THE CAUSES OF GLOBALIZATION

Geoffrey Garrett Yale University April 2000

Thanks to Stephen Brooks, Michael Dooley, Jeffry Frieden and Ronald Rogowski for helpful discussions on various aspects of this paper. I would also like to thank Alexandra Guisinger, Nathan Jensen, Jason Sorens, Andrew Youn and especially Nancy Brune for invaluable research assistance.

There is little disagreement these days that globalization is changing the world, rapidly, radically, and in ways that may be profoundly disequilibrating. But beyond this already trite clich, almost everything else concerning the phenomenon is subject to intense debate in the context of an explosion of interest in and research on the subject.1 This article explores what we know about the causes of globalization. In a followup arrticle, I will address globalization's consequences for domestic societies (in terms of inequality and economic insecurity), for national autonomy (with respect to regulation, spending and taxation, exchange rate regimes, etc.), and for international governance (IMF, WTO, etc.). In both articles, I define globalization somewhat narrowly as the international integration of markets in goods, services and capital. Other facets of the phenomenon such as increased labor mobility and cultural homogenization are surely important, but I leave their analysis to others.2 I examine four contending perspectives on the "big picture" what explains the rapid pace of international market integration in recent decades? The first perspectives claims that what we are witnessing today is, in fact, "nothing new" because current levels of market integration are only now returning to those in the last great era of economic internationalization at the turn of the 20th century. This view has been accepted as a statement of fact in numerous influential studies [Katzenstein, Keohane and Krasner 1998: 29, Krasner 1999: 220-223, Rodrik 1997, Sachs and Warner 1995]. I argue, however, that notwithstanding the aggregate similarities between the two periods, core features of the contemporary world economy are without historical precedent. Largescale portfolio lending to banks in developing countries for purposes other than raw

material extraction, two-way manufacturing trade between the north and south, and complex multinational production regimes were simply unheard of a century ago. The remaining perspectives on global trends debate the causes of this unprecedented wave of international market integration. The root of the analytic problem lies in the commingling of three secular trends technological innovations lowering the costs of moving goods and more notably information around the world, growing international economic activity, and the liberalization of foreign economic policies. What are the causal relationships among these three trends? The second perspective, "technological determinism", contends that the shrinkage of time and space has been so dramatic and so pervasive that there is essentially nothing that can be done to stop it. According to this view, technological changes have propelled international economic activity, and governments have been largely irrelevant. Thus, policy liberalization should be understood as governments' acknowledging the futility of trying to resist globalization, rather than acting as a prime mover behind market integration. Management gurus such as Kenichi Ohmae [1995] have propounded this view but political scientists such as Richard Rosecrance [1999] and Susan Strange [1998] use it as the starting point of their analyses. The case for a technologically determined view of globalization is far stronger with respect to international finance than to multinational production or trade. In the era of 24 hour global trading in a seemingly limitless array of financial instruments, governments can only hope marginally to influence control cross-border liquid capital movements. In contrast, though the Internet creates novel problems, it remains far easier for governments to regulate cross-border movements of physical goods and the buying and selling of fixed

assets. Hence, policy decisions to liberalize trade and foreign direct investment are likely to have been more consequential for international integration in these markets. The third big picture perspective on globalization takes a more moderate view of the effects of technological change. Most mainstream economists (informing the Washington consensus and best sellers [Yergin and Stanislaw 1999]) believe that potential efficiency gains from international integration have increased substantially as a result of technological progress in recent decades. From this perspective, governments can still insulate their countries from external market forces if they so choose. But the "increased opportunity costs of closure" have become sufficiently large as to tip the balance in favor of the liberalization of foreign economic policy in country after country. It is hard to argue that increasing opportunity costs of closure provide a persuasive account of the globalization of finance. The hypothetical efficiency gains of openness seem in practice to be at least offset by the costs associated with the uncertainty and volatility of international financial markets. At the other end of the spectrum, increasing costs of closure probably has been the major motivation for liberalization in the area of foreign direct investment. FDI is an important driver of growth. It provides a transmission mechanism for the diffusion of technological innovations and less tangible benefits such as managerial skills. The trade case is less clear-cut. On the one hand, there are clearly important one-time gains from trade liberalization (in terms of lowering prices for example). But modern economic theory is ambiguous as to whether freer trade is beneficial for economic growth, and the empirical evidence is also inconclusive. The final big picture perspective on globalization also accepts the critical role of government policy, but argues that the phenomenon is essentially a political construct

that does not improve the economic condition of society as a whole. For example, Dani Rodrik has raised numerous eyebrows among his economist colleagues by claiming that there is no evidence that either freer trade [Rodriguez and Rodrik 1999] or capital mobility [Rodrik 1998] is good for economic growth. This is grist for the mill of political scientists such as Eric Helleiner [1994] who propose power and ideology explanations of globalization. On this ideological change view, the roots of contemporary globalization lay in the neoliberal Reagan/Thatcher revolutions. They were spread throughout the developed world by the European Union and the Bank of International Settlements, and extended to developing counties by the IMF and the World Bank. It is easy, however, to endogenize these ideological changes in terms of technological determinism in international finance and increased opportunity costs of closure with respect to multinational production. In the former case, there might be domestic political incentives for governments to maintain restrictions on cross-border capital movements that are futile in an economic sense. Such policies send negative signals to the financial markets, however, and many governments may be unwilling to take this risk. In terms of foreign direct investment, the fact that multinational firms have become critical drivers of technological innovation, learning and economic growth affords them a very "privileged position" [Lindblom 1977] in domestic policy debates. Trade liberalization, in contrast, has not been technologically determined, and the opportunity costs of closure continue to be debated. Changing preferences and coalitional politics may therefore have played a great role here than with respect to international finance or multinational production. One possible explanation is that exporters have become much more interested in opening their home markets, mitigating the traditional

political bias in favor of protection, both because of fears of retaliation against them abroad and because many exporters import large portions of the goods they import large quantities of inputs in making finished products. Students of both international relations and comparative politics may well object at this point that my analysis gives short shrift to their central concerns and insights. International political economists have devoted enormous attention to cooperative and institutionalized efforts to reduce barriers to international economic exchange among countries. The bread and butter of comparative political economy, on the other hand, concerns explaining cross-national differences in economic policies and outcomes. My general response to these objections is that political scientists tend not to explore in sufficient detail the economics of globalization before they move on to analyzing its politics. Moreover, they have an inherent bias towards assuming both that government policies have real effects and that they are chosen for political reasons. In the case of the trend towards international market integration, it is important to problematize both assumptions. Technological determinists believe government policy is essentially irrelevant to globalization; the increased opportunity costs of closure approach suggests that governments have liberalized their economies simply because it is the efficient thing to do. Before asserting that globalization is a political phenomenon, we should assess the merits of these parsimonious explanations derived from economic analysis. With respect to international political economy, I do not wish to dispute either that free rider and coordination problems may hinder international market integration or that international institutions may mitigate these problems. But I am skeptical that this Alexrod-Keohane paradigm gives us much leverage over the big picture of the

contemporary trend to globalization. It would be hard to make the case with respect the liberalization of international finance and the multinationalization of production simply because policy liberalization in these areas has not required international cooperation or international institutions (i.e. the evidence suggests that they are not international prisoners dilemmas or even coordination problems). The prima facie case for the importance of international institutions is stronger with respect to trade integration. The WTO, NAFTA and the EU all contain mechanisms for generating common standards and policing free riding. In order to argue that these institutions caused trade integration, however, one would have to contend (implausibly, in my opinion) that they were truly innovative that is, representing radically new technologies for dealing with the problems of cooperation that were heretofore unavailable. It seems more reasonable to contend that preference convergence among participating governments was a precondition for the effectiveness of these institutional solutions [Goldstein 1997, Moravcsik 1998]. Thus, we should focus on explaining why this convergence in preferences occurred. The comparative politics objection to my approach is very different. Notwithstanding the secular trend of ever-greater market integration, there clearly still are "ins" and "outs" in the putative "global economy". For globalization pundits, these differences may be merely ephemeral bumps that will soon be smoothed over on the road to a truly seamless global marketplace. Comparativists are likely to demur, arguing that cross-national variations in international market integration are sticky and well worth exploring in their own right.

This move to assaying cross-national differences in market integration is important, but in my judgment it is a second order move that should follow analysis of the broader over time trend to more integration. A good portion of the cross-national variation in international integration is certainly explained by essentially unalterable features of countries such as their size and geographic location. There are also well developed theoretical approaches to the problem that emphasize the impact of a country's economic structure on societal preferences and coalitions [Frieden and Rogowski 1996] and the role of political institutions ranging from trade unions to constitutional systems [Garrett and Lange 1995]. I offer a brief analysis of the these perspectives with respect to three prominent classes of variables levels of development, the extent of democracy, and the balance of power between the left and right. The strongest result is that countries at higher levels of development are more likely to open their borders to the international economy, which can be easily explained from a Frieden-Rogowski perspective. Of course, if growth economists are right that differences in levels of development must diminish over time (conditional convergence), this implies that cross-national variations in market integration will diminish over time. The debate would then move on to how long this might take. The remainder of this article explores in more detail the causes both of the secular trend to more globalized markets in recent decades and of persistent cross-national differences in participation in the global economy. The first section lays the foundation for my analysis by describing the landscape with respect both to international economic movements of trade and capital and to government policies concerning these flows. The

second section discusses the case for the proposition that contemporary globalization is nothing new. The merits of a technological determinist perspective on market integration are assessed in the third section. The fourth section addresses the issue of whether the opportunity costs of closure have risen in recent years. The fifth section explores the causes and consequences of ideological shifts in favor of liberalization. The sixth section then changes gears to focus on the reasons for enduring cross-national differences in market integration. The final section briefly summarizes what we know about the causes of globalization and sketches the implications of this article for analyses of the consequences of globalization.

THE PARAMETERS OF CONTEMPORARY GLOBALIZATION

No matter how many different numbers are presented or how frequently one hears them, the growth of international economic activity in the past thirty years remains staggering. Figure 1 plots the growth of global flows in trade, foreign direct investment and international portfolio investment (equities and bonds). Although the scales for trade and capital are very different, the trend lines are similar and familiar. International economic activity grew at increasingly rapid rates over the period, and the rates of growth were faster in more liquid markets (foreign exchange > portfolio > FDI > trade).3 In 1970, exports plus imports constituted roughly one quarter of worldwide GDP. By 1997, the figure had almost doubled to over 45%. Global annual flows of international portfolio investments (in bonds and equities) and FDI both constituted around 0.5% of world GDP in 1970. In 1997, the figures were approximately 5% for portfolio flows and 2.5% for FDI flows. In 1998, the global stock (i.e. accumulated flows) of FDI is estimated at $3.4 trillion roughly 10% of global output [Mallampauly and Sauvant 1999: 34-5]. Figure 1 about here

Figures 2 and 3 shows a strong correlation between the growth of international economic flows and the liberalization of foreign economic policies around the world. The correlation between global trade flows and (un-weighted) average taxes on trade (revenues from tariffs, duties, etc. as a percentage of total trade) between 1973 and 1995 was -0.89. The reduction in tariff-type barriers was to some measure offset by increasing use of non-tariff barriers in the OECD at least [Garrett 1998a: 811]. Moreover, although trade taxes more than halved over the period, they still averaged 8% of total trade revenues in 1995. Nonetheless, the global trend line is surely indicative of the fact that global trade flows and trade liberalization around the world have moved in lock step in recent decades. Figures 2 & 3 about here Figure 3 reveals a similar pattern with respect to international capital flows (combined portfolio and FDI) and the portion of countries in the world with open capital accounts (i.e. no significant restrictions on cross border capital movements according to the IMF).4 There is, however, one interesting divergence in these trends evident in the figure. International capital flows took off in mid 1980s (fueled largely by mushrooming portfolio flows), with a brief blip down during the international recession at the end of the decade. But the trend to open capital accounts postdates the take off in capital flows by about five years it was only in the 1990s that countries in large numbers opened their capital accounts.5 This suggests that flows preceded policy change consistent with the technological determinism thesis.6 As Dani Rodrik [2000] and Robert Wade [1996] have emphasized, one should not conclude from the steep growth curves on international economic flows and policy

liberalization that a truly seamless worldwide market is emerging. Table 1 summarizes flows and policy data for all the countries for which data are available in the 1990s (see Appendix 1 for the national level data).7 The first thing to note about this table is that the standard deviations for the different measures of globalization on the whole world sample (the bottom panel) were typically larger than the means on these variables implying considerable cross-national variation in market integration. The coefficient of variation (i.e. standard deviation/mean) was in fact only substantially less than one for trade flows. The variations in trade taxes may seem surprising given the spate of regionally and multilaterally-coordinated efforts at trade liberalization in recent decades. But while customs unions like the EU impose common external trade barriers on non-members, the GATT-WTO regime continues to allow for more flexibility. For example, data collected by Michael Finger and his World Bank colleagues [Finger et al. 1996: 67] show that the standard deviation of national average applied MFN tariff rates for a sample of 53 countries after the Uruguay Round was 9.2% (with a mean of 10.6%).8 Table 1 about here Table 1 also examines market integration in countries at different levels of development (and in the case of high income countries, distinguishing the stable industrial democracies of the OECD from other well developed nations). Comparing the means for the OECD countries with those for the lowest income nations (1997 GNP per capita < $786, comprising almost all of Africa, as well as the worlds two most populous countries, China and India) provides simple and stark evidence that there are in and outs in the purportedly global economy. Mean trade flows in the two groups were comparable (though the composition of these flows was clearly very different, with the

10

poor category relying disproportionately on the export of natural resources). This probably reflects the fact that, as standard gravity models show, factors such as country size and proximity to neighbors (which have nothing to do with level of development) have marked bearing on trade volumes. The high and low-income groups differed dramatically, however, on every other dimension of market integration. FDI flows were more than twice as large in the OECD as in the low income group; international portfolio investment was almost 25 times as large; trade taxes were less than 1/25 as large a portion of trade volumes; and capital accounts were more than ten times as likely to be open. Even within the OECD category, however, considerable differences in market integration remain. At one end of the spectrum, Belgium and the Netherlands are the OECDs most globalized economies. There are also numerous instances of relative non-integration. The US and Japan are very small traders (at least relative to the massive sizes of their economies), and FDI flows are scant in Japan. Even after a decade of radical market opening in the 1980s, Australia, Canada and New Zealand remain considerably more protectionist than the OECD norm (based on trade taxes on manufactures); Greece and Spain only liberalized their capital accounts at the end of the 1990s. But soothsayers would probably want to highlight instances of high and growing market integration among the poorest countries as harbingers of the world of tomorrow. China, for example was a major recipient of FDI inflows in the 1990s. Moreover, 1990s Indonesia resembled OECD norms on most of the basic indicators of globalization. But it would simply be inaccurate to portray these as more than isolated though clearly important exceptions to the rule that the worlds poorest countries remain largely

11

disconnected from the international economy. For example, while popular commentary might lead one to believe that software engineers working for Microsoft and Sun Microsystems and telecommuting from Bangalore and Hyderabad to Seattle and Silicon Valley are the norm in the Indian economy, on most basic indicators the country remains an essentially closed economy. At the other end of the spectrum, Table 1 also highlights the distinctiveness of the small wealthy non-OECD countries that are typically conduits for trade and international finance (Hong Kong and Singapore), small oil exporters (Kuwait and the United Arab Emirates) or tax havens (the Bahamas and the Cayman Islands). Very high levels of trade and capital flows, higher indeed than even the most integrated OECD nations, characterize these countries.9 Ohmae and Rosecrance believe that these region states or virtual states are the wave of the future. But it is hard to see how Brazil or China could ever become Singapore or the Cayman Islands. Table 2 asks a different question about developments at the national level in the 1990s: did different facets of market integration go together? There is some relatively weak evidence in the affirmative. As most modern economists believe, it does appear that trade and FDI are complements, rather than substitutes (the correlation between the two was a moderate 0.40). The correlation between FDI and international portfolio investment was weaker but still positive (0.27). Countries that imposed fewer trade taxes also were somewhat more likely to have open capital accounts (the correlation was 0.33). Table 2 about here

12

But Table 2 also suggests that the policies governments pursued with respect to openness or closure to trade and international capital were essentially uncorrelated with international economic flows. One possible explanation for these weak flows-policies cross-national correlations (as opposed to the strong over time ones in Figures 2 and 3) is that policies only affect flows at the margins. For example, standard gravity models of trade demonstrate that smaller and wealthier countries tend to be bigger traders. In order to control for these effects, I estimated a simple regression equation that included both variables as well as trade taxes as predictors of trade volumes: TRADE = 0.20TRTAX + 32.75lnGDPPC*** 15.33lnGDP*** + 158.09 (0.31) (5.74) (2.26) OLS regression with robust standard errors, R-squared = 0.40, 108 observations, *** statistically significant at the .01 level. GDPPC is GDP per capita and GDP is national GDP, both expressed as 1990-1997 averages in constant dollars. Surprisingly, the equation lends no more support to the view that countries that impose higher trade taxes tend to reduce trade flows. This is a strange finding because trade taxes must deter trade at the margins. It may well be the case that better econometric specifications (e.g. the use of panel data and more control variables) would delineate this effect [Guisinger 2000]. I also ran a similar regression for the partial correlation between capital account openness and capital flows which is more consistent with the proposition that capital account openness promotes international capital flows:10 CAPFLOWS = 2.65lnGDPPC*** 0.22lnGDP + 3.11OPENCA* 13.06 (0.49) (0.20) (1.61) OLS regression with robust standard errors, R-squared = 0.35, 128 observations, * statistically significant at the .10 level.

13

Of course, at this point I should reiterate that the simple analyses presented in this section are not intended to be definitive. They do serve the useful purpose, however, of highlighting two things. First, there is something behind all the globalization hoopla. The global trend towards international market integration and policy liberalization has been rapid in recent decades. Second, significant cross-national differences in integration remain, some of which may well be attributable to differences in foreign economic policy choices. Let me now explore the causes of these two phenomena, beginning with the secular global trend to more internationally integrated markets.

THE UNIQUENESS OF CONTEMPORARY MARKET INTEGRATION

Economic historians have been quick to point out that on many basic indicators the world economy is no more globalized today than it was a hundred years ago.11 From this perspective, the big story of the 20th century was the dramatic reduction of international economic activity in the middle decades. Maurice Obstfeld and Alan Taylor's summary judgment is representative of the nascent conventional wisdom: The era of the classical gold standard, circa 1970 to 1914, is rightly regarded as a high-water mark in the free movement of capital, labor and commodities among nations. After World War I, the attempt to rebuild a world economy along pre-1914 lines was swallowed up in the Great Depression and in the new world war the Depression bred. Only in the 1990s has the world economy achieved a degree of economic integration that rivals the coherence already attained a century earlier [Obstfeld and Taylor 1997: 1]. The staggering costs of the 1914-1945 period are certainly a central fact of the 20th century from which we no doubt still have much to learn. But is it appropriate to portray the contemporary era as merely a return to the pre-existing "equilibrium" level of globalization? There is already a revisionist economic history claiming that, despite apparent similarities, international market integration today is qualitatively different than

14

it was a hundred years ago. According to Michael Bordo, Barry Eichengreen and Douglas Irwin [1999], for example, facile comparisons with the late 19th century notwithstanding, the international integration of capital and commercial markets goes further and runs deeper than ever before.12 The evidence in support of this view seems straightforward. In the 1870-1914 period, the bulk of and the fastest growth in world trade was in raw materials (agriculture and minerals), as the industrial revolution reduced the costs for the first industrial nations of extraction and transportation from their colonies. Today, international trade is dominated by manufactures, not only among the OECD countries but also both ways between north and south as well. Trade in services was unheard of one hundred years ago, but it is of considerable and rising importance these days. The nature of international capital movements also clearly differs between the two epochs of internationalization. Most international lending in the earlier period was directed to raw material extraction and transportation to market, particularly in developing countries. In the contemporary period, international finance supports the gamut of production activities around the globe. The uniqueness of the contemporary international economy is nowhere more apparent than with respect to the multinationalization of production. The basic features of todays multinational firms captured in management jargon such as breaking up the international value chain and global strategic alliances have no historical parallels.13 One clear indication of the proliferation of multinational production is the estimate that intra-firm trade (i.e. among international affiliates of the same firm) comprises roughly one-third of all global trade [Jones 1996: 56].

15

One need not embrace all the hyperbole of international management gurus to accept the fundamental point that there does indeed seem to be something new and distinctive about the contemporary era of international market integration. But this only raises the questions of what has caused the mushrooming of international economic activity in recent decades.

TECHNOLOGICAL DETERMINISM
The core question addressed in this section is: if governments wish to restrict cross-border economic activity, can they do so?14 The analytic difficulty in answering this question is that one cannot draw any firm conclusions about the feasibility of closure from the extent of government interventions designed to insulate domestic markets from international activity. The global trend to declining barriers could be the product either of a voluntary choice by governments to liberalize or the product of their resignation that they cannot affect cross border trade, production and capital movements even if they try. On the other hand, countries that persist with protectionist barriers might do so not because they actually affect economic behavior, but rather because they send signals of support to constituencies adversely affected by market integration.

International Finance
The technological determinism thesis regarding international finance is straightforward [Bryant 1987]. Nowhere is globalization's ballyhooed shrinkage of time and space more apparent than in international finance. Ever faster and bigger semiconductors, fiber optics and the Internet have radically cut the costs of transmitting information in the past twenty years. Financiers can literally operate wherever and whenever they like, cutting deals in whatever financial instruments they can dream up. It is the specter of truly footloose liquid capital that generates images of hapless

16

governments seeking to regulate yesterday's financial instruments within their borders while the essentially homeless market makers they are trying to control have already moved on to newer and more exotic types of transactions. The first piece of evidence cited to demonstrate the difficulty of regulating international financial flows predates the IT revolution.15 The euromarkets (financial transactions in a national currency that occur outside the home country) became central to international finance in the mid 1960s when the US government responded to the weakening of America's balance of payments by imposing various policy restrictions on cross-border capital flows. In response, American banks moved their operations to London to avoid the new regulations.16 When faced with the enlivened euromarkets that it had unwittingly created, the US government had little choice but to do away with its capital controls which it did in the early 1970s. There were significant costs to offshore operations in the 1960s in terms of moving the relevant information halfway around the world (a three-minute New York-London telephone call, for example, cost over $30). Nonetheless, American bankers thought that the benefits of evading domestic regulation outweighed these costs. Today, of course, even individual consumers pay less than 50 cents for the same international call. This is why the predicament of governments trying to regulate international capital flows seems even more parlous than was the case thirty years ago. Lester Thurow [1997: 72] describes an infamous 1990s analog of the euromarkets story: (t)he Japanese government tried to prevent the trading of some of the modern complex financial derivatives that depended upon the value of the Nikkei Index in Toyko. As a result, the trading simply moved to Singapore, where it had exactly the same effects on the Japanese stock market as if it were done in Toyko. This was dramatically brought home to the world when a single trader for Barings securities in Singapore (Nick

17

Leeson) was able to place a $29 billion bet on the Nikkei Index and lose $1.4 billion when the index did not trade within the ranges that he expected. More generally, very few economists these days believe that governments can effectively control capital outflows (Krugman [1999] is a notable exception). The situation is more complicated with respect to capital inflows. Michael Dooley [1995] concluded from an extensive study of the empirical literature on the 1980s that capital controls did have real consequences for cross-border economic flows. More recently and visibly, key policy makers with exemplary credentials as academic economists including the IMFs interim Managing Director, Stanley Fischer [New York Times, January 8, 1998], Joseph Stiglitz [New York Times, February 1, 1998], then Chief Economist of the World Bank, and Alan Blinder [1999: 57], former Vice Chairman of the Board of Governors of the Federal Reserve have all argued that one clear lesson of the Asian crisis is that capital controls can and should be used to mitigate the adverse affects of volatility and uncertainty in international financial markets. Much of the optimism about the effectiveness of capital controls is based on 1990s Chile. According to Stiglitz, You want to look for policies that discourage hot money but facilitate the flow of long-term loans, and there is evidence that the Chilean approach does this" [NYT Sunday, Feb 1, 1998]. Chile is a darling of neoclassical development economists because of its manifestly successful efforts radically to reduce government intervention in the economy in the past two decades. But one area in which the Chilean government violated neoclassical principles concerned the imposition of capital controls. In 1991 government imposed the requirement that all (non-equity) foreign capital inflows be accompanied by a non-interest bearing one-year deposit equal to 30% of the initial

18

value of the investment.17 Since the deposit was only for one year, it was essentially a tax whose effective cost to investors declined the longer their money stayed in Chile. Instead of trying to fight the losing battle of stemming capital flight once a crisis hit, the Chilean controls were designed to reduce the prospect of a financial crisis by steering capital from shorter-term to longer-term investments. According to the Chilean central bank, this policy was very effective in channeling capital flows from investments with shorter-term to longer-term maturities. In the first year after the controls were introduced, Banco Chile estimates that capital inflows with maturities of less than one year declined from almost three-quarters of total inflows to less than 30% [Edwards 1999: 74]. Other economists are considerably more skeptical as to the effectiveness of even this type of "smart" controls in the IT age. According to Peter Garber [1998: 30]: a system of reserve requirements that penalizes short term inflows in favor of longer term investments can be evaded through offshore swaps with call features; an apparently longterm flow can thereby be converted into an overnight foreign exchange loan. Furthermore, Sebastian Edwards [1999] arguably the leading expert on capital controls in Latin America concludes that the Chilean controls were remarkably ineffective. Edwards argues that the Chilean controls clearly failed with respect to two of the government's stated objectives. They did not slow down currency appreciations caused by capital inflows, nor did they allow the government to fight inflation by maintaining higher domestic interest rates. But Edwards even goes so far as openly to dispute Banco Chiles claims about long term investments. He argues that the portion of short-term foreign loans in the Chilean portfolio in the latter 1990s was no smaller than in those of other comparable countries with open current accounts [Edwards 1999: 75].

19

Kenneth Rogoff [1999: 35] seems sympathetic with Edwards conclusion and argues that the pre-requisites for making Chilean-type capital controls work are very exacting. Rogoff reasons that to be effective: "domestic banks must be prevented from writing offshore derivative swap contracts with foreign holders of long-term Chilean debt." But this is exceedingly difficult given the multiplicity of potential offshore transactions. Rogoff continues, that "(b)y including suitable margin and call conditions, such contracts can effectively make a Chilean bank the true holder of the long-term income stream, and the foreign bank the holder of a short-term loan." It is probably premature to declare that capital controls are wholly ineffectual. Nonetheless, few would disagree with the more tempered proposition that the IT revolution has made it much harder for governments to controls international capital movements even if they want to for economic or political reasons.

Trade and Multinational Production

No one would deny that technological change has significantly affected international trade in recent decades. But the case for a technologically determined view of trade liberalization is weak. The simple reason is that moving physical goods across national borders is a relatively transparent activity that governments can therefore monitor and slow down if they so choose. To be sure, national borders are long, and smuggling is an age-old strategy for circumventing barriers to trade. But illegal trafficking in goods such as narcotics is the exception rather than the rule. Moreover, while trade in services is obviously less transparent, this does not seem an insurmountable obstacle to government regulation in countries with reasonable accounting standards.

20

There is, however, one scenario in which technological change might significantly decrease the feasibility of trade protectionism. The Internet is "digitizing" some heretofore-physical goods. In the world of e-commerce, it is not only be possible to buy music, movies and books on-line, the "goods" can also be "shipped" via the Internet. It is hard to see how such trade could be regulated using traditional policy instruments. There must be a limit, however, as to how much goods commerce can be morphed into bits and bytes. A popular statistic today is that about half the Americans buying cars use the Internet for some part of the process. But no one is suggesting that the cars will be delivered electronically anytime soon. One can undertake a similar thought experiment with respect to whether IT makes it harder for governments to regulate the activities of multinational firms. Buying a lasting stake in foreign assets or building new plants abroad are perhaps even more transparent activities that are easier for governments to regulate than moving goods across borders. However, the IT revolution has been intimately connected with the rise of international strategic alliances among firms, the key feature of which is that they do no entail transferring any equity. One could thus paint a scenario in which multinational firms could evade government restrictions on their activities by forging informal alliances rather than swapping equity. But as Oliver Williamson [1975] pointed out long ago, there are good corporate governance reasons why alliances tend to be for specific purposes rather than ongoing management structures for the broad range of firms' activities, and it is hard to see the balance being radically changed by the Internet.

21

Summary
The case for a technologically determined view of globalization is strongest with respect to international finance. There is a credible argument that since the onset of the information technology revolution there is essentially nothing governments can do to stop global financial flows. On such a view, there is no mystery to the spate of national-level moves to capital account liberalization in the 1990s; if capital controls don't work, why risk sending negative signals to the financial markets by persisting with them? The case for technological determinism is considerably weaker with respect to trade and the multinationalization of production. Governments that wish to impede the movements of goods across national borders can do so; they can also regulate the ownership of domestic firms and the external behavior of their multinationals. This may change somewhat in an era of mature e-commerce, but it is unlikely that the ability of governments to regulate trade and multinational production will be wholly emasculated any time soon.

THE COSTS OF CLOSURE

Anyone who has taken an introductory international economics course knows the logic behind the mantra that removing barriers to cross-border economic activity is always welfare-improving. But a more complicated picture emerges if one reads cutting edge research on the costs and benefits of international openness. The potential benefits of capital mobility in terms of the efficient allocation of investment are clear, but theses gains may often not be realized because of the incomplete information problems that are endemic to international financial markets. Freer trade certainly gives consumers lower prices and allows economies to exploit comparative advantages and scale economies. But there are sound theoretical arguments that trade liberalization can hinder

22

economic growth in cases of imperfect markets or positive externalities from domestic production, and the empirical evidence on the subject is inconclusive. Indeed, foreign direct investment is the only case where economists universally endorse the basic neoclassical approach to market integration.

Trade
A simple answer as to why we have seen so many moves towards freer trade in recent decades is that the opportunity costs of closure have increased as a result of rapid technological change. To take the classic example, the advent of super-freighters led to a decline in sea freight unit costs of almost 70% from the early 1970s to 1996 [World Bank 1997: 37]. More generally, the portion of national economies that are considered "nontradable" has decreased dramatically in recent decades.18 Indeed, this is a direct reflection of technological progress because something is non-tradable by definition if the difference between the local and the international price is greater than the cost of bringing it to the domestic market. A simple corollary of the increasing proportion of national economies that are tradable is that the deadweight losses associated with protectionism have increased apace. It is important to note, however, that these oft-cited benefits of freer trade are in essence one-time gains. Once the price for a product in a domestic economy is as low as the world price that is the end of the story. Development economics, however, has been concerned with a dynamic issue: whether freer trade stimulates economic growth in the medium term. In the 1950s and 1960s, the conventional view was that protecting infant industries from international competition was the appropriate development strategy for

23

most countries. After all, it seemed to have worked not only in post WWII Latin America and Western Europe, but also in the ante bellum US. Import substitution industrialization has fallen into disrepute since the 1970s. The Washington consensus has moved firmly to support the view that freer trade is good for growth.19 Influential articles using large-n statistics claim to show empirically that there are large positive growth effects to freer trade around the world [Balassa 1985, Sachs and Warner 1995]. The theoretical justification for the purported dynamic gains from trade comes from new growth theory in which technological innovation is endogenous. In these models, freer trade could increase innovation by creating scale economies in export sectors that allow for higher R&D expenditures, by speeding up technological diffusion in import-competing sectors, and by giving domestic firms access to the best and cheapest intermediate inputs. The "trade is good for growth" argument, however, is subject to important criticisms. Theoretically, it is easy to construct models in which freer trade retards growth. Strategic trade theory is a well known example, but its relevance is limited by the fact that it only claims benefits of protection in sectors with extremely high start up costs and very large minimum efficient scales of production (commercial aircraft and pharmaceuticals are exemplars) [Krugman 1987]. Of more potential importance is the argument (which in many ways formalizes the intuitions behind ISI) that freer trade leads to the undersupply of beneficial production externalities in import-competing sectors [Bhagwati 1968]. Consider the following simple example. At t0, a country can import widgets more cheaply than it can produce them. But if it goes ahead and produces these widgets (by

24

protecting the infant industry), a number of other things will happen the technology used in widget production may be useful in other sectors, for example. At t1, the country might be better off if it had protected widgets than if it had imported them.20 Indeed, the theoretical uncertainty about trade and growth is sufficiently great to lead Robert Lawrence and David Weinstein [1999: 8] no friends of protectionism to conclude in a recent study: Theory is actually quite ambiguous on the dynamic effects of trade. There are some reasons to expect that increased international competition could accelerate productivity growth but also some reasons to expect the reverse. Turning to the empirical evidence, many economists have argued that conventional hero-villain characterizations of the decline of Latin America and the East Asian miracle are simply inappropriate. Dani Rodrik [1999] argues that it is wrong to blame ISI for Latin Americas economic problems in the 1970s and 1980s the effects of the oil crises were far more important. Paul Krugman [1994] and Jeffrey Sachs [1996] argue that trade had very little to do with the East Asian miracle. High savings rates and high levels of educational attainment mattered far more. Others contend that trade policy was central to the East Asian model, but that the relevant policy was the protection of infant industries from import competition, rather than trade liberalization [Amsden 1989, Wade 1990].21 Moreover, the large n-studies of the trade-growth nexus have also been trenchantly criticized on methodological grounds [Edwards 1993, Rodriguez and Rodrik 1999]. One fundamental objection is that the causality is the reverse of that assumed. Fast growth and higher income levels promote trade. Jeffrey Frankel and David Romer [1999], for example, consider this causality question so important that they deliberately

25

try to exclude from their growth regressions any parts of trade volumes that could be attributed to either wealth or trade policy by using an instrumental variables approach in which unalterable aspects of geography are proxies for natural levels of trade. Frankel and Romer might be right that countries that are closer to each other grow more quickly, but this could hardly be used as support for the position that trade liberalization is good for growth and this is why governments have chosen to open their economies. In sum, it is undeniable that technological change has increased the gains from trade as they are conventionally understood. Thus, there is a good argument to be made for the proposition that the trend towards trade liberalization around the world in recent decades is explicable in terms of the increased opportunity costs of closure. However, things get much murkier if one considers the potential impact of trade liberalization on countries medium term growth trajectories.

International Finance
Though it is less prominent in the policy discourse on globalization, conventional international economics endorses not only free trade, but also free finance, as being in the interests of all countries. Maurice Obstfeld effectively summarizes the textbook argument [1998: 2-3]: International financial markets allow residents of different countries to pool various risks a country suffering a temporary recession or natural disaster can borrow abroad. Developing countries with little capital can borrow to finance investment, thereby promoting economic growth without sharp increases in savings rates The other main potential positive role of international capital markets is to discipline policymakers who might be tempted to exploit a captive domestic capital market. Unsound policies would spark speculative capital outflows and higher domestic interest rates.

26

By extension, the cheaper and easier it is to move information across borders, the greater are the efficiency gains of openness. Thus, there is a simple argument that the trend to financial market integration can be explained in terms of the heightened costs of financial closure. This is the view of the IMFs Interim Committee, which went so far in September 1997 as to recommend that all members commit themselves through a treaty revision to open capital accounts paralleling their extant commitments to current account convertibility.22 Bradford De Long, former Deputy Assistant Secretary for Economic Policy in the Clinton administration, believes that the benefits of capital mobility have been mammoth: the ability to borrow abroad kept the Reagan deficits from crushing US growth like an egg, and the ability to borrow from abroad has enabled successful emerging market economies to double or triple the speed at which their productivity levels and living standards converge to the industrial core [quoted in Bhagwati 1998: 10]. Most economists these days, however, are less bullish about the benefits of unfettered capital mobility. Any potential benefits of financial integration must be balanced against a series of costs generated by the fact that financial transactions are plagued by problems of incomplete and asymmetric information. Moreover, these problems are only exacerbated rather than mitigated as the costs of transmitting information decrease.23 The most important contemporary manifestation of these problems is that international financial markets are subject to wild swings in sentiment that are if not wholly irrational [Morris and Shin 1999] certainly unpredictable. Financial crises have been with us for centuries [Kindleberger 1984]. But they seem to have become more frequent

27

and more damaging in recent years from the Latin American debt crises of the early 1980s to the East Asian flu of the late 1990s. The causes of the Asian crisis are hotly debated, ranging from unalloyed panic [Sachs and Radelet 1998] to bad fundamentals [Corsetti, Pesenti and Roubini 1999]. But even proponents of the latter view accept that the volume and speed of global financial flows have rendered most emerging markets (and as the EMS crises of 1992-3 showed, even stable developed countries as well) vulnerable to essentially instantaneous switches between good" (rapid growth fueled by vast capital inflows) and "bad" (widespread capital flight precipitating deep recession) equilibria with no apparent change in underlying economic conditions. As Kenneth Rogoff [1999: 25] explains: If creditors suddenly become unwilling to roll over short-term loans as they fall due, a country may find itself in a financial squeeze even if, absent a run, it would have no problems servicing its debts. Devotees of the this "multiple equilibrium view believe that this is precisely what happened in the case of, say, Mexico in 1994 or Korea in 1997. For example, creditor panic at a relatively small devaluation of the peso in December 1994 suddenly made it impossible for Mexico to roll over its short-term debt, quickly precipitating a crisis. Instead of humming along in a "good" growth equilibrium as Mexico seemed to be doing prior to the crisis, it suddenly bounded into a "bad" recessionary equilibrium. Thus, there are good reasons to think that while free international finance is hypothetically allocationally efficient, informational problems likely generate numerous costs as well. How much do we know about calibrating the trade-offs between allocational efficiency and damaging volatility, and hence about the net effects of financial market integration? Even a defender of capital account liberalization like Fischer [1998: 8] admits that the answer is not much: The difference between the analytic understanding of capital- and currentaccount liberalization is striking. The economics profession knows a great deal about current account liberalization, its desirability, and effective

28

ways of liberalizing. It knows far less about capital account liberalization. It is time to bring order both to thinking and policy on the capital account. Despite this acknowledgment, Fischer [1998: 2] is nonetheless happy to endorse financial liberalization for all countries, arguing that the best evidence in their favor is that that essentially all of the OECD countries now have open capital accounts. But of course, the causality may well run in the other direction wealthier countries are more likely to liberalize finance, rather than financial openness increasing wealth. Rodrik [1998] has gone further. He claims that there is no good evidence that capital mobility is good for growth. His methodology is simple: he adds a capital account policy variable to a typical growth regression equation of the type pioneered by Robert Barro [1997] (that is, controlling for initial level of wealth, educational levels, regional effects, etc). Rodriks [1998: 61] conclusion is stark: Capital controls are essentially uncorrelated with long-term economic performance once we control for other determinants. One need not go all the way with Rodrik to conclude that the case is at best weak that there are clear economic benefits to financial market integration, and that these benefits have increased in recent years. Thus, it is hard to argue that increasing opportunity costs of closure can have been an important driver of financial globalization.

Foreign Direct Investment

Like trade and financial integration, the textbook argument for the increasing costs of closure to foreign direct investment centers around the efficient allocation of resources, and the fact that these gains have increased as a result of technological change in recent decades. Unlike the other two facets of market integration, however, there is little dispute

29

in the economics community that the effects of FDI are unambiguously positive from the standpoint of economic growth. Technological change has had a marked impact on multinational firms. The costs of product innovation have skyrocketed in many sectors (particularly those with the highest value-added such as aviation, computers, pharmaceuticals, etc.). This has greatly increased the minimum efficient scale of production for numerous industries and hence the benefits of multinationalization. Declining transportation costs have also made multinational production more efficient because they lower the costs of moving goods among locations in diversified and complex production regimes. As in the case of finance, however, it is arguable that the IT revolution has had the biggest impact on multinational firms. The Internet has radically reduced the costs of coordinating complex supply, production and distribution networks that are geographically decentralized. The automobile industry is a classic example. It may long have been efficient for Volkswagen to buy gear boxes in the US, build engines in Germany, assemble cars in Brazil, and sell the finished product cars all over the world. But the challenges of coordinating all this activity are immense, especially if VW wants to pursue just-in-time production/low inventory best practices. Being able to coordinate all elements of the supply and distribution chains on the World Wide Web has been a boon for firms that have incentives to decentralize their activities.24 But is more multinational activity good for the national economies among which it is distributed? Theory and evidence are strongly supportive.25 Interestingly, the case does not need to rely on the notion that attracting foreign investors is beneficial to capital poor developing countries. This would suggest, for example, that FDI within the OECD would

30

have little impact on growth, whereas the evidence is that FDI is good for growth even in the wealthiest nations [Graham and Krugman 1991]. Rather, the key argument is that foreign direct investment is a conduit for the transfer of technology and less tangible knowledge assets such as management practices.26

Summary
This section has argued that the argument that increasing opportunity costs of closure have driven globalization is most persuasive with respect to the multinationalization of production and least persuasive for international financial integration. Trade occupies an intermediate place because while the static gains from trade liberalization are well known, it is less clear whether trade is good for growth.

IDEOLOGICAL CHANGE
The political center of gravity around the world with respect to economic issues fiscal prudence, deregulation and privatization, but also international market integration has shifted to the right in the past twenty years. The time line would highlight successively the Reagan-Thatcher revolutions, Francois Mitterrands neoliberal U-turn, Antipodean market making and market opening, the rise to power of the "Chicago boys" in Latin America, the collapse of communism, and the embrace of the third way-ism by governing social democrats in countries as diverse as Australia, Brazil, Britain, Germany and Poland. But is this ideological shift merely a description of political economic changes driven by other factors, or does it have independent causal weight as analysts like Helleiner [1994] would have it?27 In this section, I endogenize the trend towards market integration in terms of changes in the domestic balance of political power, rather than via the diffusion of economic ideas or coercion by international institutions.

31

The previous two sections have argued that technological determinism provides a compelling explanation for the trend towards international financial integration in the contemporary period, and that the efficiency incentives to liberalize FDI are large. Important political implications fall out of these economic arguments are clear. Charles Lindblom [1977: 172-3] famously argued that business enjoys a privileged position under capitalism because public functions in the market system rest in the hands of businessmen. He continues: (i)t follows that jobs, prices, production, growth, the standard of living and the economic security of everyone all rest in their hands A major function of government, therefore, is to see to it that businessmen perform their tasks governments cannot command business to perform .. They must therefore offer benefits to businessmen in order to stimulate the required performance. It is easy to see how this privileged position has been enhanced for international financiers and multinational firms.28 If multinational firms perform essential growth functions, governments have little choice but to pursue policies of which they approve such as removing impediments to their cross-border activities. Of course, some domestic constituents may oppose the selling of national assets to foreign entities. But if the aggregate economic benefits of FDI are sufficiently large, governments have strong incentives to support the multinationalization of production (and to find other ways to compensate those who feel adversely affected by this process). If governments believe that there is simply no way effectively to regulate cross-border capital flows, and if investment capital is a scarce good, they might as well accept this reality and focus their energies on dealing with the consequences of capital mobility. These are political arguments in that they contend that the increasing power of financiers and multinational firms has led governments to remove barriers to international

32

activity. But they also entail predicting policy choice from economic effects without knowing anything more about the details of domestic political interactions. To the extent that the previous two sections suggest that economics does not provide a parsimonious explanation for trade liberalization either in terms of technological determinism or increased opportunity costs of closure it may be more fruitful to analyze this trend in terms of struggles among distributive coalitions. There has been a proliferation of sophisticated work in political science in the past decade studying interest group and coalitions politics in the trade area much of it stimulated by Ronald Rogowskis [1989] seminal application of Hecksher-Ohlin-StolperSamuelson models to the political arena. But in a recent excellent review of this literature James Alt and his collaborators [Alt et. al. 1996] acknowledge that neither HOSS nor the contending approaches (Ricardo-Viner specific factors or increasing returns to scale) tells us very much about likely trade policy outcomes. In particular, the authors point out that the question of why the apparently strong political bias to protectionism has been significantly mitigated in recent years remains a mystery. It is relatively easy to explain the inherent political bias towards trade protectionism [Magee, Brock and Young 1989]. Consumers are the primary beneficiaries of reductions in barriers to imports because this will lower the prices of goods and services they buy. Both the owners and employees of protected industries, however, will be adversely affected by import competition profits, wages and jobs will be reduced. In the conventional story, the benefits of free trade are relatively small and spread throughout society, whereas the costs of free trade are concentrated in import-competing industries for whom trade policy is a life and death issue (plants may close, whole

33

industries may shrink radically, those affected will have their lives seriously altered). In turn, collective action problems cripple consumers in the political battle against the wellorganized intense interests of import-competing industries.29 This approach cannot, however, explain the over time trend towards trade liberalization. As Alt and his collaborators acknowledge, introducing more sophisticated models of trade preferences doesnt help much either. The limitation of preference/coalition-based approaches is that increased demands for liberalization are likely to be offset by increased demands for protection. As Frieden and Rogowski observe [1996: 42]: It (lower costs of moving products and information) leads to intensified demands for trade on the part of those firms and individuals closest to their countrys comparative advantage On the other hand, easier trade sharpens the desire for protection on the part of those farthest from their countrys own comparative advantage [Frieden and Rogowski 1996: 42]. Let me now offer an argument that might help explain why the balance of political power has tilted in favor of freer trade. In Magees formulation, competitive exporters sit on the sidelines in the battle between pro-trade consumers and protectionist import-competers. The assumption is that exporters dont care about domestic trade policy; they only want access to foreign markets. But in the contemporary world, exporters seem to be active participants in the domestic trade game. Consider the Clinton administration's threat in the early 1990s of imposing 100% import tariffs on luxury Japanese automobiles in response to what it considered protectionist barriers in the Japanese auto parts market. The whole point of this strategy was to mobilize support among influential Japanese exporters (i.e. the automakers) for the liberalization of their

34

home market. Indeed, exporters have generally been strong supporters of trade liberalization in Japan since the 1980s [Rosenbluth 1996]. The multinationalization of production may also have had a significant impact on the policy preferences of exporters. The more exporters import intermediate inputs to produce final products, the greater their stake in lowering the prices of imported goods (and hence removing protectionist barriers). But export interests are unlikely to be hamstrung by the kind of collective action problems that afflict consumers. One could reasonably expect that the increased pro-free trade activism of exporters has tipped the political balance in favor of freer trade. What, then, do we know about the causal impact of political change on the global trend towards market integration? At some level, it is surely right that the shift to the right on economic issues has been a proximate cause of international market integration. I have argued, however, that in the cases of financial integration and the multinationalization of production, these causal processes are political only in a quite narrow sense. The increasing power of finance and multinational firms caused by technological changes has resulted in public policies that are more consistent with their interests that is, increasingly open markets. Things are more complicated, and more political in the sense of the constellation of preferences and interest coalitions, with respect to trade. Again, however, it seems that the move to trade liberalization has its roots in technological changes that have changed the preferences of exporters with respect to protection of the domestic economy.

CROSS-NATIONAL VARIATIONS IN MARKET INTEGRATION

Understanding the big picture the over time worldwide trends toward more internationally integrated markets is clearly of critical importance to any analysis of the

35

causes of globalization. This justifies the amount of attention I have given to this subject in the preceding sections. But the magnitudes of enduring cross-national disparities in international economic flows and foreign economic policies are sufficiently large that they cannot be dismissed as mere noise on the path to a single seamless global market (much as some pundits would like to believe that this is the case). In this section, I present a simple comparative analysis of the political economy of the two foreign economic policy choices for which it is possible to gather reasonable data for a large number of countries around the world trade taxes/total revenues and whether countries impose significant restrictions on capital account transactions. Studying the politics of protectionism is a subject with a very long and distinguished pedigree, but I know of no efforts to compare all the countries of the world in the same analysis (for recent reviews of the voluminous empirical literature, see Ray [1990] and Rodrik [1994]). There are a couple of global studies of capital account openness, but these are either relatively apolitical [Leblang 1997] or quite preliminary [Garrett, Guisinger and Sorens 2000]. I explore the effects of four types of variables that have received considerable attention in the political economy literature economic size, the level of development, the balance of power between pro and anti-market forces (measured in terms both of partisan control of government and unionization rates), and the effects of formal political institutions (in this case, the extent to which political regimes are democratic).30 I employ the simplest possible cross-sectional research design - regressing foreign economic policy outcomes in the 1990s on the explanatory variables (using lagged, 1980s, values to mitigate the possibility of reverse causality).31

36

This type of analysis is useful for the simple reason that it gives us a first cut at discriminating among the numerous plausible explanations for cross-national variations in market integration. In the 1990s, for example, foreign economic policies were much more liberal in the OECD nations than in the worlds poorest countries (see Table 1). There are at least two clear differences between the two groups. The OECD countries are wealthy and have long histories of stable democracies; the poor countries have much shorter (if any) democratic histories. The regressions reported below directly address the question: to what extent does democracy or income level explain these variations? But they might also generate some insights into important what if questions about the future: will we indeed witness the creation of a seamless global economy if/when most of the world's countries become wealthier and stably democratic? It is important to note at this point, however, that if my preceding analysis is right there is a crucial difference between trade taxes and capital controls. Trade taxes remain an effective way of regulating international trade and it is not clear that the macroeconomic benefits of liberalization are overwhelming. In contrast, the impact of capital account restrictions on international capital movements was probably quite limited in the 1990s. Thus, unlike the trade case where the level of protection has significant effects of the material well being of different segments of society, the politics of capital account liberalization are likely to be more symbolic (sending signals to domestic constituencies about the governments broader orientation to the international economy). Both might be subject to distributional conflict, but in the case of capital account policy this conflict may well be more symbolic than real.

37

Trade Taxes
The first panel of Table 3 presents the results for the determinants of trade taxes in the 1990s. The first thing to note about this table is the impact of the size of countries on trade policy. Countries with larger populations have lower trade taxes because they have larger domestic markets and hence less to gain from openness in terms of price reductions, specialization or realizing scale economies than smaller countries. Based on the estimates in the first column of the table, trade taxes in a country with one hundred million people would have constituted almost 10 percentage points more of total trade volumes in the 1990s than in a country with ten million people. Table 3 about here The powerful effects of size are not surprising. But they are quite interesting with respect to speculation about the future of protections. Economists have suggested that the minimum feasible size of countries has declined substantially in recent years as a result of the lower costs of trade and other cross-border economic activity [Alesina and Spolaore 1997]. Thus, dramatic increase in the number of countries in the world in the past decade may also have reinforced the trend to globalization by leading to a reduction in policy barriers to international trade.32 Nonetheless, it would be unrealistic to assume that anytime soon the world will be comprised of thousands of very small, essentially free trading states. The growth of the EU, with the fears about new forms of protectionism that it has engendered, is a clear counter-example to the trend towards the breakup of larger states. The second clear finding from Table 3 is that countries at higher levels of development were less protectionist. Based again on the on first model, trade taxes/total trade revenues would have been 14 percentage points lower in a country with a per capita

38

income of $10,000 in the 1980s than in a nation with GDP per capita of $1,000. There are numerous possible explanations of this result. Countries with higher income per capita are likely to have relatively more owners of capital and skilled labor and to have relatively more specialized production profiles. Pace Alt et. al. [1996], all of these actors prefer liberalization over protectionism. It may also be the case that in higher income countries, the median voter consumes more imports, again making liberalization more likely. Moreover, governments in more developed countries seem better able to raise taxes from their citizens, allowing them to rely less on trade taxes. Once one controls for the effects of level of economic development, whether or not countries were democracies had no impact on the level of trade protection. This is consistent with the notion that democracy has two countervailing effects on economic policy [Przeworski and Limongi 1993]. On the one hand, democracy makes leaders more accountable to their citizens promoting trade liberalization to the extent that this is good for society as a whole. On the other hand, democracy also empowers distributional coalitions with intense interests, making higher levels of protectionism more likely [Olson 1993]. If these effects are largely offsetting, perhaps the preferences of different groups in society, rather than the formal political institutions governing their aggregation that matter most for policy choice. Alternatively, one could argue that more fine-grained institutional analysis is required concerning, for example, electoral systems [Rogowski 1987] or federalism and the separation of powers [McGillivray 1997]. Even if broad regime type does not seem consequential in this case, there is evidence that other types of mediating institutions do matter to trade liberalization. While left government did not significantly affect the size of trade taxes, the second column of

39

Table 3 shows that countries with higher union density were more protectionist.33 A country with 50% of its (non-agricultural) labor force unionized in 1985 is estimated to have had trade taxes that were almost 2.5 percentage points higher than a country within a union density of 10%. This finding makes eminent sense on the reasonable assumption that trade unions tend to over represent workers in less internationally competitive firms, industries and sectors. In terms of over time trends, the unionization results would also imply that if the OECD trend towards lower rates of unionization since the 1970s is a global phenomenon, this might also have added to worldwide trend towards trade liberalization.

Capital Account Openness

The results for capital account openness are quite similar to those for trade taxes. Countries with larger populations were less likely to have open capital accounts in the 1990s a country with a 100 million people would have had open capital accounts for three more years in the 1990s than one with 10 million. The estimated effect of moving from $1000 to $10,000 in GDP per capita was essentially the same, decreasing the number of years with open capital accounts in the 1990s by about three. Although the precise reasons for these effects may be somewhat different than was the case for trade liberalization, the same general dynamics are likely to obtain. The extent to which a country was democratic had no impact on capital account openness. However, greater trade unionization was associated with more capital account closure. A forty-point increase in union density would have reduced the period of openness in the 1990s by 1.6 years. Unlike the trade case, left governments were also

40

significantly associated with closure, decreasing the number of years with capital account openness in the 1990s by 0.9. But if the technological determinism thesis is correct, capital account policy these days has very little impact on the actual cross border movements of capital. Why, then, are the regression results so similar to those for trade protection? One simple answer would be to assume that governments and domestic constituencies continue to think, naively, that they do. Given that the preferences of different broad classes of actors are likely to be similar with respect to trade and capital account policy, we would then expect the two policy choices to be driven by similar dynamics. A more realistic rendering of this type of argument is that governments understand that some policy choices are more important for the general signals about the governments broader intentions (rather than for their specific effects in a given policy area). Capital account liberalization may be a case in point. Restrictions send signals to domestic constituents who feel that they would be adversely affected by openness that the governments cares about their concerns and is willing to act to defend their interests. But imposing controls on the capital account also send signals to mobile capital that the country imposing the restrictions is in important senses unfriendly. In smaller, wealthier countries with lower rates of unionization and more conservative governments, the costs of the negative market signal may well dominate the benefits of the positive signal to domestic constituencies whereas the opposite in true larger, poorer countries, particularly where the left and trade unions are strong.

CONCLUSION
The central analytic problem one faces when trying to understand the causes of globalization is to untangle the interrelations among three important phenomena rapid

41

technological change, mushrooming cross-border economic activity, and a spate of initiatives to liberalize foreign economic policies at the national, regional and global levels. This article explored two ways to try to tease out the causal pathways among these variables. First, I mined the vast literatures in international economics about the economic effects of trade, multinational production and international finance to reason backwards to the causes of integration in each of these markets. Second, I examined today's large cross-national variations in international market integration to ascertain whether they are likely over time to erode, ultimately resulting in a truly seamless global marketplace. Figure 4 summarizes my assessment of the contending big picture arguments about the causes of globalization. Notwithstanding important similarities with the last great era of internationalization a hundred years ago, global market integration today is qualitatively different and deeper today. Technological changes lowering the costs of moving goods and more importantly information have been the primary exogenous stimulus behind contemporary globalization. There are, however, three different pathways between this stimulus and market integration. Figure 4 about here Technological determinism provides a parsimonious explanation for the integration of international financial markets. The IT revolution has rendered capital controls much less effective than ever before. Governments that are unwilling to risk sending negative signals to the markets therefore have strong incentives to remove capital controls. They will still have to deal with the adverse consequences of financial integration, but this

42

would be the case irrespective of whether or not they sought to impose controls on cross border capital flows. The desire of multinational firms to expand their international activities has grown as the costs of moving goods and information have decreased. But the multinationalization of production would not have been possible without governments' removing barriers to foreign ownership of domestic assets. Governments have been willing to do so because FDI generates externalities such as the transfer of technology and management best practices that stimulate economic growth. Some segments of society may still be unhappy with foreign ownership of domestic assets, but if this has beneficial macroeconomic policies it may be relatively easy for governments to compensate the opponents of the multinationalization of production. Finally, trade liberalization has not been technologically determined governments can and still do impose policy restrictions on cross border trade in goods and services. Moreover, while the one-time gains of freer trade (in terms of lower prices, etc.) are obvious, whether this is also beneficial for or harmful to economic growth in the longer run is debatable. It is thus likely that more traditional political factors have played a larger role in trade liberalization than in the other two facets of market integration. I highlighted the fact that exporters have become increasingly interested in reducing protectionism at home either to reduce the prospect of foreign retaliation or because they rely heavily on imports as productive inputs and suggested that this may have tipped the domestic political balance in favor of liberalization. Turning to my cross-national analysis, four basic points stand out. First, one should not expect all national markets ever to appear equally globalized. The incentives for

43

larger countries to be open are simply considerably weaker than those facing smaller countries. Larger countries are always likely to be less integrated into international markets than smaller ones. Second, levels of development have a marked impact on the propensity for international market integration. Wealthier countries are more likely to be open to and integrated into global markets probably because more of their citizens are likely to benefit from this. Economists may be correct that in the long run income levels will tend to converge around the world. Thus, it is possible to envisage a scenario in which some of today's great disparities in market integration are lessened. But it is unlikely that the large cross-national differences in per capita income of the current era will disappear anytime in the foreseeable future. Third, there is little support in the cross-national evidence that democratization is conducive to market integration. As others have noted, democracy has ambiguous and countervailing effects on economic policy choice, including international openness. On the one hand, democracy makes leaders more accountable to their citizens, which would promote openness to the extent that market integration is welfare improving. But on the other hand, democracy empowers distributional coalitions with vested interests in resisting market liberalization. Finally, there is some evidence that traditional indicators of the balance of political power within countries have affected their openness to the international economy. Countries with left wing governments and powerful trade unions tend to be more closed, though the substantive magnitude of these effects is considerably smaller than those for country size and level of development. One might debate whether trade unions are in

44

secular decline or how centrist the nominal left is becoming, but answers to these questions are unlikely to have major effects on international market integration. Let me finish this article by suggesting some potential implications of my analysis of the causes of globalization for its consequences for domestic politics. International financial integration is essentially an irresistible force. The live questions, therefore, are how large the adverse consequences of market uncertainty and volatility are, and whether governments have the incentives and the capacity to mitigate these consequences through domestic policies. The multinationalization of production, in contrast, is likely to be welfare improving for most countries. One would thus expect that dealing with its consequences would not be a big issue in most countries. Finally, governments can still restrict trade if they want to. Trade liberalization may be welfare enhancing, but the benefits are likely to be smaller than those associated with the multinationalization of production. In turn, freer trade has significant distributional implications for different segments of domestic society, to which governments may seek to respond with policies of domestic redistribution. Assessing how governments balance trade liberalization with domestic compensation remains an important question. I will explore all of these issues in "the consequences of globalization".

45

REFERENCES
Alesina, Alberto and Enrico Spolaore. 1997. On the Number and Size of Nations. Quarterly Journal of Economics 112: 1027-1056 Alt, James E., Jeffry Frieden, Michael J. Gilligan, Dani Rodrik and Ronald Rogowski. 1996. The Political Economy of International Trade. Comparative Political Studies 29: 689-717. Amsden, Alice H. 1989. Asias Next Giant. New York: Oxford University Press. Balassa, Bela. 1985. Exports, Policy Choices and Economic Growth in Developing Countries after the 1973 Oil Shock. Journal of Development Economics 18(2) 2335. Baldwin, Richard E. and Philippe Martin. 1999. Two Waves and Globalization: Superficial Similarities, Fundamental Differences. NBER Working Paper 6904. Barro, Robert. 1997. Determinants of Economic Growth: A Cross-Country Empirical Study. Cambridge: MIT Press. Bates. Robert H. 1981. Markets and States in Tropical Africa. Berkeley: Unviersity of California Press. Beck, Thorsten, George Clarke, Alberto Groff, Philip Keefer, and Patrick Walsh. 2000. New tools and new tests in comparative political economy: The Database of Political Institutions. Development Research Group, The World Bank. Groff: Federal Department of Foreign Affairs (Switzerland). [DPI 2000] Bhagwati, Jagdish. 1968. Distortions and Imiserizing Growth: A Generalization. Review of Economics Studies 35: 481-485. Bhagwati, Jagdish. 1998. The Capital Myth. Foreign Affairs 77(3): 7-12. Blinder, Alan S. 1999. Eight Steps to a New Financial Order. Foreign Affairs 78(5):5163. Blomstrom, Magnus, Robert E. Lipsey and Mario Zejan. 1994. What Explains Developing Country Growth? NBER Working Paper 4132. Bordo, Michael D., Barry Eichengreen and Douglas A. Irwin. 1999. Is Globalization Today Really Different than Globalization a Hundred Years Ago? NBER Working Paper 7195. Borensztein, Eduardo, Jose De Gregorio and Jong-Wha Lee. 1998. How Does Foreign Direct Investment Affect Economic Growth? Journal of International Economics 45: 115-135. Brooks, Stephen. 2000. The Globalization of Production and International Security. Ph.D. dissertation, Yale University. Bryant, Ralph C. 1987. International Financial Intermediation. Washington, DC: Brookings Institution. Corsetti, Giancarlo, Paolo Pesenti and Nouriel Roubini. 1999. Paper Tigers: A Model of the Asian Crisis. European Economic Review 43: 1211-1236. Dooley, Michael P. 1995. A Survey of Academic Literature on Controls over International Capital Transactions. NBER Working Paper 5352. Drezner, Daniel. 1998. Globalizers of the World, Unite! The Washington Quarterly 21(1): 207-225. Dunning, John. 1993. Multinational Enterprises and the Global Economy. Reading, MA: Addison-Wesley.

46

Easterly, William, Robert King, Ross Levine and Sergio Rebelo. 1994. Policy, Technology Adoption, and Growth. NBER Working Paper 4681. Economist. 1999a. Finance: Trick or Treat. 91-92, October 23. Economist. 1999b. Meddling in the Currency Market. 96, November 20. Edwards, Sebastian. 1999. How Effective Are Capital Controls? Journal of Economic Perspectives 13(4): 65-84. Edwards, Sebastian. 1993. Openness, Trade Liberalization and Economic Growth in Developing Countries. Journal of Economic Literature 31: 1358-1393. Feldstein, Martin, and Charles Horioka. 1980. Domestic Savings and International Capital Flows. The Economic Journal 90: 31429. Findlay, Ronald. Relative Backwardness, Foreign Direct Investment and the Transfer of Technology. Quarterly Journal of Economics 92: 1-16. Finger, J. Michael, Merlinda D. Ingco and Ulrich Reincke. 1996. The Uruguay Round, Statistics on Tariff Concessions Given and Received. Washington DC, World Bank. Fischer, Stanley. 1998. Capital Account Liberalization and the Role of the IMF. In Should the IMF Pursue Capital Account Convertibility? Essays in International Finance 207, Department of Economics, Princeton University. Frankel, Jeffrey A. 1993. On Exchange Rates. Cambridge: MIT Press. Frankel, Jeffrey A. and David Romer. 1999. Does Trade Cause Growth? American Economic Review 89: 379-399. Frankel, Jeffrey A. and Alan MacArthur. 1988. Political vs. Currency Premia in International Real Interest Differentials: A Study of Forward Rates for 24 Countries. European Economic Review 32: 1083-1121. Frieden, Jeffry A. and Ronald Rogowski. 1996. The Impact of the International Economy on National Policies. In Robert O. Keohane and Helen V. Milner (eds.), Internationalization and Domestic Politics. New York: Cambridge University Press. Garber, Peter M. 1998. Buttressing Capital Account Liberalization with Prudential Regulation and Foreign Entry. In Should the IMF Pursue Capital Account Convertibility? Essays in International Finance 207, Department of Economics, Princeton University. Garrett, Geoffrey. 2000. Capital Mobility, Exchange Rates and Fiscal Policy in the Global Economy. Review of International Political Economy 7(1): 154-171. Garrett, Geoffrey. 1998a. Global Markets and National Politics: Collision Course or Virtuous Circle? International Organization 52: 787-824. Garrett, Geoffrey. 1998b. Partisan Politics in the Global Economy. New York: Cambridge University Press. Garrett, Geoffrey, Alexandra Guisinger and Jason P. Sorens. 2000. The Political Economy of Capital Account Liberalization. Yale University, MS. Garrett, Geoffrey and Peter Lange. 1995. Internationalization, Institutions and Political Change. International Organization 49: 627-656. Goldstein, Judith. 1997. International Law and Domestic Institutions. International Organization 50: 541-564. Graham, Edward M. and Paul R. Krugman. 1991. Foreign Direct Investment in the United States (2nd ed.). Washington, DC: Institute for International Economics.

47

Gruber, Lloyd. 2000. Ruling the World. Princeton: Princeton University Press. Guillen, Mauro. 2000. Is Globalization Civilizing, Destructive or Feeble? Annual Review of Sociology (forthcoming). Guisinger, Alexandra. 2000. The Supporting Conditions for Trade Liberalization. MS. Yale University. Gurr, Ted R. et. al. 1998. Polity III Update - Regime Characteristics, 1800-1998 (dataset). [Polity 1998] Helleiner, Eric. 1994. States and the Reemergence of Global Finance. Ithaca: Cornell University Press. International Labor Organization. 1997. World Labor Report, 1997-1998. Geneva: ILO. International Monetary Fund. 2000. International Financial Statistics (cd-rom). Washington, DC: IMF. [IFS 2000] Jones, Geoffrey. 1996. The Evolution of International Business: An Introduction. New York: Routledge. Katzenstein, Peter J., Robert O. Keohane and Stephen D. Krasner. 1998. International Organization and the Study of World Politics. International Organization 52: 647-686. Kindleberger, Charles P. 1984. A Financial History of Western Europe. London: Allen and Unwin. Kobrin, Stephen. 1997. The Architecture of Globalization: State Sovereignty in a Networked Global Economy. In John Dunning (ed.), Governments, Globalization and International Business. Oxford: Oxford University Press. Krasner, Stephen D. 1999. Sovereignty. Princeton: Princeton University Press. Krueger, Anne O. 1997. Trade Policy and Economic Development. American Economic Review 87: 1-22. Krugman, Paul R. 1994. The Myth of Asias Miracle. Foreign Affairs October/November. Krugman, Paul R. 1999. Capital Control Freaks. Slate 9/27. Krugman, Paul R. 1987. Is Free Trade Passe? Journal of Economic Perspectives 11(2) 131-44. Krugman, Paul R. and Maurice Obstfeld. 1991. International Economics: Theory and Policy (2nd ed.). New York: Harper Collins. Kurzer, Paulette. Business and Banking. Ithaca: Cornell University Press. Lawrence, Robert Z. and David E. Weinstein. 1999. Trade and Growth: Import-Led or Export-Led? NBER Working Paper 7264. Lindblom, Charles E. 1977. Politics and Markets. New York: Basic Books. Maddison, Angus. 1995. Monitoring the World Economy. Paris: OECD. Magee, Stephen P., William A. Brock and Leslie Young. 1989. Black Hole Tariffs And Endogenous Policy Theory. New York : Cambridge University Press, 1989. Mallampally, Padma and Karl. P. Sauvant. 1999. Foreign Direct Investment in Developing Countries. Finance and Development March: 34-37. Marston, Richard C. 1995. International Financial Integration. New York: Cambridge University Press. McGillivray, Fiona. 1997. Party discipline as a determinant of the endogenous formation of tariffs. American Journal of Political Science 41: 584-607.

48

Mishkin, Frederic S. 1999. Global Financial Instability. Journal of Economic Perspectives 13(4): 3-20. Moravcsik, Andrew. 1998. The Choice for Europe. Ithaca: Cornell University Press. Morris, Stephen and Hyun Shin. 1999. Coordination Risk and the Price of Debt. MS. Yale University. Obstfeld, Maurice. 1998. The Global Capital Market: Benefactor or Menace? NBER Working Paper 6559. Obstfeld, Maurice and Alan M. Taylor. 1997. The Great Depression as a Watershed: International Capital Mobility in the Long Run. NBER Working Paper 5960. Ohmae, Kenichi. 1995. The End of the Nation State. New York: Free Press. Olson, Mancur. 1993. Dictactorship, Democracy and Development. American Political Science Review 87: 567-576. Przeworski, Adam and Fernando Limongi. 1993. Political Regimes and Economic Growth. Journal of Economic Perspectives 7(3): 51-69. Ray, Edward John. Empirical Research on the Political Economy of Trade. In Colin Cater et al. (eds), Imperfect Competition and Political Economy. Boulder: Westview. Rodriguez, Francisco and Dani Rodrik. 1999. Trade Policy and Economic Growth: A Skeptic's Guide to the Cross-National Evidence. NBER Working Paper 7081. Rodrik, Dani. 2000. How Far Will International Economic Integration Go? Journal of Economic Perspectives 14(1): 177-186. Rodrik, Dani. 1999. The New Global Economy and Developing Countries. Washington: Overseas Development Council. Rodrik, Dani. 1998. Who Needs Capital-Account Convertibility? In Should the IMF Pursue Capital Account Convertibility? Essays in International Finance 207, Department of Economics, Princeton University. Rodrik, Dani. 1997. Has Globalization Gone Too Far? Washington, DC: Institute for International Economics. Rodrik, Dani. 1994. What Does the Political Economy Literature on Trade (Not) Tell Us that We Ought to Know? NBER Working Paper 4870. Rogoff, Kenneth. 1999. International Institutions for Reducing Global Financial Instability. Journal of Economic Perspectives 13(4): 21-42. Rogowski, Ronald. 1989. Commerce and Coalitions. Princeton: Princeton University Press. Rogowski, Ronald. 1987. Trade and the Variety of Democratic Institutions. International Organization 41: 203-223. Romer, Paul M. 1986. Increasing Returns and Long-Run Growth. Journal of Political Economy 94: 1002-10037. Rosecrance, Richard N. 1999. The Rise of the Virtual State: Wealth and Power in the Coming Century. New York: Basic Books. Rosenbluth, Frances. 1996. TITLE In Robert O. Keohane and Helen V. Milner (eds.), Internationalization and Domestic Politics. New York: Cambridge University Press. Sachs, Jeffrey D. 1996. Reforms in Eastern Europe and the Former Soviet Union in Light of the East Asian Experiences. NBER Working Paper 5404 Sachs, Jeffrey D. and Steven Radelet. 1998. The East Asian Financial Crisis. Brookings Papers on Economic Activity 1: 1-74.

49

Sachs, Jeffrey D. and Andrew Warner. 1995. Economic Reform and the Process of Global Economic Integration. Brookings Papers on Economic Activity 1995(1): 1118. Strange, Susan. 1998. Mad Money: When Markets Outgrow Governments. Ann Arbor: University of Michigan Press. Thurow, Lester C. 1997. A Global Economy. Foreign Policy Forum 1: 61-83. Tsebelis, George. 1995. Decision Making in Political Systems. British Journal of Political Science 25: 289-326. UNCTAD. 1995. World Investment Report 1995. Geneva: UN. Yergin, Daniel and Joseph Stanislaw. 1999. The Commanding Heights: The Battle Between Government and the Marketplace That Is Remaking the Modern World. New York: Simon and Schuster. Wade, Robert. 1996. Globalization and Its Limits. In Suzanne Berger and Robert Dore (eds.), National Diversity and Global Capitalism. Ithaca: Cornell University Press. Wade, Robert. 1990. Governing the Market. Princeton: Princeton University Press. Williamson, Oliver E. 1975. Markets and Hierarchies. New York: Free Press. Wood, Adrian. 1994. North-South Trade, Employment and Inequality. New York: Oxford University Press. World Bank. 1997. Global Economic Prospects and the Developing Countries: 1997. Washington: World Bank. World Bank. 1999. World Development Indicators (and cd-rom). Washington, DC: World Bank. World Trade Organization. 1996. Annual Report: 1996. Geneva: World Trade Organization.

50

Table 1. Cross-National Variations in Globalization in the 1990s Economic Flows International portfolio inv./GDP (%) Economic Policy Open capital account (years in Trade taxes/total trade (%) 1990s)

Trade/GDP (%) I. High income OECD Mean Stand. Dev. II. High income other (oil exporters and tax havens) III. Upper middle income

FDI/GDP (%)

67 37

3.3 2.1

7.2 6.9

0.9 1.3

7.5 2.8

165 95

5.6 4.7

5.7 4.4

20.4 22.4

3.7 4.6

98 50 IV. Lower middle income 87 36 V. Low income 66 34 World 83 48

3.7 3.1

1.9 1.7

14.2 13.3

2.9 3.9

3.2 3.2 1.4 1.4 3.0 2.8

1.6 3.2 0.3 0.4 3.0 4.5

19.9 14.4 25.7 13.8 16.3 15.2

2.2 3.2 0.7 1.6 2.6 3.6

Income categories from WDI 1999. High income - OECD: 1997 GNP per capita > $9656; other high income (e.g. oil exporters and island tax havens): same as OECD; upper middle income: $3126-9656; lower middle income: $786-3125; and low income < $786. Data are unweighted averages for all acountries in ech category. See Appendix 1 for the complete data set.

51

Table 2. Trade and Capital Flows in the 1990s Years with Open Capital Accounts 0.22 0.09 0.02 -0.33

Trade/GDP FDI/GDP Portfolio/GDP Trade taxes/trade

FDI/GDP 0.40

Portfolio/GDP Trade taxes/trade 0.15 0.16 0.27 0.12 -0.08

Figures are correlations among countries based on data in Appendix 1.

52

Table 3. The Determinants of International Openness

Ln(population)c Ln(GDP pc)d Left governmente Unionizationf Democracyg Intercept R2 Obs.

4.31*** -6.21*** 1.88 -0.03 92.01*** 0.51 103

Trade taxesa 5.76*** -7.59*** -0.06*** 0.44 103.46*** 0.65 53

Open capital accountsb -1.42*** -2.18*** 1.35*** 2.23*** -0.90* -0.04*** 0.04 0.00 -6.20* -13.91*** 0.38 0.57 146 59

Robust OLS regressions; *** p < .01, ** .01 < p < .05, * p < .10. Average trade taxes in 1990s (from Appendix 1). Number of years with open capital accounts in 1990s (from Appendix 1). Natural log of average 1980-1990 population [WDI 1999]. Natural log of average 1980-1990 GDP per capita (in constant 1997 dollars) [WDI 1999]. e. Average jeft party control of executive government, 1980-1990 [DPI 2000]. f. Rate of unionization of non-agricultural workers in 1985 [ILO 1997]. g. Average democracy autocracy scores, 1980-1990 [Polity 1998]. a. b. c. d.

53

Appendix 1. Globalization in the 1990s

Trade/GDP (%)a I. High income OECD Australia Austria Belgium Canada Denmark Finland France Germany Greece Iceland Ireland Italy Japan Luxembourg Netherlands New Zealand Norway Portugal Spain Sweden Switzerland United Kingdom United States Average Std. deviation II. High income other Aruba Bahamas, The Brunei Cayman Islands Cyprus Hong Kong, China Israel Kuwait Macao Malta Netherlands Antilles Qatar Reunion Singapore Slovenia United Arab Em. 38 77 129 63 66 58 44 47 42 66 123 43 18 182 100 58 71 68 42 64 68 53 22 67 37 FDI/GDP (%)b 3.1 1.8 7.9 2.6 3.3 3.0 3.4 1.4 1.0 0.9 3.8 1.0 0.6 . 7.6 5.7 3.4 2.3 2.5 5.7 4.7 5.1 1.9 3.3 2.1 Int. portfolio inv./GDP (%)c 4.5 6.3 36.0 5.5 7.1 6.0 4.5 5.8 . 1.1 4.0 6.3 3.4 . 7.6 3.8 4.8 7.4 4.9 8.3 9.4 10.9 4.3 7.2 6.9 Trade taxes/trade (%)d 3.2 1.1 0.0 2.7 0.1 0.7 0.0 0.0 0.1 5.1 0.0 0.0 1.2 0.0 0.0 2.2 0.6 0.5 0.6 0.7 1.1 0.1 1.4 0.9 1.3 Years with Open Capital Accountse 10 8 10 10 10 8 6 10 3 2 7 6 10 . 10 10 4 6 2 6 7 10 10 7.5 2.8

. . . . 104 277 78 101 129 192 . . . 361 121 123

9.8 . . . 1.5 . 2.4 5.1 . 5.6 . . . 13.7 1.2 .

. . . . 1.7 . 3.1 6.8 . 8.8 . . . 12.4 1.2 .

. 61.8 . 42.7 12.0 . 1.0 . . 16.6 28.1 . . 1.3 . 0.0

0 0 4 . 0 10 1 7 . 0 0 10 0 10 0 10

54

Trade/GDP (%) Average Std. deviation III. Upper middle income Antigua and Barbuda Argentina Bahrain Barbados Botswana Brazil Chile Croatia Czech Republic Estonia Gabon Grenada Hungary Korea, Rep. Lebanon Malaysia Mauritius Mexico Oman Panama Poland Saudi Arabia Seychelles Slovak Republic St. Kitts and Nevis St. Lucia Trinidad and Tobago Turkey Uruguay Venezuela Average Std. deviation IV. Lower middle income Albania Algeria Belarus Belize Bolivia 165 95

FDI/GDP (%) 5.6 4.7

Int. portfolio inv./GDP (%) 5.7 4.4

Trade taxes/trade (%) 20.4 22.4

Years with Open Capital Accounts 3.7 4.6

203 16 192 97 93 18 60 113 113 150 90 105 70 64 83 173 127 45 88 189 50 78 124 116 129 144 85 40 43 54 98 50

12.1 1.7 0.3 0.8 2.1 1.2 6.8 2.0 3.2 4.2 2.1 7.5 5.6 . . 6.6 1.1 1.3 1.1 . 2.6 0.9 8.0 1.7 7.8 7.3 5.8 0.5 0.7 3.4 3.7 3.1

0.1 4.1 0.9 1.8 1.3 3.0 2.1 0.3 2.6 3.1 . 0.2 3.8 . . 0.9 1.4 4.2 . 6.6 0.6 4.5 0.3 2.7 0.6 0.1 0.1 1.6 1.0 1.6 1.9 1.7

. 8.0 9.0 . 16.4 1.6 9.6 8.4 3.6 1.1 16.6 20.0 7.3 7.2 40.2 14.1 39.3 6.1 2.9 10.7 6.7 . 46.1 . 40.2 27.9 7.2 4.1 5.6 8.6 14.2 13.3

10 1 10 0 1 0 0 0 0 3 0 0 0 0 9 7 3 0 10 10 0 7 9 0 0 0 5 0 7 3 2.9 3.9

60 51 114 113 49

2.9 0.1 0.8 3.0 4.3

. 0.0 0.2 1.0 0.1

13.6 16.9 8.5 . 6.5

0 0 0 0 10

55

Bosnia/Herzegovina Bulgaria Cape Verde Colombia Costa Rica Djibouti Dominica Dominican Republic Ecuador Egypt, Arab Rep. El Salvador Equatorial Guinea Fiji Georgia Guatemala Guyana Iran, Islamic Rep. Iraq Jamaica Jordan Kazakhstan Kiribati Latvia Lithuania Macedonia, FYR Maldives Marshall Islands Micronesia, Fed. Sts. Morocco Namibia Papua New Guinea Paraguay Peru Philippines Romania Russian Federation Samoa South Africa Sri Lanka St. Vincent/Gren. Suriname Swaziland Syrian Arab Rep. Thailand Tonga Tunisia Ukraine

Trade/GDP (%) . 95 77 35 84 113 116 88 57 54 54 151 117 73 43 185 46 . 123 135 91 132 106 106 91 130 . . 58 113 98 48 26 76 55 53 108 47 76 120 33 167 67 83 79 89 69

FDI/GDP (%) . 1.3 3.0 3.7 3.5 0.4 11.6 2.8 2.6 1.3 0.2 . 4.0 . 0.8 . 0.0 . 5.2 1.6 4.4 1.2 3.5 2.1 . 3.1 . . 1.0 4.0 3.9 1.4 5.1 2.1 1.2 0.9 . 1.0 1.5 10.6 11.4 7.8 0.7 2.0 0.8 2.2 0.5

Int. portfolio inv./GDP (%) . 0.8 . . 0.2 . 0.7 . . 0.2 0.5 . . . 0.6 . . . . . 0.8 . 3.7 0.7 . . . . 0.2 1.7 16.8 . . 3.1 0.4 1.7 . 3.3 2.0 0.3 0.5 0.3 . 2.6 0.6 0.3 0.6

Trade taxes/trade (%) . 6.1 . 11.6 16.0 . . 40.4 12.0 10.2 . . 25.1 12.6 . . 9.0 . . 29.2 . . 2.8 3.4 . 37.8 . . 16.7 31.8 24.1 16.1 10.5 25.9 4.0 11.7 . 2.3 21.6 40.8 . 47.4 11.2 17.0 49.3 27.7 .

Years with Open Capital Accounts 2 0 0 0 4 10 0 1 8 2 3 0 0 3 10 3 0 0 3 2 0 7 4 6 0 10 4 4 0 0 0 3 6 0 . 0 . 0 0 0 0 0 0 0 2 0 0

56

Uzbekistan Vanuatu Yugoslavia, FR Average Std. deviation V. Low income Afghanistan Angola Armenia Azerbaijan Bangladesh Benin Bhutan Burkina Faso Burundi Cambodia Cameroon Central African Rep. Chad China Comoros Congo, Dem. Rep. Congo, Rep. Cote d'Ivoire Eritrea Ethiopia Gambia, The Ghana Guinea Guinea-Bissau Haiti Honduras India Indonesia Kenya Kyrgyz Republic Lao PDR Lesotho Liberia Madagascar Malawi Mali Mauritania Moldova Mongolia

Trade/GDP (%) 94 123 . 87 36

FDI/GDP (%) . 13.2 . 3.2 3.2

Int. portfolio inv./GDP (%) . . . 1.6 3.2

Trade taxes/trade (%) . 56.5 . 19.9 14.4

Years with Open Capital Accounts 0 10 0 2.2 3.2

. 118 94 85 24 58 77 39 34 49 41 42 47 35 60 44 120 69 108 29 124 54 45 49 33 78 23 52 65 78 54 149 . 47 61 54 100 120 109

. 4.8 . . 0.1 . . . 0.1 . 0.3 1.0 1.4 5.4 0.5 . . 1.2 . . 2.5 . 0.6 . 0.2 1.7 0.7 2.1 0.2 . 2.8 1.5 . 0.4 . 1.6 0.7 1.0 0.4

. . . . 0.1 . . . . . 0.5 . . 0.6 . . . 0.1 . . . . . . . 0.0 . 1.3 0.1 . 0.0 . . . . . 0.0 0.0 .

. . . . . . 1.3 23.6 21.7 . 19.0 . 19.8 14.5 . 34.7 . 29.1 . 17.1 42.8 33.1 46.3 . . . 24.3 5.0 13.6 . . 54.6 . 47.2 16.3 . . . 12.5

0 0 3 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 8 0 0 1 0 2 0 7 3 3 0 0 3 0 0 0 0 1 0

57

Mozambique Myanmar Nepal Nicaragua Niger Nigeria Pakistan Rwanda Sao Tome/Principe Senegal Sierra Leone Solomon Islands Somalia Sudan Tajikistan Tanzania Togo Turkmenistan Uganda Vietnam Yemen, Rep. Zambia Zimbabwe Average Std. deviation World Average Std. deviation a. b. c. d. e.

Trade/GDP (%) 65 4 49 78 38 80 37 33 111 62 46 120 48 . 187 54 69 . 31 76 61 75 66 66 34

FDI/GDP (%) 1.9 . 0.1 1.4 1.8 4.3 1.0 0.2 . 1.0 0.6 4.9 . . . 1.5 0.2 . 1.8 . . 1.0 0.3 1.4 1.4

Int. portfolio inv./GDP (%) . . . 0.1 . 1.1 0.8 0.0 . 0.1 . . . . . . 0.1 . . . . . 0.6 0.3 0.4

Trade taxes/trade (%) . 14.1 28.5 19.0 . . 26.1 29.5 . . 40.6 54.7 . . . . . . . . 16.5 21.9 18.2 25.7 13.8

Years with Open Capital Accounts 0 0 0 3 0 1 0 0 0 0 0 0 0 0 0 0 1 0 2 0 4 3 0 0.7 1.6

83 48

3.0 2.8

3.0 4.5

16.3 15.2

2.6 3.6

Average 1990-1996. Average 1990-1997. Average 1990-1997. Average 1990-1995. Number of years in 1990s with open capital accounts.

58

Figure 1. Global Market Integration, 1970-1997

world investment (% output) 50 world trade (% output) 45 40 35 30 25 20 1970 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 1975 1980 1985 FDI 1990 1995

trade

portfolio

Trade is exports plus imports, from WDI 1999. FDI is inflows and outflows of foreign direct investment; portfolio is assets and liabilities of international portfolios investments. Both from IMF IFS 2000.

59

Figure 2. Trade Flows and Trade Policy, 1973-1995

50 20

45 world trade (% output))

17 trade taxes/total trade (%)

40

14

35

11

30

25 1973

5 1978 1983 trade volume 1988 trade taxes 1993

All data from WDI 1999. Trade taxes are un-weighted annual averages for all countries. Correlation = -0.89

60

Figure 3. Capital Flows and Capital Account Policy, 1970-1997

8.00 7.00 6.00 0.40 5.00 0.35 4.00 0.30 3.00 2.00 1.00 1970 0.25 0.50
portion of all countries with open capital accounts

world FDI and int. portfolio flows (% output)

0.45

0.20 1975 1980 capital flows 1985 1990 1995

open capital accounts

Open capital accounts as defined by IMF, Exchange Arrangements and Exchange Restrictions (various). Correlation = 0.78.

61

Figure 4. The Causes of Globalization Facet of Market Integration Trade Multinational International Production Financial Integration Lower Lower transportation IT revolution transportation costs costs and IT revolution

Causal Perspectives Exogenous Stimulus Nothing New

Intra-industry trade, trade in services, 2way north-south manufacturing trade

Multinational supply, production and distribution networks, international strategic alliances

24 hour global trading, limitless derivative transactions

Technological Determinism

Movements of goods and services can still be controlled

Cross-border equity transactions can still be regulated

Offshore markets very difficult to regulate

Increased Costs of Closure

/?

Theoretical efficiency gains offset by in practice by uncertainty and volatility

FDI transfers Lower prices, etc., but uncertain growth technology, management practices, effects know-how

New Preferences and Coalitions

Exporters more interested in domestic liberalization

Privileged position of MNCs

Privileged position of financial capital

62

NOTES
1

In 1980, there were fewer than 300 articles or books with the word global or globalization in the title. In 1995, the number was over 3000 Guillen 2000: Table 2]. 2 See Drezner [1998] and Guillen [2000] for recent reviews of some of these issues. 3 Global foreign exchange transactions increased fully fifty-fold between 1980 and 1998 to reach two trillion dollars per day [Economist 1999a: 91, 1999b: 96]. This outstrips the foreign exchange reserves of all the OECD countries combined. World GDP in 1997 was approximately 34 trillion dollars. 4 International economists are quick to point out that increased international capital flows are not necessarily indicative of capital market integration (i.e. the absence of obstacles to international capital flows). But there is no consensus as to the extent to which capital markets are integrated these days. Feldstein and Horioka [1980] argued in a seminal paper that since national savings drove national investment in the 1970s OECD countries, there must have been considerable barriers to international capital movements. This result has been replicated many times (with some modifications) in the subsequent two decades. Frankel [1993] and Marston [1995], in contrast, use the relevant (covered) interest rate comparisons to argue precisely the opposite that OECD capital markets were highly integrated by the late 1980s. Extending these analyses outside the OECD is fraught with difficulty and preliminary results are inconclusive [Montiel 1995]. 5 Though systematic data are not readily available, this liberalization trend in the 1990s is also apparent with respect to the regulation of foreign direct investment. For example, UNCTAD [1995: xx] reported that in the 1991-1994 period there were 373 significant changes in FDI regulations enacted in countries throughout the world, and all but five of these were in the direction of fewer restrictions on inflows and outflows. 6 Also note that even in 1997, more than half the countries in the world still imposed significant restrictions on the capital account. 7 Note that these means are un-weighted averages for all countries. Hence, the flows data are not comparable with the global flows statistics reported in Figure 1. 8 Note also, that for these countries, the correlation between applied tariff rates and the trade tax measure used here was high (r = 0.75). 9 Note also that the tax havens make up for imposing no burdens on capital with very high trade taxes. 10 The data do not allow me to conduct the same exercise on a global sample with respect to the partial correlation between capital account openness and capital market integration (using covered interest rate differentials or savings-investment correlations). For the OECD countries, however, Frankel and McArthur [1988] demonstrated that even in the 1980s capital account openness was strongly positively correlated with greater capital mobility (measured in terms of smaller covered interest rate differentials). 11 Much of this work relies on and was inspired by the pathbreaking empirical research of Angus Maddison [1995]. 12 See also Baldwin and Martin [1999] for a similar argument. 13 For summaries, see Brooks [2000: chapter 4], Dunning [1993] and Kobrin [1997]. 14 I consider the issue in the context of individual governments vs. market actors. If there were evidence that individual governments are powerless to stop globalization, this would raise the issue of whether international cooperation would be more effective. I will address this question in The Consequences of Globalization. 15 For a clear and concise discussion of this case, see Krugman and Obstfeld [1991: 605-608]. 16 Of course, this would not have been possible had the British government mimicked the American regulations. They had little incentive to do so, however, because Britain could and did gain by becoming the world center for offshore financial transactions. Moreover, the British government must have known that there would have been many others willing to offer an unregulated environment catering to the offshore activities of American banks. 17 These controls were ultimately lifted in the aftermath of the Asian crisis. 18 Moreover, lower transportation costs also increase the gains from specialization and from scale economies, since the costs of importing a countrys comparative disadvantage decrease. 19 See, for example, Former Vice President of the World Bank, Anne Kruegers [1997] presidential address to the American Economics Association.

63

Of course, this argument requires that governments are relatively good at picking winners sectors and technologies that generate long future income streams. It is easier to do this when a country can mimic technological advances already made elsewhere than if it is at the technological frontier. For example, this may explain why Japanese efforts to pick winners seem to have become less successful over time. 21 In contrast, Lawrence and Weinstein conclude that while the East Asian economies certainly did practice import substitution, this actually hindered their development. As a result, their growth performance "was even more of a miracle than we thought [Lawrence and Weinstein 1999: 24]." 22 Following the Asian crisis, the IMF has backed away somewhat from this unconditional position. It now argues that countries should only open their capital accounts when the appropriate domestic institutions are in place most importantly, transparent and well-regulated domestic banking systems. 23 Mishkin [1999] provides an accessible review of these issues. Adverse selection occurs before a transaction when bad credit risks are more likely to seek out loans (even at very high interest rates) because they are less concerned with paying back their creditors. Moral hazard takes place after the transaction. Borrowers have incentives to invest in riskier projects than were agreed to at the time of contract. As a result of both problems, lenders will likely make fewer loans than they should, or would, if they were perfectly informed about the attributes of potential borrowers. 24 The Economist, Construction and the Internet - New wiring (1/15/00) argues that this type of Internet coordination has even had dramatic effects in sectors, like construction, that would apparently seem a long way from the cutting edge of e-commerce. 25 The benefits of foreign direct investment may also have implications for the costs of trade closure. The modern view about trade and FDI is that they are complements rather than substitutes (see, for example, WTO [1996: 53-55]). The reasoning is straightforward. If multinational firms are to realize the benefits of international systems of production and distribution, they need to be able to move inputs and intermediate goods among their operations in different countries via trade. 26 Findlay [1978] is the seminal theoretical article. For empirical support for the proposition that FDI has a positive impact on medium term growth, see Blomstrom [1994] and Easterly [1994]. Borensztein [1998] argues that this effect is contingent upon a minimal level of human capital perhaps explaining the relative absence of FDI in Africa and its apparently limited effects on growth rates on the continent. 27 See Gruber [2000] for a more sophisticated rendering of policy diffusion with respect to market integration. 28 Kurzer [1993] was among the first political scientists to see this connection. 29 Bates's [1981] seminal argument, of course, is that there is a countervailing urban consumer bias in many developing countries because small agricultural producers, unlike manufacturers in the stable industrial democracies, are plagued by pervasive collective action problems. 30 I would have liked to test the proposition that foreign economic policy liberalization should be less pronounced in countries with fewer veto players [Tsebelis 1995] but unfortunately the data are not available outside the OECD countries. 31 Of course, more rigorous analysis using panel data will be required before any more definitive conclusions can be drawn. 32 World population growth without the creation of new nations, of course, would have the opposite effect 33 Note that even though the sample of countries is roughly halved by the inclusion of union density, the results on the other variables are essentially unaffected.

20

64