These are, of course, all indicators of informational flows. What about cross-border movements of people? To start with long-run movements, despite all the hubbub about immigration, first-generation immigrants account for only 3 percent of the world's population. For medium-run movements, we can look at university students, whom we might expect to be relatively mobile. Actually, students studying overseas account for just 2 percent of all university students! If one adds up all kinds of cross-border movements of people, it is estimated that about 90 percent of the world's people will never leave the country in which they were born.13 World 2.0, indeed!
Okay, okay, you say, but where it really counts—the flows of goods and money—the world's economies must be much more integrated. After all, the broad debate about globalization hasn't been about coverage of foreign news or concentrations of foreign students; it's been about companies shutting down domestic factories and call centers and investing overseas or outsourcing those activities. Especially given trade's accelerating growth over the last two centuries, as described in chapter 1, product market integration must be close to complete. Or so a motley crew ranging from proglobalizers such as Thomas Friedman to rabid antiglobalizers have intuited.
Not quite. Trade intensity, measured by products and services exported from one country to another as a percentage of GDP, hit an all-time peak of 29 percent in 2008, before dropping to 23 percent in 2009 (a decline that is also a sign of its fragility). These are higher internationalization numbers than the ones cited earlier for informational and people flows, but they also inflate and distort the reality of trade. I should elaborate.
Consider Apple's iPod. Although it is labeled as made in China, China is just the final assembly platform for four hundred–plus components from East Asia and elsewhere, and adds only a few dollars of value, 1 to 2 percent of the retail selling price of $299.14 Most of the value added—an estimated $163—goes to American companies and workers, with Apple alone pocketing almost half of that. Yet every such iPod sold in the United States is recorded as contributing about $150 to the U.S. trade deficit with China.15 And since virtually all the components are shipped at least twice if not several times across national borders, the total trade officially recorded around that sale is a multiple of the $150 number!
Given such distortions and exaggerations, deflating official trade statistics to reflect value added rather than revenues is a key operational priority—not just according to me but according to the director-general of the World Trade Organization (WTO), Pascal Lamy. As Lamy recently explained to me, without such a correction, we are left with an illusory sense of surging trade when much of what is happening is that value chains have been getting sliced up ever more finely across national borders. And that exaggeration can, among other things, encourage protectionism—a concern of his that I share and will shortly return to, in a way that goes beyond the trade numbers.
The nonofficial academic estimates that we do have suggest that foreign content accounts for about 50 percent of the value of China's exports and 25 to 30 percent for world exports as a whole.16 Sticking with these estimates—although Lamy thinks that they are too low, perhaps by 5 to 10 percent—and using them to deflate the numbers for nominal exports suggests that global exports account for only 20 percent of all the value produced in the world (GDP).17 This is still a large number, but it's far below the roughly 90 percent export-to-GDP ratio we'd expect to see—without any double counting—if borders and distance didn't matter at all.18
If you remain skeptical, consider that trade economists—people who spend their professional lives focused on trying to understand trade flows—focus not on celebrating how much trade there is, but rather on trying to explain why trade levels are much lower than simple models would lead us to expect. We'll dig deeper into this mystery of the missing trade, as trade economists call it, in the next chapter.
When we move away from trade and consider the flow of investment around the world, the picture is strikingly similar. Economists define foreign direct investment (FDI) as a company from one country making a physical investment in building or buying operations in another country. FDI flowing across borders accounted for only 9 percent of all fixed investment (what economists call “gross fixed capital formation”) in 2009. Of course, FDI flows also fluctuate a great deal, and peaked recently in 2007, at 16 percent of fixed investment. But they have averaged about 10 percent over the last few years, suggesting that about 90 percent of all fixed investment in the world is still domestic.19
Other cross-border capital flows, based on cruder data, generally run somewhat higher but resonate with the same basic story. Only 15 to 20 percent of venture capital money is deployed outside the investing fund's home country.20 Only about 20 percent of the stock market (equity) is owned by foreign investors.21 And cross-border ownership of bank deposits and government debt remains closer to 25 percent and 35 percent, respectively.22 Since this is precrisis data, capital was far from completely mobile internationally even at the height of global financial fever!
One notable and distressing exception to financial flows' comparatively greater internationalization involves noncommercial flows. The cross-border component of private charitable giving has been estimated at less than 10 and even 5 percent of the total, although billionaire philanthropists with global outlooks appear to be trying to change that.23 Charity, even more than commerce or finance, begins at home.
Taking Stock of Semiglobalization
Figure 2-1 compiles the more systematic statistics among the ones cited above.24 It is worth reminding ourselves at this point that these are global averages that mask huge amounts of cross-country variation, measure by measure. In other words, globalization is very uneven. Cross-country differences are the focus of the next chapter; for now, note that there is some useful information to be gleaned from the global averages.
The five purely financial measures of cross-border integration just discussed show average internationalization levels of 21 percent, and the other nine measures average 10 percent. Of the latter, the people-related measures both fall below 10 percent; the product market–related measures—FDI and adjusted exports—vary between 10 and 20 percent; and the informational measures are more dispersed, varying from zero to 20 percent. Overall, capital is more footloose across national borders than products, which in turn are more footloose than people. This ranking can be related to differences in the spreads of capital costs, product prices, and wages around the world. Although capital costs typically diverge by only a few percentage points or less in different countries, the prices for even a product as standardized as a Big Mac burger can easily vary 20 percent or more from one country to the next, and wages might be twenty times higher in an advanced Western country than in a country like India.25
Figure 2-1: Internationalization levels
But the biggest headline from this analysis relates not to differences across product, capital, people, and information flows, but to a commonality: actual levels of internationalization across all these different types of markets fall very far short of the levels implied by World 2.0.
Presentation of these data to the several dozen audiences mentioned earlier has also elicited two particularly common critiques that I need to address. First, people dismiss the data presented above as too narrow and, specifically, overly focused on economic flows. (This was Tom Friedman's critique of a subset of this data that I presented in a 2007 article in Foreign Policy titled “Why the World Isn't Flat”; I responded by pointing out that at least I had some data.)26 Actually, rather than just picking a few random measures, I went through A. T. Kearney/Foreign Policy's 2007 Globalization Survey—the most recent available—and tried to figure out which of the broad dimensions of globalization it covered could be measured the way I wanted to (i.e., in terms of the cross-border component as a percentage of the total). The Globalization Survey's measures of political engagement were the ones that proved most resistant but, given their other attributes, their excl
usion is not too much of a concern.27 So I have deliberately tried to be broad—subject to the constraint of measurability—rather than narrow.
A second critique, worth considering at somewhat greater length, focuses on the fact that the data presented above provide a sense of levels of globalization but not of changes over time. So despite levels of cross-border integration that currently correspond to World 3.0, very rapid increases in those levels could conceivably make World 2.0 a better guide to the future. (This may have been the point that General Powell was trying to make.) Or very rapid decreases could achieve the same for World 1.0.
Long-run trends provide little support for the assertion of “World 2.0 tomorrow if not today.” While measures of globalization discussed in this section have generally increased over the last few decades, they haven't all been setting new records. The percentage of the world's population composed of immigrants, for example, is the same now as it was in 1910. Also, some precrisis measures of cross-border capital flows/stocks were actually comparable to earlier peaks more than a hundred years ago—and thanks to the crisis, are now lower.
Second, the measures that have been setting new records, such as trade intensity, have a long way to go before approaching the levels associated with complete integration and World 2.0. Even if one were to simply extrapolate the faster growth rates experienced by trade than by GDP in recent decades, trade-to-GDP would still take decades to reach the point where one could say that national borders no longer affected international trade.
Third, the increases that have been achieved often exhibit clear geographic patterns that are inconsistent with World 2.0. Trade has become increasingly regionalized in recent decades, to the point that international trade within regions exceeds trade across regions. Based on rougher data, FDI also seems more regionalized than not. Even the Internet, it is generally agreed, exhibits increasing localization and regionalization for reasons ranging from increasing peer-to-peer traffic to the development of regional alternatives to the United States' earlier role of switching hub for the world. Strong distance dependence of this sort is inconsistent with World 2.0 but is an integral component of World 3.0, as the next chapter will stress.
Fourth, the precipitous declines in various cross-border flows in the aftermath of the crisis remind us of the limits of linear extrapolation. We shouldn't assume that cross-border integration will, along most dimensions, continue to progress. Evidently globalization can go into reverse—as it did between World War I and World War II.
In the wake of the financial crisis, this last point has led some to proclaim that instead of the imminent attainment of World 2.0, we're poised to revert all the way back to World 1.0. Once again, this claim can't survive contact with data. Consider in some detail a cross-border flow that has shown a particularly big percentage decline in the wake of the financial crisis, foreign direct investment. The United Nations Conference on Trade and Development, UNCTAD, estimates in its 2010 World Investment Report that FDI fell from nearly $2 trillion in 2007 to $1.7 trillion in 2008 and $1 trillion in 2009—or by nearly 50 percent. But before we get all apocalyptic, it is worth remembering that this is not an unprecedented decline. The 2000–2002 period supplies a more extreme example from within the same decade: while global GDP actually grew over that period, the stock market crashed in the wake of the Internet bubble—and FDI fell by more than 50 percent! Why? Because surges in FDI activity are typically driven by mergers and acquisitions activity, which tends to peak with stock market prices.
All of this suggests that the decline in FDI between 2007 and 2009 greatly overstates the true changes—so far—in the fundamentals facing the global economy. UNCTAD's own estimates suggest that FDI flows will recover—albeit not right away, to $1.2 trillion in 2010, $1.4 trillion in 2011, and $1.8 trillion in 2012. In other words, by 2012, FDI will surpass the 11 percent of gross fixed capital formation that it averaged between 2000 and 2009. But even if one discounts such a recovery by making the extreme assumption that FDI levels will maintain their (low) 2009 level of 9 percent of gross fixed capital formation, that still exceeds the averages of 6 percent for the 1990s, 3 percent for the 1980s, and 2 percent for the 1970s. So the argument that the aftermath of the crisis has somehow returned us to the world of a few decades ago—when World 1.0 did work better as a worldview—is greatly overstated. The caveat, elaborated in chapter 4 and later in this book, is that we may yet revert to World 1.0 if we succumb to protectionism.
Globaloney Analyzed
How surprised are you by the data thus far? If you overestimated actual levels of integration, you are not alone: the polls that I've conducted reveal overestimation as a widespread tendency rather than just a predictable bias of the prophets of World 2.0. Thus, in a survey of several hundred corporate managers conducted for me by Harvard Business Review, the respondents guessed an average level of internationalization of 30 percent for a subset of the measures in figure 2-1, versus an actual average of 10 percent! Interestingly, experience offered no substitute for actually looking at the data: very senior or internationally experienced executives were as likely to overestimate as much as all the rest. Such overestimation characterizes every group that I have surveyed so far, and the crisis seems to have only dented this tendency instead of dispelling it. The breadth and depth of this bias will cause me to refer to it frequently so, for the sake of compactness, I adopt Clare Booth Luce's useful tag of “globaloney.”
Why are intelligent, informed, interested people so prone to globaloney? It is important to answer this question because globaloney is more than just a harmless way of calling more attention to the international domain. Globaloney, even if it stops short of the exaggerations of World 2.0, can be hazardous to global welfare by creating complacency among proglobalizers and provoking paranoia among antiglobalizers (cf. Lamy's point, cited earlier, about the effects of double counting in trade statistics).
The obvious reason for globaloney is that much of the debate about globalization takes place in a data-free zone. Compiling the numbers I've presented in this chapter wasn't easy. Broadcast media don't usually make the effort, and scholarly writings tend to be years if not decades out of date—as well as incomprehensible to nonspecialists. Without outside help, we have difficulty imagining, much less actually knowing, how others live, given how localized our own daily lives still are. It's simpler to assume that things are pretty much the same everywhere—which is consistent with a sense that total globalization is a fait accompli.
Having said that, the notion that data availability is the problem is itself problematic. Something other than data must account for the success of The World Is Flat, since its 450-plus pages contain not a single table, chart, or footnote to back up its pronouncements. I still find the comfort of Friedman's many fans with this data-free approach the most flabbergasting aspect of the flattening.
Psychology reminds us, of course, that beliefs aren't based just on data. The specter of apocalyptic occurrences (such as World 2.0) can mesmerize people; as the French writer of fables, Jean de la Fontaine, famously observed, “Everyone believes very easily whatever they fear or desire.” This suggests some additional reasons why globaloney may be so popular: fear, as in the cases of those who worry about world domination by multinationals (including many Marxists), or who are simply concerned about losing their jobs to foreign competition; and desire, as in the cases of executives who like to see themselves as superheroes capable of solving business problems anywhere and, more broadly, all those who hope to gain personally from additional globalization.
Whether one actually believes in globaloney or not, professing to do so also carries social cachet. In many circles, you're “not with it” if you don't subscribe to Friedman's “world is flat” notion of globalization. I can testify to some personal experience in this regard: as a flat-world skeptic, I've often found myself treated as a relic, especially by reporters. I still remember the TV interview that began with the question, commingling disbelief and pity, “Pr
ofessor Ghemawat, why do you still believe the world is round?” I am afraid I burst out laughing in response. It was in the aftermath that I decided to write my “Why the World Isn't Flat” article.
Technotrances
All these factors are important, but perhaps the most powerful reason most people buy into apocalyptic notions of globalization has to do with our almost religious worship of technology. I've already mentioned how empowering the Internet seems to many people. Since productivity growth and incomes really took off in the nineteenth century (see figure 1-1), people in the throes of industrial progress have been dazzled again and again by technology, enthralled with it, even locked into what I like to call a technotrance. (Younger readers will get the analogy with techno music and its effects on brain activity.)
As far back as the 1850s, the Scottish missionary and explorer David Livingstone remarked that “the extension and use of railroads, steamships, telegraphs, break down nationalities and bring peoples geographically remote into close connection commercially and politically. They make the world one, and capital, like water, tends to a common level.”28 U.S. philosopher John Dewey, writing in 1927, argued that steam, electricity, and the telephone were not only creating a new world by compressing space but might also challenge local forms of self-government.29 In 1929, U.S. carmaker Henry Ford gushed that “the airplane and wireless know no boundary … they are binding the world together in a way no system can.”30 In 1933, English novelist and historian H. G. Wells, also an air enthusiast, imagined a world state based on air power that used the Basic English of aviators as its language and the air dollar as its currency—and had Basra, Iraq, as its capital.31
World 3.0 Page 4