype (June 2015) If policy makers and businesses get it right, linking the physical and digital worlds could generate up to $11.1 trillion a year in economic value by 2025. Global flows in a digital age: How trade, finance, people, and data connect the world economy (April 2014) The movement of goods and services, finance, and people has reached previously unimagined levels. Global flows are creating new degrees of connectedness among economies—and playing an ever-larger role in determining the fate of nations, companies, and individuals. Financial globalization: Retreat or reset? (March 2013) Cross-border capital flows remain far below their pre-crisis peak, and growth in financial assets around the world has stalled. Continued retrenchment could jeopardize investment unless policy makers can “reset” the financial system for a healthier flow of financing that supports economic growth. Disruptive technologies: Advances that will transform life, business, and the global economy (May 2013) This report identifies 12 technologies that could drive massive economic transformations in the coming years. It looks at how these technologies could change our world, as well as their benefits and challenges, and offers guidelines to help business and policy leaders respond. www.mckinsey.com/mgi E-book versions of selected MGI reports are available at MGI’s website, Amazon’s Kindle bookstore, and Apple’s iBooks Store. Download and listen to MGI podcasts on iTunes or at www.mckinsey.com/mgi/publications/multimedia/ Cover images all © Getty Images. Contents page images all © Getty Images. http://www.mckinsey.com/insights/high_tech_telecoms_internet/digital_america_a_tale_of_the_haves_and_have_mores http://www.mckinsey.com/insights/high_tech_telecoms_internet/digital_america_a_tale_of_the_haves_and_have_mores http://www.mckinsey.com/insights/corporate_finance/the_new_global_competition_for_corporate_profits http://www.mckinsey.com/insights/corporate_finance/the_new_global_competition_for_corporate_profits http://www.mckinsey.com/insights/business_technology/the_internet_of_things_the_value_of_digitizing_the_physical_world http://www.mckinsey.com/insights/business_technology/the_internet_of_things_the_value_of_digitizing_the_physical_world http://www.mckinsey.com/insights/globalization/global_flows_in_a_digital_age http://www.mckinsey.com/insights/globalization/global_flows_in_a_digital_age http://www.mckinsey.com/insights/globalization/global_flows_in_a_digital_age http://www.mckinsey.com/insights/global_capital_markets/financial_globalization http://www.mckinsey.com/insights/global_capital_markets/financial_globalization http://www.mckinsey.com/insights/business_technology/disruptive_technologies http://www.mckinsey.com/insights/business_technology/disruptive_technologies http://www.mckinsey.com/insights/business_technology/disruptive_technologies McKinsey Global Institute March 2016 Copyright © McKinsey & Company www.mckinsey.com/mgi @McKinsey_MGI McKinseyGlobalInstitute www.mckinsey.com/mgi Weaponized Interdependence: How Global Economic Networks Shape State Coercion Henry Farrell, Abraham L. Newman International Security, Volume 44, Number 1, Summer 2019, pp. 42-79 (Article) Published by The MIT Press For additional information about this article [ Access provided at 29 Apr 2021 23:39 GMT from Arizona State University ] https://muse.jhu.edu/article/730804 https://muse.jhu.edu/article/730804 In May 2018, Donald Trump announced that the United States was pulling out of the Joint Comprehensive Plan of Action agreement on Iran’s nuclear program and re- imposing sanctions. Most notably, many of these penalties apply not to U.S. ªrms, but to foreign ªrms that may have no presence in the United States. The sanctions are consequential in large part because of U.S. importance to the global ªnancial network.1 This unilateral action led to protest among the United States’ European allies: France’s ªnance minister, Bruno Le Maire, tartly noted that the United States was not the “economic policeman of the planet.”2 The reimposition of sanctions on Iran is just one recent example of how the United States is using global economic networks to achieve its strategic aims.3 While security scholars have long recognized the crucial importance of energy markets in shaping geostrategic outcomes,4 ªnancial and information markets Henry Farrell is Professor of Political Science and International Affairs at George Washington University. Abraham L. Newman is Professor at the Edmund A. Walsh School of Foreign Service and the Government Department at Georgetown University. The authors are grateful to Miles Evers, Llewellyn Hughes, Woojeong Jang, Erik Jones, Miles Kahler, Nikhil Kalyanpur, Matthias Matthijs, Kathleen McNamara, Daniel Nexon, Gideon Rose, Mark Schwartz, and William Winecoff, as well as the anonymous reviewers for comments and criticism. Charles Glaser provided especially detailed and helpful comments on an early draft. Previous versions of this article were presented at the 2018 annual convention of the Interna- tional Studies Association and at the Johns Hopkins University School of Advanced International Studies Research Seminar in Politics and Political Economy on April 17, 2018. The authors are also grateful to the participants and audiences at both events for feedback. 1. The legal principles through which exposure is determined are complex. For a useful introduc- tion, see Serena B. Wille, “Anti-Money-Laundering and OFAC Sanctions Issues,” CFA Institute Conference Proceedings Quarterly, Vol. 29, No. 3 (2011), pp. 59–64, doi.org/10.2469/cp.v29.n3.2. 2. Anne-Sylvaine Chassany, Michael Peel, and Tobias Buck, “EU to Seek Exemptions from New U.S. Sanctions on Iran,” Financial Times, May 9, 2018. 3. Henry Foy, “EN� President Steps Down in Move to Win U.S. Sanctions Waiver,” Financial Times, June 4, 2018. 4. Llewelyn Hughes and Austin Long, “Is There an Oil Weapon? Security Implications of Changes in the Structure of the International Oil Market,” International Security, Vol. 39, No. 3 (Winter 2014/ 15), pp. 152–189, doi.org/10.1162/ISEC_a_00188; Jeff D. Colgan, “Fueling the Fire: Pathways from Oil to War,” International Security, Vol. 38, No. 2 (Fall 2013), pp. 147–180, doi.org/10.1162/ ISEC_a_00135; Charles L. Glaser, “How Oil Inºuences U.S. National Security: Reframing Energy Security,” International Security, Vol. 38, No. 2 (Fall 2013), pp. 112–146, doi.org/10.1162/ISEC_a _00137; and Llewelyn Hughes and Phillip Y. Lipscy, “The Politics of Energy,” Annual Review of Po- litical Science, Vol. 16, No. 1 (May 2013), pp. 449–469, doi.org/10.1146/annurev-polisci-072211- 143240. International Security, Vol. 44, No. 1 (Summer 2019), pp. 42–79, https://doi.org/10.1162/ISEC_a_00351 © 2019 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology. 42 Weaponized Interdependence Weaponized Interdependence Henry Farrell and Abraham L. Newman How Global Economic Networks Shape State Coercion are rapidly coming to play similarly important roles. In Rosa Brooks’s evoca- tive description, globalization has created a world in which everything be- came war.5 Flows of ªnance, information, and physical goods across borders create both new risks for states and new tools to alternatively exploit or miti- gate those risks. The result, as Thomas Wright describes it, is a world where unprecedented levels of interdependence are combined with continued jock- eying for power, so that states that are unwilling to engage in direct conºict may still employ all measures short of war.6 Global economic networks have security consequences, because they in- crease interdependence between states that were previously relatively autono- mous. Yet, existing theory provides few guideposts as to how states may leverage network structures as a coercive tool and under what circumstances. It has focused instead on trade relations between dyadic pairs and the vulnera- bilities generated by those interactions.7 Similarly, work on economic sanctions has yet to fully grasp the consequences of economic networks and how they are being weaponized. Rather, that literature primarily looks to explain the success or failure of direct sanctions (i.e., sanctions that involve states denying outside access to their own markets individually or as an alliance).8 Power and vulnerability are characterized as the consequences of aggregate market size or bilateral interdependencies. In addition, accounts that examine more dif- Weaponized Interdependence 43 5. Rosa Brooks, How Everything Became War and the Military Became Everything: Tales from the Penta- gon (New York: Simon & Schuster, 2017). 6. Thomas J. Wright, All Measures Short of War: The Contest for the Twenty-First Century and the Fu- ture of American Power (New Haven, Conn.: Yale University Press, 2017). 7. Joanne Gowa, “Bipolarity, Multipolarity, and Free Trade,” American Political Science Review, Vol. 83, No. 4 (December 1989), pp. 1245–1256, doi.org/10.2307/1961667; Brian M. Pollins, “Does Trade Still Follow the Flag?” American Political Science Review, Vol. 83, No. 2 (June 1989), pp. 465– 480, doi.org/10.2307/1962400; John R. Oneal et al., “The Liberal Peace: Interdependence, Democ- racy, and International Conºict, 1950–85,” Journal of Peace Research, Vol. 33, No. 1 (February 1996), pp. 11–28, doi.org/10.1177/0022343396033001002; and Dale C. Copeland, Economic Interdependence and War (Princeton, N.J.: Princeton University Press, 2014). 8. Robert A. Pape, “Why Economic Sanctions Do Not Work,” International Security, Vol. 22, No. 2 (Fall 1997), pp. 90–136, doi.org/10.2307/2539368; Kimberly Ann Elliott, “The Sanctions Glass: Half Full or Completely Empty?” International Security, Vol. 23, No. 1 (Summer 1998), pp. 50–65, doi.org/10.2307/2539262; Daniel W. Drezner, The Sanctions Paradox: Economic Statecraft and Interna- tional Relations (New York: Cambridge University Press, 1999); David A. Baldwin, “The Sanctions Debate and the Logic of Choice,” International Security, Vol. 24, No. 3 (Winter 2000), pp. 80–107, doi.org/10.1162/016228899560248; Jonathan Kirshner, “Economic Sanctions: The State of the Art,” Security Studies, Vol. 11, No. 4 (Summer 2002), pp. 160–179, doi.org/10.1080/714005348; Fiona McGillivray and Allan C. Stam, “Political Institutions, Coercive Diplomacy, and the Duration of Economic Sanctions,” Journal of Conºict Resolution, Vol. 48, No. 2 (April 2004), pp. 154–172, doi.org/10.1177/0022002703262858; and Daniel W. Drezner, “Outside the Box: Explaining Sanc- tions in Pursuit of Foreign Economic Goals,” International Interactions, Vol. 26, No. 4 (2001), pp. 379–410, doi.org/10.1080/03050620108434972, which does consider secondary sanctions, as does the policy literature we discuss below. fuse or secondary sanctions have focused more on comparative effectiveness than on theory building.9 In this article, we develop a different understanding of state power, which highlights the structural aspects of interdependence. Speciªcally, we show how the topography of the economic networks of interdependence intersects with domestic institutions and norms to shape coercive authority. Our account places networks such as ªnancial communications, supply chains, and the in- ternet, which have been largely neglected by international relations scholars, at the heart of a compelling new understanding of globalization and power.10 Globalization has transformed the liberal order, by moving the action away from multilateral interstate negotiations and toward networks of private ac- tors.11 This transformation has had crucial consequences for where state power is located in international politics, and how it is exercised. We contrast our argument with standard liberal accounts of complex inter- dependence. The initial liberal account of interdependence paid some atten- International Security 44:1 44 9. See Peter D. Feaver and Eric B. Lorber, Coercive Diplomacy and the New Financial Levers: Evalu- ating the Intended and Unintended Consequences of Financial Sanctions (London: Legatum Institute, 2010); Orde F. Kittrie, “New Sanctions for a New Century: Treasury’s Innovative Use of Financial Sanctions,” University of Pennsylvania Journal of International Law, Vol. 30, No. 3 (Spring 2009), pp. 789–822; and Daniel W. Drezner, “Targeted Sanctions in a World of Global Finance,” Interna- tional Interactions, Vol. 41, No. 4 (2015), pp. 755–764, doi.org/10.1080/03050629.2015.1041297. Sec- ondary sanctions coexist with other tools to control international ªnancial ºows. For a useful recent overview, see Miles Kahler et al., Global Governance to Combat Illicit Financial Flows: Measure- ment, Evaluation, Innovation (Washington, D.C.: Council on Foreign Relations, 2018). 10. Of course, there is a burgeoning scholarship on cybersecurity, which is relevant to the internet. See Sarah E. Kreps and Jacquelyn Schneider, “Escalation Firebreaks in the Cyber, Conventional, and Nuclear Domains: Moving beyond Effects-Based Logics,” Cornell University and U.S. Naval War College, 2018; Joseph S. Nye Jr., “Deterrence and Dissuasion in Cyberspace,” International Security, Vol. 41, No. 3 (Winter 2016/17), pp. 44–71, doi.org/10.1162/ISEC_a_00266; Rebecca Slayton, “What Is the Cyber Offense-Defense Balance? Conceptions, Causes, and Assessment,” In- ternational Security, Vol. 41, No. 3 (Winter 2016/17), pp. 72–109, doi.org/10.1162/ISEC_a_00267; Henry Farrell and Charles L. Glaser, “The Role of Effects, Saliencies, and Norms in U.S. Cyberwar Doctrine,” Journal of Cybersecurity, Vol. 3, No. 1 (March 2017), pp. 7–17, doi.org/10.1093/cybsec/ tyw015; and Jon R. Lindsay, “The Impact of China on Cybersecurity: Fiction and Friction,” Interna- tional Security, Vol. 39, No. 3 (Winter 2014/15), pp. 7–47, doi.org/10.1162/ISEC_a_00189. This liter- ature, however, largely fails to address the network characteristics of the internet, focusing instead on variation in traditional metrics such as the offense-defense balance, the ability to deter or com- pel, and the treatment of the network characteristics of the internet either as a constant or a straightforward determinant of state-level vulnerability or strength (so that technologically ad- vanced states such as the United States will have a different set of strengths and vulnerabilities than states that rely less on technology). An earlier proto-literature on “netwar” examines how leaderless networks are becoming more important in world politics, but is primarily descriptive in nature. See John Arquilla and David Ronfeldt, The Advent of Netwar (Santa Monica, Calif.: RAND Corporation, 1996). There is a technical literature that discusses networks, but it tends not to dis- cuss the strategic aspects we focus on below. For an important exception, see Réka Albert, Hawoong Jeong, and Albert-László Barabási, “Error and Attack Tolerance of Complex Networks,” Nature, July 2000, pp. 378–382, doi.org/10.1038/35019019. 11. Kathryn Judge, “Intermediary Inºuence,” University of Chicago Law Review, Vol. 82, No. 2 (Spring 2015), pp. 573–642, https://chicagounbound.uchicago.edu/uclrev/vol82/iss2/1. tion to power, but emphasized bilateral relationships. Subsequent liberal accounts have tended either to avoid the question of power, focusing on mu- tual cooperative gains, to suggest that apparently lopsided global networks obscure more fundamental patterns of mutual dependence, or to posit a net- worked global order in which liberal states such as the United States can exer- cise “power with” (the power to work together constructively with allies) to achieve liberal objectives.12 Our alternative account makes a starkly different assumption, providing a structural explanation of interdependence in which network topography gen- erates enduring power imbalances among states. Here we draw on sociologi- cal and computational research on large-scale networks, which demonstrates the tendency of complex systems to produce asymmetric network structures, in which some nodes are “hubs,” and are far more connected than others.13 Asymmetric network structures create the potential for “weaponized inter- dependence,” in which some states are able to leverage interdependent rela- tions to coerce others. Speciªcally, states with political authority over the central nodes in the international networked structures through which money, goods, and information travel are uniquely positioned to impose costs on oth- ers. If they have appropriate domestic institutions, they can weaponize net- works to gather information or choke off economic and information ºows, discover and exploit vulnerabilities, compel policy change, and deter un- wanted actions. We identify and explain variation in two strategies through Weaponized Interdependence 45 12. See Robert O. Keohane and Joseph S. Nye Jr., Power and Interdependence, 4th ed. (New York: Longman, 2012); Kal Raustiala, “The Architecture of International Cooperation: Transgovern- mental Networks and the Future of International Law,” Virginia Journal of International Law, Vol. 43, No. 1 (Fall 2002), pp. 1–92; Anne-Marie Slaughter, “Global Government Networks, Global Information Agencies, and Disaggregated Democracy,” Michigan Journal of International Law, Vol. 24, No. 4 (Summer 2003), pp. 1044–1075, https://repository.law.umich.edu/mjil/vol24/ iss4/7; Anne-Marie Slaughter, A New World Order (Princeton, N.J.: Princeton University Press, 2004); and Anne-Marie Slaughter, The Chessboard and the Web: Strategies of Connection in a Networked World (New Haven, Conn.: Yale University Press, 2017). The classic critique of liberalism’s empha- sis on mutual gains from cooperation is Stephen D. Krasner, “Global Communications and National Power: Life on the Pareto Frontier,” World Politics, Vol. 43, No. 3 (April 1991), pp. 336–366, doi.org/10.2307/2010398. 13. Albert-László Barabási and Réka Albert, “Emergence of Scaling in Random Networks,” Sci- ence, October 1999, pp. 509–512, doi.org/10.1126/science.286.5439.509; M.E.J. Newman and Juyong Park, “Why Social Networks Are Different from Other Types of Networks,” Physical Re- view E, September 2003, pp. 1–8, doi.org/10.1103/PhysRevE.68.036122; Aaron Clauset, Cosma Rohilla Shalizi, and M.E.J. Newman, “Power-Law Distributions in Empirical Data,” SIAM Review, Vol. 51, No. 4 (December 2009), pp. 661–703, doi.org/10.1137/070710111; Emilie M. Hafner-Burton, Miles Kahler, and Alexander H. Montgomery, “Network Analysis for International Relations,” International Organization, Vol. 63, No. 3 (Summer 2009), pp. 559–592, doi.org/10.1017/ S0020818309090195; and Stacie E. Goddard, “Embedded Revisionism: Networks, Institutions, and Challenges to World Order,” International Organization, Vol. 72, No. 4 (Fall 2018), pp. 763–797, doi.org/10.1017/S0020818318000206. which states can gain powerful advantages from weaponizing interdepen- dence; they respectively rely on the panopticon and chokepoint effects of net- works. In the former, advantaged states use their network position to extract informational advantages vis-à-vis adversaries, whereas in the latter, they can cut adversaries off from network ºows. To test the plausibility of our argument, we present detailed analytic nar- ratives of two substantive areas: ªnancial messaging and internet communica- tions.14 We selected these areas as they are signiªcant to a range of critical security issues including rogue-state nonproliferation, counterterrorism, and great power competition. Moreover, global ªnance and the internet are often depicted as being at the vanguard of decentralized economic networks. As such, they offer an important test of our argument and a contrast to the more common liberal perspective on global market interactions. At the same time, ªnancial messaging and internet communications see im- portant variation in the level and kind of control that they offer to inºuential states. In the former, the United States, in combination with its allies, has sufªcient jurisdictional grasp and appropriate domestic institutions to oblige hub actors to provide it with information and to cut off other actors and states. In internet communications, the United States solely has appropriate jurisdic- tional grasp and appropriate institutions to oblige hub actors to provide it with information, but does not have domestic institutions that would allow it to de- mand that other states be cut out of the network. This would lead us to expect that in the case of ªnancial messaging, the United States and its allies will be able to exercise both the panopticon and chokepoint effects—so long as they agree. In contrast, in internet communications, the United States will be able to exercise the panopticon effect even without the consent of its allies, but it will not be able to exercise the chokepoint effect. This variation allows us to dem- onstrate the limits of these network strategies and also show that they are not simply coterminous with United States market size or military power. Empirically, the cases draw on extensive readings of the primary and second- ary literature as well as interviews with key policymakers. Our argument has signiªcant implications for scholars interested in thinking about the future of conºict in a world of global economic and information net- works. For those steeped in the liberal tradition, we demonstrate that institu- tions designed to generate market efªciencies and reduce transaction costs can International Security 44:1 46 14. Anecdotal evidence suggests similar processes are at work in a number of other areas, includ- ing dollar clearing and global supply chains. See, for example, Cheng Ting-Fang and Lauly Li, “‘Huawei Freeze’ Chills Global Supply Chain,” Nikki Asian Review, December 8, 2018, https:// asia.nikkei.com/Economy/Trade-war/Huawei-freeze-chills-global-supply-chain. be deployed for coercive ends. Focal points of cooperation have become sites of control. For those researchers interested in conºict studies and power, we show the critical role that economic relations play in coercion. Rather than re- hashing more conventional debates on trade and conºict, we underscore how relatively new forms of economic interaction—ªnancial and information ºows—shape strategic opportunities, stressing in particular how the topogra- phy of global networks structures coercion. Here, we use basic insights from network theory to rethink structural power, linking the literatures on economic and security relations to show how coercive economic power can stem from structural characteristics of the global economy. Finally, the article begins to map the deep empirical connections between economic networks—for ex- ample, ªnancial messaging, dollar clearing, global supply chains, and inter- net communication—and a series of pressing real-world issues—including counterterrorism, cybersecurity, rogue states, and great power competition. We begin by explaining how global networks play a structural role in the world economy. Next, we describe how these networks, together with domes- tic institutions and norms, shape the strategic options available to states, focus- ing on what we describe as the panopticon and chokepoint effects. We provide detailed parallel histories of how networks in ªnancial communication and in- ternet communication developed and were weaponized by the United States. We conclude by considering the policy implications of clashes between coun- tries such as the United States that have weaponized interdependence and other states looking to counter these inºuences. Statecraft and Structure: The Role of Global Networks As globalization has advanced, it has fostered new networks of exchange— whether economic, informational, or physical—that have remade domestic economies, densely and intimately interconnec