The Journal of International Trade and Diplomacy 1 (2), Fall 2007: 1-33

HOW TO SAVE GLOBALIZATION FROM ITS CHEERLEADERS

Dani Rodrik*
I. INTRODUCTION When future economic historians write their textbooks, they will no doubt marvel at the miraculous turn the world economy took after 1950. Over the long stretch of history, neither the Industrial Revolution nor the subsequent economic catch-up of the United States and other western offshoots looks as impressive (Figure 1). The period since 1950 has witnessed more rapid economic growth than any other period before, with only the classical gold standard era between 1870 and 1913 coming close. Even more striking, there has been a quantum jump in the growth rate of the most rapidly growing countries since 1950. Prior to 1950, growth superstars experienced growth rates that barely surpassed 2 percent per annum (in per capita terms) over long stretches. Compare this with the post-1950 growth champions: Japan, South Korea, and China; each grew at 6-8 percent per annum during 1950-73, 1973-90, and 1990-2005, respectively. Even allowing for the shorter time slices, this indicates that the world economy became a much more enabling environment for economic growth after 1950. Clearly, the architects of this new world economic system got something right. Going forward, there can be few things more important than to maintain a global economic environment that is as enabling in the future as it has been in the recent past. This requires that we interpret the reasons behind the post-1950 boom appropriately. A simple its all due to globalization view receives little support from Figure 1. It is significant that the world economy experienced a more significant boost during 1950-73 than it did during either the post-1990 period of gung-ho
*

Kennedy School of Government, Harvard University. E-mail: [email protected]. I thank several commentators on my blog, in particular Paine, Steveb, and Andrew, for catching mistakes and typos.

Dani Rodrik

globalization or the transition period between 1973-90. Second, and perhaps even more tellingly, the countries that did best under each one of these periods were hardly poster children for open markets and laissez-faire economics. These countries combined orthodoxy on some (mostly macroeconomic) policy fronts with a good bit of heterodoxy on others (especially in microeconomic policies). Japan, South Korea, and China each played by very different rules than those enunciated by the guardians of orthodox globalizationmultilateral institutions such as the World Bank, IMF, and GATT/WTO and by Western-based academics.
Figure 1: The Expanding Growth Frontier
Historical experience with growth
9 8 7 6 5 4 3 2 1 0

GDP per capita growth rate of fastest growing country/region (annual average, %)

World GDP per capita growth rate (annual average, %)

Source: Maddison (2001) and World Development Indicators.

In this paper I present a forward-looking evaluation of globalization. I accept as my premise that globalization, in some appropriate form, is a major engine of economic growth (as Figure 1 amply demonstrates). However, I will argue that several paradoxical features require us to rethink its rules. First, as I already indicated, globalizations chief
2 JITD, Fall 2007

How to Save Globalization From Its Cheerleaders

beneficiaries are not necessarily those with the most open economic policies. Second, globalization has come with frequent financial crises and considerable amounts of instability, which are both costly and in principle avoidable. Third, globalization remains unpopular among large segments of the people it is supposed to benefit (especially in rich countries). It is not that these features have gone unnoticed in the recurrent debate on globalization. In fact, we can talk of a new conventional wisdom that has begun to emerge within multilateral institutions and among Northern academics. This new orthodoxy emphasizes that reaping the benefits of trade and financial globalization requires better domestic institutions, essentially improved safety nets in rich countries, and improved governance in the poor countries. With these institutions in place (or in construction), it remains safe and appropriate to pursue a strategy of more of the same, but better to continue to open markets in trade and finance, while strengthening institutions. Enhanced trade adjustment assistance (and perhaps more progressive taxation) in the advanced countries, the Doha trade agenda, IMF surveillance over exchange rate policies, the World Banks governance agenda, aid-for-trade, and international financial codes and standards are some of the visible markers of this approach. This strategy is predicated on the presumption that insufficiently open markets continue to pose an important constraint on the world economy. Its proponents concerns therefore center on the question: what institutional reforms are needed at home and internationally to render further market opening politically acceptable and sustainable? Is this presumption really valid? I shall argue here that lack of openness is (no longer) the binding constraint for the global economy. I will provide a range of evidence on trade and capital flows that indicates that the obstacles faced by developing countries do not originate from inadequate access to markets abroad or to foreign capital. The gains to

JITD, Fall 2007

3

Dani Rodrik

be reaped by further liberalization of markets are meager for poor and rich countries alike. This leads me to an alternative approach to globalization, one that focuses on enhancing policy space rather than market access. Such a strategy would focus on devising the rules of the game to better manage the interface between national regulatory and social regimes. A good argument can be made that it is lack of policy spaceand not lack of market accesswhich is (or likely to become soon) the real binding constraint on a prosperous global economy. This argument can be buttressed by current evidence from rich and poor countries along with reference to historical experience with the previous wave of globalization. What do we mean by policy space and can we really create it without running into the slippery slope of creeping protectionism? By the end of the paper, I hope I will have given the reader some reason to believe that an alternative conception of globalizationone that is more likely to maintain an enabling global environment than the path we are on currentlyis worth thinking about and potentially workable. II. THE PARADOXES OF GLOBALIZATION AS WE KNOW In 2001 the World Bank published a volume entitled Globalization, Growth, and Poverty: Building an Inclusive World Economy. In it, the Bank identified four countries as star globalizerscountries that had greatly increased their integration with the world economy and at the same time had grown rapidly and made progress with poverty reduction. The countries were China, India, Vietnam, and Uganda. With the possible exception of Uganda, these still constitute Exhibit A of the case for globalizations benefits. However, as Table 1 shows, these countries policies can hardly be described as being of the free-trade type. In fact, by standard measures, such as the height of import tariffs and prevalence of non-tariff barriers,
4 JITD, Fall 2007

How to Save Globalization From Its Cheerleaders

India, China, and Vietnam were among the most heavily protected countries in the early 1990s. China and Vietnam were not even members of the WTO and therefore could engage in policiessuch as trade subsidies and quantitative restrictionsthat are unavailable to other countries. In each one of these cases, whatever trade liberalization took place happened with a significant delay after the onset of economic growth. For example, China significantly reduced its trade barriers in the mid-1990s and beyond, but this came after at least fifteen years of rapid growth. It is true of course that these countries greatly increased their volumes of trade and inward foreign investment, but that is precisely the paradox. They did so despiteand in fact because oftheir heterodox strategies. Simply put, countries that have benefited the most from globalization are those that did not play by the rules.
Table 1. Trade Policies in the World Banks Star Globalizers of the 1990s
growth rate China Vietnam India Uganda 7.1 5.6 3.3 3.0 average tariffs (early to mid-90s) 31.2 30-50 50.5 14.4 NTBs? yes yes yes WTO member? (early to mid-90s) no no yes yes

Poverty: Building an Inclusive World Economy, 2001.

Note: List of star globalizers taken from World Bank, Globalization, Growth, and

By contrast, Latin America, which tried harder than any other part of the world to live by the orthodox rules, experienced on the whole a dismal performance since the early 1990s. This occurred despite the boost provided by the natural bounce-back from the debt crisis of the 1980s. Here the paradox is not just that Latin America did worse than Asia, it is also that Latin America did worse than its pre-1980s performance (Figure 2). Lets recall that the pre-1980s were the era of import substitution, protectionism, and macroeconomic populism. That the region did better with these discredited policies than it has under open-

JITD, Fall 2007

5

Dani Rodrik

market policies is a fact that is quite hard to digest within the conventional paradigm.
Figure 2. Comparative Growth Rates
7.0% 6.0% 5.0% 4.0% 3.0% 2.0% 1.0% 0.0% -1.0% East Asia & Pacific South Asia

Economic growth
6.4% 5.6%

3.3% 2.8%

3.3%

3.3%

1.2%

1.0%

1960-1980

1980-90 -0.8%

1990-2003 Latin America & Caribbean

Source: World Development Indicators.

Let us now turn to international finance. Here the paradox is that financial globalization, which was supposed to enable developing countries to augment their savings, raise their investment levels, and grow more rapidly, has in fact done little of any of it. Such is the absence of evidence of these direct benefits that the new wave of research on financial globalization focuses on a search for its indirect benefits (Kose et al, 2006). Neither has financial globalization enhanced risksharing and consumption smoothing for developing nations; if anything, the evidence points in the reverse direction (Kose et al, 2007). Instead, one noticeable consequence of financial globalization has been a series of costly financial crashesin Mexico, Thailand, Indonesia, South Korea, Russia, Argentina, Turkey, and many other countries. This has forced developing nations to accumulate huge amounts of foreign reserves in order to purchase self-insurance against the fickleness of financialmarket sentiment. As Figure 3 shows, reserves in developing countries
6 JITD, Fall 2007

How to Save Globalization From Its Cheerleaders

(measured in months of imports) now stand at 3-4 times the levels observed in the advanced countries. In view of the low yields and high social opportunity cost of these reserves, the costs of this strategy amount to almost 1 percent of GDP in developing nations annually (Rodrik, 2006).
Figure 3. Trends in Foreign Reserves

9 8 7 6 5 4 3 2 1
1956

Foreign reserves (excluding gold) in months of imports industrial and non-oil developing countries

1962

1976

1982

1996

2002

1950

1952

1954

1958

1960

1964

1966

1968

1970

1972

1974

1978

1980

1984

1986

1988

1990

1992

1994

1998

2000

Industrial Countries

Developing Countries (excl. oil-exporting countries)

Source: Rodrik (2006).

The world economy has avoided similar financial crises in recent years. This indicates that developing nations have become more resilient to financial turbulence. That is due, in no small part, to the mountains of liquidity that developing nations now sit on. It is also due to the fact that many emerging markets are now running trade surplusesi.e., lending money to the rest of the world. In other words, in order to protect themselves from the whiplash of financial crises, developing countries have been forced not only to shun its benefits, but to make transfers to rich countries on top!

JITD, Fall 2007

2004

7

Dani Rodrik

The third paradox is that globalization remains restricted in precisely those areas where further relaxation of barriers would yield the greatest economic benefits. Barriers on labor mobility in particular are inordinately higher than they are anywhere else. It is easy to show that even minor reductions in labor-market barriers would generate gains that are vastly larger than those from the conventional areas under negotiation in the WTO and elsewhere. Consider for example a temporary South-to-North labor mobility scheme that would expand the industrialized countries labor force by about three percent (0.03). (The existing share of immigrants and of foreign-born in advanced country labor markets is about ten percent on average.) Lets use, as our estimate of the earnings gap, the figure $17,500, which is Mark Rosenzweigs estimate (in unpublished work) of the earning premium for a Mexican worker in the United States. Given a Northern labor force of around half billion (0.5), the net gains generated for nationals of poor countries would amount to $262.5 billion per annum (0.5 x 0.03 x 17,500). By contrast, current estimates of the gains to developing countries from the completion of the Doha Round do not exceed $30 billion. The disparity between these two numbers reflects the fact that poor countries already have fairly good access to rich country markets where goods are concerned, but are virtually shut out from Northern labor markets (with only a few exceptions). Many tears have been shed about the recent demise of the Doha Round. Meanwhile multilateral negotiations on reducing barriers to labor mobility are not even on the agenda. A recent proposal in the United States Senate to institute a temporary guest worker program was eventually killed, alongside the proposed immigration reform. Finally, lest one think that globalization has produced paradoxical outcomes just for developing countries, it is worth pointing out that globalization remains quite unpopular in most advanced countries. A recent Financial Times poll reports that individuals in the United States and other rich countries who believe globalization is a positive force are

8

JITD, Fall 2007

How to Save Globalization From Its Cheerleaders

greatly outnumbered (by a factor of between 2 to 3) by those who think it is a negative force. The same poll reports that large majorities of people in the US and in Europe want higher taxation for the rich and even pay caps for corporate executives to counter what they believe are unjustified rewards and the negative effects of globalization (Financial Times, July 22, 2007). This is only the most recent of a long series of polls that show globalizations benefits are either not filtering down, or if they are, not being adequately appreciated. It has become harder over the years to ascribe such views to sheer ignorance and the search for easy culprits. It is remarkable that many outspoken and prominent boosters of globalization such as Paul Krugman (2007), Lary Summers (2006), and Alan Blinder (2006) have recently acknowledged that globalization is contributing to inequality and insecurity (see in particular Krugman 2007 for an explanation of why his views have changed). The search is on for tax and adjustment policies that would respond to the middle-class malaise that is widely felt to endanger the sustainability of globalization. III. THE NEW CONVENTIONAL WISDOM In response to the developments I have just reviewed, a new conventional wisdom has been emerging during the last few years. The main contours of this emergent consensus can be described as follows: Globalization is indeed contributing to rising inequality, stagnant median wages, and the growing sense of insecurity in the advanced economies, even if it is still unclear as to what extent globalization is the dominant influence. This is in sharp contrast to the views expressed by most establishment economists during the trade and wages debate of two decades ago, in which blame was ascribed to skill-biased technological change rather than to globalization, but the rise of China and of global outsourcing has made those earlier views untenable.

JITD, Fall 2007

9

Dani Rodrik

Trade and financial openness are unlikely to lead to economic growth on their own, and may occasionally even backfire, in the absence of a wide range of complementary institutional and governance reforms. This is in sharp contrast to the views expressed in the literature on trade and growth of some 10-15 years ago, in which the assertion was that trade liberalization in particular has an unconditional and strong effect on economic growth on its own--even absent other reforms. Once again, the evidence has rendered the older views untenable. Therefore, globalization requires a range of institutional complements in both rich and poor countries in order to deliver its benefits in full and remain sustainable. In the advanced countries of the North, the complementary measures relate in large part to improved social safety nets and enhanced adjustment assistance. In the developing countries, the requisite institutional reforms range all the way from anti-corruption to labor market and financial market reforms. This new conventional wisdom finds expression in a multipronged effort to deepen globalization in its current form. One element is the completion of the Doha development round, with its focus on agricultural liberalization. At present, the Doha process seems dead on its tracks due to a combination of unwillingness of the rich countries to offer substantial cuts in agricultural supports and market access as well as the reticence of developing nations to offer low enough bindings on their own tariffs. Another element is the promotion of cautious capital account opening in developing countries, by a coalition of the IMF and financial interests in the developing countries themselves. A third flank is the governance agenda of the multilateral institutions, focusing on anticorruption at the World Bank and financial regulation and supervision at the IMF. A fourth is the ongoing discussion in the U.S. and other advanced countries on a menu of proposals to take the pain out of globalization: increased progressivity in taxation, enhanced adjustment

10

JITD, Fall 2007

How to Save Globalization From Its Cheerleaders

assistance, portability of health insurance, and wage insurance to cover part of the income losses due to dislocation. Many of these efforts are useful in their own right, of course. Essentially we can conceive of this strategy as the answer to the following question: what institutional reforms are needed at home and internationally to render further market opening politically acceptable and sustainable? The maintained hypothesis behind it is that the greatest bang for the global reform buck lies in pushing for increased openness and market access, while ensuring that the adverse consequences of openness are taken care of. In the next section, I will question the validity of this view, and suggest that the binding constraint on maintaining a healthy global economy lies elsewhere. IV. HOW CONSTRAINING ON ECONOMIC GROWTH ARE THE REMAINING BARRIERS TO INTERNATIONAL ECONOMIC INTEGRATION? Under certain conditions, economic integration can be a powerful force for economic convergence, and it can promote rapid economic growth in poorer regions. The historical experience of the United States is telling in this regard. The U.S. experienced economic convergence within its own national economy, but did so only after a truly common and integrated set of product, capital, and labor markets were established nationally. What the U.S. experience shows is that achieving integration of this kind is not just a matter of eliminating barriers to interstate commerce (which are prohibited in the constitution itself). National economic convergence entailed a lot more: the elimination (through a civil war) of divergent social institutions in the South, the establishment of a truly free labor market with no interstate restrictions, the creation of the Federal Reserve system, a national financial regulator and lender-of-last resort, a national system of fiscal transfers, a Federal legal system, and a Supreme Court to prevent individual states from re-imposing restrictions. Put simply, a national market required a national polity. It was only then that
JITD, Fall 2007 11

Dani Rodrik

the

transaction

costs

associated

with

jurisdictional

and

legal

discontinuities across state lines eroded sufficiently to permit economic convergence. The European Union today presents another interesting case, since it displays the significant growing pains associated with efforts to create seamless integration. The EU is currently engaged in a process of legal, institutional, political integration that is somewhat similar to the process that took the United States some two centuries. Some key markers in this process are the vast supranational regulatory apparatus that is in construction (the acquis communautaire is more than 90,000 pages in length and growing by the day); the European Court of Justice (which has the power to strike down national laws and regulations); an almost Europe-wide monetary system (which eliminates monetary autonomy and restricts fiscal policy); the various cohesion and structural funds aimed at pulling up lagging regions; and of course full labor mobility within the union. Even with all this, convergence within the EU is far from complete, and remains a stop-and-go process. In the absence of the legal and political integration of the type that the U.S. has already achieved and the EU is attempting to construct, transaction costs condemn the global economy to a patchwork of national economies. This is the model of shallow integration in contrast to the US/EU models of deep integration (the terms have been coined by Robert Lawrence). In a world of shallow integration, the prospects for convergence are doomed to remain incomplete. Capital flows are hindered by sovereign risk and the absence of international financial regulation and bankruptcy procedures. Financial panics and crashes are rendered more likely by the absence of a true international lender-oflast resort. Labor can flow in very small quantities, and often only illegally. Differences in national regulatory, legal, and currency regimes also impose severe transaction costs on international trade (estimated by Anderson and van Wincoop [2004] to be on the order of 40% in tariff equivalents). Net capital flows end up being too small, and often go in

12

JITD, Fall 2007

How to Save Globalization From Its Cheerleaders

the wrong directionthat is, from poor to rich nations. Trade flows similarly remain too small, compared to intranational trade, and national borders exert a significant depressing effect on trade even in the absence of import duties or other government-imposed barriers. Many of the paradoxical syndromes of globalization that I discussed above are a direct consequence of this incompleteness. What is the implication of all this for developing countries? The vast majority of countries do not face the realistic option of full economic integration with their rich trade partners; legal and political integration la EU or U.S. model is not on offer, and even if it were, national sovereignty is perceived to be too valuable for many to give up. They need to recognize, therefore, that they are living in a second-best world, in which international economic integration remains incomplete due to the transaction costs I just discussed. Living in a second-best world requires second-best strategies; if markets cannot solve your problems of labor surplus and capital shortage (because the former is not free to leave and the latter comes only in small quantities), you need roundabout policies. You may need to postpone import liberalization in order to protect employment for a while. You may need to subsidize your tradables to achieve more rapid structural change. In fact, you may need a whole range of industrial policies in order to build technological and productive capacity at home (Rodrik, 2004). This line of reasoning help us understand why some countries that sharply lowered their barriers to trade and capital flows are still waiting for the rewards, while others who have been much more cautious have done so much better. Consider for example the contrast between El Salvador and Vietnam. Both countries returned to peace and political stability after a long period of civil war. Vietnam started its reforms in the late 1980s, while El Salvadors reforms came in the early 1990s. El Salvador quickly eliminated all quantitative barriers to imports, slashed tariffs, established convertibility on the capital account, and dollarized its economy. It became an open economy in both trade and financial

JITD, Fall 2007

13

Dani Rodrik

senses of the term. It also became the recipient of large amounts of remittances from its expatriates in the United States. Vietnam, meanwhile, followed a Chinese-style reform, based on gradual external liberalization, pragmatism, and a concerted effort to diversify the economy through public encouragement and investment where needed. Vietnam did not rush into the WTO, and has only just become a member. Looking at these policies wearing conventional lens, it would be hard to see how El Salvadors policies could have been improved. Yet private investment and growth remained lackluster in El Salvador. Meanwhile Vietnam achieved phenomenal success in terms of both growth and poverty reduction. These examples can be multiplied many times over. Perhaps the most notable development failure of the last fifteen years is Mexico. This is a country that has free and preferential access to the U.S. market for its exports, can send several millions of its citizens across the border as workers, receives huge volumes of direct investment and is totally plugged into U.S. production chains, and for which the U.S. Treasury has acted as a lender of last resort. It is hard to imagine a case where globalization gets any better. Yet even though trade and investment flows have expanded rapidly, the results have been underwhelming, to say the least, where it mattersin economic growth, employment, poverty reduction, and real wage growth. NAFTA, it turns out, is another instance of shallow integration. The standard response when cases like El Salvador and Mexico are brought out is to point out, in line with the new conventional wisdom I summarized previously, that these countries did not undertake the complementary reforms needed to make globalization work. What specific reforms are in question depends on who is talking and when, but the usual line is that both countries need more judicial and (in the case of Mexico especially) structural reform. However, this is hardly a satisfactory response in light of the fact that successful countries that did not open themselves up as fully to international trade and finance

14

JITD, Fall 2007

How to Save Globalization From Its Cheerleaders

had, if anything, even worse institutional preconditions. It is difficult to argue that Vietnam or Chinatwo authoritarian socialist economies with extensive state ownership and widespread corruptionhad the institutional prerequisites which Mexico and El Salvador lacked. This standard riposte reflects once again the habit of using first-best reasoning when circumstances demand second-best thinking. It is of course trivially true, but largely besides the fact, that if Mexico and El Salvador had first-world institutions, they would be as rich as the advanced countries. Successful growth strategies are based on making the best of what you have, not on wishing you had what you lack (Rodrik, 2007a). Given successive rounds of multilateral trade liberalization and the extensive unilateral liberalization that developing countries have already undertaken, the shallow integration model has already run into strong diminishing returns. This is one reason why Doha has stalled. There are simply not enough gains to get people excited. Consider the numbers in Table 2, which show World Bank estimates of the real income gains from complete liberalization of global merchandise trade (Most Doha scenarios do not get us anywhere near there, so these numbers represent an upper bound). By 2015, developing countries stand to reap gains on the order of one percent of GDP, while developed country gains are substantially smaller. At this point, further global trade liberalization is hardly a force for economic convergence. These and similar numbers are credible even if the general-equilibrium models on which they are based remain largely a black box, because they are based on reasonable assumptions and accord with back-of-the envelope calculations. There are also studies that generate estimates that are orders of magnitude larger, and which have been used to counter the kind of skepticism I have expressed with regard to the value of pushing hard on the liberalization frontier. But these have to be approached with extreme caution. Some of them have to be treated as works of advocacy rather than credible scholarship. Notable in this vein

JITD, Fall 2007

15

Dani Rodrik

is a study by Bradford et al. (2006) which puts the prospective gains from removing remaining barriers to trade in goods and services in the U.S. at $4,000 to $12,000 per household. These are huge numbers if they could be believed but such numbers are arrived at by making the most favorable assumptions possible so that the numbers come out large. One set of estimates rely on increasing returns to scale, but ignores that the presence of scale economies make all kinds of outcomes possibleincluding losses from liberalization! Another is based on assuming that further liberalization will eliminate price differences across national markets in goods and services, but leaves open the question of what the policy levers are to get us there. The best interpretation we can put on these estimates is that they represent the payoff to what I have called deep integration. They presume a degree of legal and institutional harmonization with practices of other countries that I believe even the authors would balk at if spelled out explicitly. Another potentially large source of gains from enhanced global liberalization is enhanced product variety and expansion of the range of available inputs. Work by Broda and Weinstein (2006) and by Broda, Greenfield, and Weinstein (2006) has shown that this channel can add several percentage points of GDP to income and substantially enhance productivity growth. But these studies too have to be interpreted with caution. At the root of these gains lies the entry into world markets of fast-growing countries like Taiwan, South Korea, China, and India, which have diversified their exports at a dizzying pace. As discussed previously, none of these counties owe their success to trade liberalization proper or to playing the globalization game according to the standard guidebook. If the realistic gains from further trade liberalization are small, what about international finance? Once again, we can imagine a world where international capital flows make a significant contribution to fostering development in the poorer countries of the worldthat would be a world

16

JITD, Fall 2007

How to Save Globalization From Its Cheerleaders

Table 2. Impacts on Real Income from Full Liberalization of Global Merchandise Trade, by Country/Region, 2015 (Impacts in 2015 Relative to the Baseline, in 2001 dollars)
Real Income gain ($ billion) Australia and New Zealand EU 25 plus EFTA United States Canada Japan Korea and Taiwan Hong Kong and Singapore Argentina Bangladesh Brazil China India Indonesia Thailand Vietnam Russia Mexico South Africa Turkey Rest of South Asia Rest of East Asia Rest of LAC Rest of ECA Middle East and North Africa Selected SSA countries Rest of Sub-Saharan Africa Rest of the World High-income countries Developing countriesWTO definition Developing countries Middle - income countries Low-income countries East Asia and Pacific South Asia Europe and Central Asia Middle East and North Africa Sub-Saharan Africa Latin America and the Caribbean World Total 6.1 65.2 16.2 3.8 54.6 44.6 11.2 4.9 0.1 9.9 5.6 3.4 1.9 7.7 3.0 2.7 3.6 1.3 3.3 1.0 5.3 10.3 1.0 14.0 1.0 2.5 3.4 201.6 141.5 85.7 69.5 16.2 23.5 4.5 7.0 14.0 4.8 28.7 287.3 Gain due just to change in terms of trade ($ billion) 3.5 0.5 10.7 -0.3 7.5 0.4 7.9 1.2 -1.1 4.6 -8.3 -9.4 0.2 0.7 -0.2 -2.7 -3.6 0.0 0.2 -0.8 -0.9 0.0 -1.6 -6.4 0.5 -2.3 0.1 30.3 -21.4 -29.7 -16.7 -12.9 -8.5 -11.2 -4.0 -6.4 -1.8 2.2 0.6 as % of baseline income in 2015 1.0 0.6 0.1 0.4 1.1 3.5 2.6 1.2 0.2 1.5 0.2 0.4 0.7 3.8 5.2 0.6 0.4 0.9 1.3 0.5 1.9 1.2 0.3 1.2 1.5 1.1 1.5 0.6 1.2 0.8 0.8 0.8 0.7 0.4 0.7 1.2 1.1 1.0 0.7

Source: Anderson, Martin, and van der Mensbrugghe (2005).

JITD, Fall 2007 17

Dani Rodrik

of deep integration. That is very far from the world we inhabit presently, with the various transaction costs imposed by fragmented sovereignty. Our world is one where countries that rely more on external capital markets grow less rapidly. That is the astonishing finding of a recent paper by Prasad et al. (2007). The bottom line of this paper is depicted in Figure 4, which shows a positive correlation across a wide range of countries between the current account surplus (net lending abroad) and the growth rate of the economy. For a variety of reasons, countries do not do that well when they attract capital from abroad. The authors show that this is a robust result, and likely has to do with various imperfections in credit and product markets which are aggravated in the presence of capital inflows (The Dutch disease is the clearest case). This is another reminder that a proper evaluation of the prospective benefits from further liberalization requires second-best thinking.
Figure 4. Countries with Less Access to Foreign Savings Grow More, Not Less!

Source: Prasad, Rajan, and Subramanian (2007).

18 JITD, Fall 2007

How to Save Globalization From Its Cheerleaders

In sum, the likely gains from further liberalization in goods and capital markets are small, as long as the world remains politically fragmented and transaction costs emanating from jurisdictional discontinuities prevent deep economic integration. V. SAVING GLOBALIZATION FROM ITS CHEERLEADERS: THE NEED FOR POLICY SPACE Strong diminishing returns may have set in on the prevailing

liberalization agenda, but the losses from a real retreat from todays globalization would be catastrophic. A collapse towards protectionism and bilateralism la 1930s can never be ruled outit has happened beforeand would be bad news for poor and rich nations alike. Therefore, we ought to place a high premium on policies that make such a retreat less likelyeven if they run contrary (in the short run at least) to a market-opening agenda. In order to maintain globalization in some version of its current form, we need to diagnose well the problems that confront it. These problems do not arise, as I argued in the previous section, from liberalization not having gone far enoughunless, that is, we are ready to envisage deep integration as a feasible option. They originate instead from something that is closer to the opposite, namely the clash between the liberalization agenda and the weakness of the institutional underpinnings that make open markets functional and politically sustainable. Once we put the problem this way, the challenge becomes not how do we liberalize further, but how do we create the policy space for nations to handle the problems that openness creates. The policy space in question would allow:

Rich nations to address issues of social insurance and concerns about

the labor, environmental, and health consequences of trade; and

Poor nations to position themselves better for globalization through

economic restructuring and diversification.
JITD, Fall 2007 19

Dani Rodrik

In this section I will make a case for such policy space by showing that globalizations constraints do bite where legitimate economic and social ends are concerned in these two sets of countriesand will bite even more if we continue to pursue a market-opening agenda. I begin with the advanced countries. Consider the following dilemmas that our present arrangements pose.

Labor standards:
Domestic labor laws protect workers from being displaced through unfair employment practices at home, such as the hiring of child labor or the employment of workers under hazardous conditions. WTO rules do not make room for similar protections when displacement occurs through trade but why should trade be allowed to contravene an established domestic norm (Rodrik 1997)?

Environmental, health and safety standards:

If European citizens want to apply a higher precautionary standard because this has an effect on trade? than other countries, should trade rules prevent them from doing so

Regulatory takings:
Why should foreign firms in the U.S. receive greater protection from and bilateral investment treaties (BITs) may require)? policy changes that affect their profits than domestic firms (as NAFTA

Redistributive provision of social insurance:

If taxation of capital and skilled professionals has historically helped international mobility be allowed to undercut this social compact? fund social insurance programs and generate equity, should their

Currency policies and unfair trade:

WTO rules recognize the concept of unfair trade in cases of explicit subsidization of exports and allow importing countries to respond

20

JITD, Fall 2007

How to Save Globalization From Its Cheerleaders

through countervailing duties. Should countries that undervalue their currencies, and hence subsidize their exports in non-fiscal ways, be allowed to get away with it?

Trade versus technological change:

Domestically, R&D and technological progress are highly regulated (cf. the stem cell controversies). Why should trade, which is analogous to technological change, be left unregulated as a rule?

These are all difficult questions, without clear-cut answers. They will likely increase in salience with the explosive growth in off-shore provision of services. The appropriate locus for their discussion and resolution is most likely the national polity, given the wide variety of standards and norms that prevail across the globe. If so, countries will need the policy space with which they can act on their deliberations. Sometimes dilemmas of the kinds illustrated above are pooh-poohed by economists as instances of self-interested pleading on the part of lobbies adversely affected by imports. However, there is a variety of evidence that points to more than narrow self-interest being at work in rich countries. For example, when Alan Krueger (1996) examined where the support for a Congressional bill aimed against child labor was coming from, he found that the support was strongest not in districts with a concentration of low-skilled labor, but in well-to-do districts with preponderantly skilled labor. People were against child labor not because it meant more competition, but because they felt it was wrong. Similarly, recent research by Hiscox and Smyth (2006) documents significant willingness-to-pay by U.S. consumers for improved labor standards in developing nations. Also, in our analysis of attitudes to trade in a large cross-section of countries, Anna Maria Mayda and I found that individuals with negative attitudes towards trade and globalization were motivated only partly by labor-market concerns and pocket-book issues (Mayda and Rodrik, 2005); values and norms mattered too. In particular, we found that individuals with high levels of attachment to their
JITD, Fall 2007 21

Dani Rodrik

neighborhood and immediate community were more likely to have negative views on trade. When economists talk about comparative advantage and gains from trade, they typically ignore whether trade opportunities involve exchanges that most people would consider unacceptable if they took place at home. It is immaterial whether the gains from trade are created, say, by a company shutting down its factory at home and setting up a new one abroad using child labor, but the archetypal person on the street reacts differently to trade-induced changes in distribution than to technology-induced changes (i.e., to technological progress). Both increase the size of the economic pie, while often causing large income transfer, but a redistribution that takes place because home firms are undercut by competitors who employ deplorable labor practices, use production methods that are harmful to the environment, or enjoy government support is procedurally different than one that takes place because an innovator has come up with a better product through hard work or ingenuity. Trade and technological progress can have very different implications for procedural fairness. This is a point that most people instinctively grasp, but economists often miss. (Notice that even in the case of technology, we have significant restrictions on what is allowablecf. human-subject review requirementsand wide-ranging debates about the acceptability of things like stem-cell research.) So globalization is a hot button issue in the advanced countries not just because it hits some people in their pocket book; it is controversial because it raises difficult questions about whether its outcomes are right or fair. That is why addressing the globalization backlash purely through compensation and income transfers is likely to fall short. Globalization also needs new rules that are more consistent with prevailing conceptions of procedural fairness. Turning to developing countries, where do the constraints bite? As I have already argued, successful development strategies often require secondbest and therefore unorthodox policies. Current thinking has moved
22 JITD, Fall 2007

How to Save Globalization From Its Cheerleaders

considerably away from a standardized Washington Consensus-style approach to a diagnostic strategy which focuses on each countrys own binding constraints (World Bank, 2005; Hausmann et al., 2005; Rodrik, 2007a; Ocampo and Vos, 2007). Differences in the nature of these constraints shape the appropriate economic strategies. For example, investment-constrained economies respond differently to capital inflows than saving-constrained economies, and need to have a different policy stance vis--vis the capital account. Moreover, as the examples of East Asian countries show, desirable policy reforms often take heterodox form because they try to make the best of pre-existing institutional capabilities and configurations. In China, non-standard policies such as dual-track pricing, township-and-village enterprises (TVEs), and special economic zones (SEZs) provided effective price incentives, some security of property rights, and outward orientation--but did so in highly unusual ways. Successful heterodoxy is a reflection of the need to overcome second-best complications. Trying to apply uniform best-practice rules or harmonizing policy differences away does not serve the needs developing and transitional economies. The need to maintain space for developmental policies is now recognized even by ardent supporters of free trade (Wolf, 2007). Some of the key areas where globalizations constraints bite for developing nations are the following:

The trade regime: WTO agreements on subsidies, trade-related

investment measures (TRIMs), and intellectual property rights (TRIPs) entail a considerable narrowing of space for the conduct of industrial policies, and preclude the adoption of strategies that worked well for growth superstars such as South Korea, Taiwan, and China. While determined governments can find ways around these restrictions, developed countries are demanding further tightening of restrictions in these and other areas. One reason developing countries such as India and Brazil have lost interest in Doha is that they are being pushed to significantly lower their own tariff bindings. Bilateral and

JITD, Fall 2007

23

Dani Rodrik

regional trade agreements, especially those negotiated with the U.S., often contain clauses on intellectual property rights and investment that go significantly beyond what is in the WTO.

The international financial regime: The promulgation of international

financial codes and standards follows a best-practice approach that is overly concerned with financial stability and the need to make economies resilient to capital flows, even when the promotion of such capital flows may not be a priority. These arrangements narrow the scope for traditional development banking and rule out credit market interventions (such as credit subsidies or directed credit) in support of industrialization. The preference of the IMF for central bank independence and free floating crowds out the use of the exchange rate as a developmental policy instrument. Currency undervaluation has been a potent tool for promoting growth in Asian and other countries (Rodrik, 2007b). The recent adoption by the IMF of enhanced surveillance guidelines on exchange-rate policies will make such strategies more difficult.

Once again, these are all areas where there are difficult trade-offs to consider. Currency undervaluation and subsidization policies may be good for developing nations, but they do impose political and economic costs on advanced countries, as I discussed previously. The point is not that there is an obvious right or wrong in each of these areas. My argument is that we need to recognize these frictions and focus our efforts on devising rules that can manage them, instead of proceeding with a market-opening agenda as if they were of little consequence. Here the lessons of history are invaluable. Economic historians agree that the earlier wave of globalization (1815-1913) collapsed because of the inability of the international system to cope with the tensions created by the expansion of global finance and trade. Harold James (2001) cites inherent instabilities in global finance, a growing social and political backlash, and the overloading of institutions that manage globalization as the leading contenders to explain the downfall of this earlier
24 JITD, Fall 2007

How to Save Globalization From Its Cheerleaders

globalization. These problems are remarkably similar to those we face today. They all have their origin in the tug-of-war between markets that are straining to become truly global and their mechanisms of governance which remain largely national and parochial. Jeffry Friedens (2006) account of the interwar period explains how the tug-of-war was eventually resolved:
The ensuing backlash [against globalization] had some predictable properties. Supporters of the classical order had argued that giving

new environment, some of those newly empowered responded that if the choice was between social reform and international economic integration, they would choose social reform thus leading to the Communists option of radical autarky. If the choice was between national assertion and global economic integration, another set of mass movements chose nation-building thus leading to fascist autarky in Europe and economic nationalism in the developing world. (Emphasis added.)

priority to international economic ties required downplaying such concerns as social reform, nation building, and national assertion. In the

In other words, national purpose reasserted itself in one form or another over the demands of international economic integration. Because the upholders of the international economic regime were blind to the conflict, the process was messy and the resulting outcomes (communism and fascism) were less than ideal. That is the real danger our globalization faces today. The risk is that the pursuit of a more perfect globalizationmore opennessendangers our imperfect, but still remarkable globalization by intensifying the conflicts that the system inevitably generates. VI. CAN POLICY SPACE BE ENHANCED WITHOUT DOING MORE DAMAGE THAN GOOD? The question posed in this sections title lies at the heart of the matter. The conventional view is that there is a slippery slope whereby even the slightest relaxation of international disciplines spawns further demands for protectionto the point that the system of free trade and finance
JITD, Fall 2007 25

Dani Rodrik

eventually unravels. This view sometimes finds expression in the bicycle theory of international trade, which states that maintaining an open economic regime requires constant efforts to liberalize. In this line of reasoning, policy space is a recipe for mischief. National polities cannot be trusted to work out reasonable internal compromises among competing domestic political forces with varying views on globalization. They need to have a straitjacket imposed from the outside. Yet there is little evidence that favors the slippery slope hypothesis in our contemporary political economy. The political balance in most countries, including developing countries, has tilted sharply in recent decades towards groups that favor links with the global economy. Notable departures from free trade, such as the Multi-Fiber Arrangement (MFA) and the voluntary export restrictions (VERs) of the 1980s, did not spawn increasing protection. In fact, they were removed once they had served their primary purpose of increasing the comfort level of rich country citizens. There are also some provisions of the GATT/WTO regime that are highly open to protectionist abuse, but these have had only limited impact on trade. The antidumping (AD) provisions of trade law are particularly notable in this respect, since they provide easy access to protection in circumstances where the economic case for protection is weak or non-existent. While countries do make use of AD provisions, it is hard to argue that the world economy has greatly suffered as a result. In retrospect, what is striking is not that AD provisions are used, but that they are used so infrequently in light of the flexibility of the rules, and that it has caused so little damage. Indeed, we could argue that AD strategies have made the trade regime more resilient by providing a safety valve for protectionist pressures. These pressures might have had more damaging consequences otherwise, if they had to make their way outside international rules rather than within them. That is precisely the principle behind the policy space approach: negotiated opt-outs, with internationally agreed procedural constraints,

26

JITD, Fall 2007

How to Save Globalization From Its Cheerleaders

are better than disorganized, unilateral opt-outs. It is better for the rules to recognize that sometimes countries need their own maneuvering room than to leave such a possibility outside the scope of the rules. An even better illustration of this principle at work is the WTOs Agreement on Safeguards (Art. XIX). The Safeguards Agreement allows countries to re-impose tariffs or quantitative restrictions under certain circumstances and for a limited time period, when countries experience a surge in imports of specific products and when such imports are determined to cause or threaten to cause serious injury to an industry at home. Aside from being temporary, the restrictions in question must be applied on a most-favored nation (MFN) basis, i.e., non-selectively, and affected exporters must receive compensation. While the principle behind Safeguards is clear, the restrictions currently placed on it make less sense. Why limit the application of the clause to instances of injury to producers, for example, or require that it be triggered only by a surge in imports? It could be that the injury in question is a conflict with deeplyheld values at home (say in the case of imports made using child labor or imposing environmental costs). Such injury could well be triggered by new information rather than by an import surge per seconsumers at home may discover unsavory facts about labor practices abroad or there may be new scientific information about safety or harm to the environment. In these instances, applying safeguard action on an MFN basis will not necessarily make sense (child labor may be a problem in Vietnam but not in Mauritius). Requiring compensation may not be appropriate either. In effect, the Agreement vests too many of the residual rights on trade interests and too few on the broader interests at stake. We can envisage broadening the Safeguards Agreement to a wider set of circumstances in which the legitimacy of trade is at issue, subject to institutional and procedural prerequisites that minimize the risk of protectionist capture (and in particular the empowering of interests who would be harmed by trade restrictions). We can also imagine a similar
JITD, Fall 2007 27

Dani Rodrik

development box provision, to facilitate the pursuit of developmental policies that may conflict with existing rules (e.g., subsidies). In effect, rich and poor nations would then be in the game of exchanging policy space instead of market access. Negotiators would be tasked not with maximizing the flow of trade and investment, but with designing rules that managed the interface among different regulatory environments. Such an approach was outlined in Rodrik (1997 and 2001) for the international trade regime, and I discuss it here to provide a concrete illustration of how these principles could be put in practice. A broadened safeguard agreementcall it an agreement on social and developmental safeguardswould enable countries to opt out from their international obligations under specified circumstances. The process for obtaining such an exemption would be a domestic one, as in the case of AD and safeguards currently, but it would be subject to multilateral review to ensure procedural requirements are met. Any interested party would be allowed to seek an exemption or opt-out. One requirement would be for the plaintiff to make a compelling case that the international economic transactions in question are in conflict with a widely shared social or developmental norm at home. For example, an NGO may try to make the case that goods imported using child labor violate domestic views about what is an acceptable economic transaction or a consumer body may want to ban imports of certain goods from a country because of safety concerns. A second procedural requirement would be to ensure that producer and other groups who have a stake in the international exchanges in question are able to present their case as well. In particular, the investigative body would be required to seek the views of those groups who would be adversely affected by the exemption. This is to ensure that the antiprotection views are given full hearing. One of the most important problems with AD and Safeguard proceedings at present is the lack of such a requirement. This prevents the full story from coming out and creates a protectionist bias in the system. When there is a truly

28

JITD, Fall 2007

How to Save Globalization From Its Cheerleaders

widely-held norm or principle at stake, it would be difficult for the protrade groups to mount an effective defense. It is hard to imagine that a business lobby representing importers would defend free trade when it involves, for example, slave labor or exceptionally harsh and exploitative working conditions. However, in other instances, there are real tradeoffs to consider, and a well-designed set of procedureswhether administrative or judicialwould help bring out the relevant considerations on all sides. Finally, the ultimate decision would rest with a semiautonomous government body that would consider the testimony given and determine (a) whether there is sufficiently broad support for the exercise of some kind of opt-out; and (b) what the best remedies are in cases where the answer to (a) is affirmative. The decision would be subject to periodic review to ensure that protection does not become permanent. It would also be open to review in a multilateral setting (say the WTO) to ensure that multilaterally-agreed procedural requirements have been met. A main advantage of the proposed scheme is that it forces deliberation and debate at the national level on the nature of the international economy, the economic gains it generates, and the circumstances under which domestic practices and needs come into conflict with it. This differs from the traditional, technocratic manner in which international governance is approached. It may seem overly messy and idealistic. But it has the virtue of bringing democracy to bear on these questions, and as such it has the potential to enhance the legitimacy of the global economy. VII. CONCLUDING REMARKS We can summarize the main arguments of this paper by using Figure 5, which underscores the mutual incompatibility of deep integration, national sovereignty, and democracy (Rodrik, 2000). I have argued that deep economic integrationa truly flat world economy to use Thomas
JITD, Fall 2007 29

Dani Rodrik

Friedmans evocative phraseis rendered infeasible by the fragmented nature of political sovereignty around the globe. Jurisdictional discontinuities impose transaction costs on international trade and finance that remain in place even when conventional barriers in the form of import duties and financial restrictions are removed. Of course, deep integration could still be attainable if national sovereigns were to restrict their actions only to those that are fully compatible with its requirements. This is the golden straitjacket option depicted in Figure 5. Its real-world counterpart was the classical gold standard era of the 19th century. This is a model that rules out democracy, since it requires that political authorities be unresponsive to national policy imperatives and domestic needs. It is not a coincidence that the gold standard collapsed following the expansion of mass franchise and spread of democracy in the major industrial powers. Facing the conflicting needs of employment creation and parity with gold, a democratic Britain made its choice in favor of the former and went off gold in 1931. We could also theoretically combine democracy with deep integration by eroding national sovereignty and carrying democratic politics to the global level. This is the global federalism model in Figure 5. It corresponds to the U.S. or EU model writ large, on a global scale. Needless to say, this outcome does not seem practical anytime soon. The only alternative we have left, therefore, is the Bretton Woods compromise, named after the golden era of 1950-1973 in which the world economy achieved unprecedented economic growth under a shallow model of economic integration. I have argued in this paper that our main challenge at the moment is to recreate this compromise, by designing a global architecture that is sensitive to the needs of countriesrich and poor alikefor policy space. This requires us to move away from a market-opening mindset, and to recognize that what nations need to do in order to maintain social peace and spur economic development in our second-best global economy often conflicts with the
30 JITD, Fall 2007

How to Save Globalization From Its Cheerleaders

free movement of goods, services and capital. The only way to save globalization is to not to push it too hard.
Figure 5. The Political Trilemma of the World Economy: Pick Two, Any Two

Deep economic integration Golden Straitjacket Global Federalism

Nation state Bretton Woods compromise

Democratic politics

JITD, Fall 2007

31

Dani Rodrik
REFERENCES James E. Anderson and Eric van Wincoop. 2004. Trade Costs, Journal of

Economic Literature 42(3): 691-751.

Anderson, Kym, Will Martin, and Dominique van der Mensbrugghe. 2005. Market and Welfare Implications of Doha Reform Scenarios, The World Bank, May 18. Blinder, Alan. 2006., Offshoring: The Next Industrial Revolution?, Foreign

Affairs, March/April.

Bradford, Scott C., Paul L. E. Grieco, and Gary Clyde Hufbauer. 2006. The Payoff to America from Globalisation, The World Economy 29(7 ): 893-916. Variety, Quarterly Journal of Economics 121(2): 541-585. Christian Broda and David. E. Weinstein. 2006. Globalization and the Gains from Christian Broda, Joshua Greenfield, and David E. Weinstein. 2006. From Groundnuts to Globalization: A Structural Estimate of Trade and Growth, NBER Working Paper 12512. Frieden, Jeffry. 2006. "Will Global Capitalism Fall Again?," Presentation for BRUEGEL's Essay and Lecture Series, Brussels, June. Hausmann, March. Hiscox, Michael J., and Nicholas F. B. Smyth. 2006. Is There Consumer Demand for Improved Labor Standards? Evidence from Field Experiments in Social Labeling, Department of Government, Harvard University. James, Harold. 2001. The End of Globalization: Lessons from the Great Depression, Harvard University Press, Cambridge: MA. Kose, M. Ayhan, Eswar Prasad, Kenneth Rogoff, and Shang-Jin Wei. 2006. Financial Globalization, A Reappraisal, International Monetary Fund Working Paper WP/06/189, August. Kose, M. Ayhan, Eswar S. Prasad, and Marco E. Terrones. 2007. , How Does Financial Globalization Affect Risk Sharing? Patterns and Channels, June. Krueger, Alan B. 1996. "Observations on International Labor Standards and Trade," NBER Working Paper W5632, June. Krugman, Paul. 2007. Trade and Inequality, Revisited, available at http://www.voxeu.org/index.php?q=node/261, June 15. 32 JITD, Fall 2007 Ricardo, Dani Rodrik, and Andres Velasco. 2005. Growth Diagnostics, John F. Kennedy School of Government, Harvard University,

How to Save Globalization From Its Cheerleaders

Maddison, Angus. 2001. The World Economy: A Millenial Perspective, OECD. Mayda, Anna Maria, and Dani Rodrik. 2005. "Why Are Some Individuals (and August. Ocampo, Jose Antonio, and Rob Vos. 2007. Divergences in Growth and Countries) More Protectionist than Others?," European Economic Review,

Development, Longman, forthcoming.

Prasad, Eswar, Raghuram Rajan, and Arvind Subramanian. 2007. Foreign Capital and Economic Growth, Brookings Papers on Economic Activity 1, March. Rodrik, Dani. 1997. Has Globalization Gone Too Far?, Institute of International Economics, Washington DC. Rodrik, Dani. 2000."How Far Will International Economic Integration Go?," Journal

of Economic Perspectives, Winter.

Rodrik, Dani. 2001. "The Global Governance of Trade as if Development Really Mattered," UNDP, New York. Rodrik, Dani. 2004. Industrial Policies for the Twenty-First Century, John F. Kennedy School of Government, Harvard University, September. Rodrik, Dani. 2007a. One Economics, Many Recipes: Globalization, Institutions,

and Economic Growth, Princeton University Press, forthcoming.

Rodrik, Dani. 2007b. The Real Exchange Rate and Economic Growth: Theory and Evidence, John F. Kennedy School of Government, Harvard University. Summers, Lawrence. 2006. Financial Times, October 29. Wolf, Martin. 2007. The growth of nations, Financial Times, July 21. World Bank. 2005. Economic Growth in the 1990s: Learning from a Decade of Reform, Washington, DC: World Bank.

JITD, Fall 2007

33