WPS4668 Policy ReseaRch WoRking PaPeR 4668 Currency Undervaluation and Sovereign Wealth Funds: A New Role for the World Trade Organization Aaditya Mattoo Arvind Subramanian The World Bank Development Research Group Trade Team July 2008 Policy ReseaRch WoRking PaPeR 4668 Abstract Two aspects of global imbalances--undervalued two institutions, with the International Monetary Fund exchange rates and sovereign wealth funds--require providing the essential technical expertise in the World a multilateral response. For reasons of inadequate Trade Organization's enforcement process. There is a leverage and eroding legitimacy, the International bargain to be struck between countries with sovereign Monetary Fund has not been effective in dealing with wealth funds, which want secure and liberal access for undervalued exchange rates. This paper proposes new their capital, and capital-importing countries, which have rules in the World Trade Organization to discipline concerns about the objectives and operations of sovereign cases of significant undervaluation that are clearly wealth funds. The World Trade Organization is the attributable to government action. The rationale natural place to strike this bargain. Its General Agreement for WTO involvement is that there are large trade on Trade in Services, already covers investments by consequences of undervalued exchange rates, which act sovereign wealth funds, and other agreements offer as both import tariffs and export subsidies, and that the a precedent for designing disciplines for these funds. WTO's enforcement mechanism is credible and effective. Placing exchange rates and sovereign wealth funds on The World Trade Organization would not be involved the trade negotiating agenda may help revive the Doha in exchange rate management, and would not displace Round by rekindling the interest of a wide variety of the International Monetary Fund. Rather, the authors groups. suggest ways to harness the comparative advantage of the This paper--a product of the Trade Team, Development Research Group--is part of a larger effort in the division to understand the implications of international economic integration and international cooperation. Policy ResearchWorking Papers are also posted on the Web at http://econ.worldbank.org. The author may be contacted at Amattoo@worldbank. org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Currency Undervaluation and Sovereign Wealth Funds: A New Role for the World Trade Organization Aaditya Mattoo (World Bank) Arvind Subramanian (Peterson Institute for International Economics; Center for Global Development and Johns Hopkins University) JEL Codes: F21, F31, F32, F53 Keywords: exchange rates, undervaluation, sovereign wealth funds, World Trade Organization, International Monetary Fund, Authors' note: We are grateful to Montek Ahluwalia, Anders Aslund, C. Fred Bergsten, Surjit Bhalla, Nancy Birdsall, William Cline, Mac Destler, Kim Elliott, Josh Felman, Anwarul Hoda, Bernard Hoekman, Gary Hufbauer, Morris Goldstein, Simon Johnson, Jason Kearns, Vijay Kelkar, Will Martin, Petros Mavroidis, Rakesh Mohan, Scott Morris, Mike Mussa, Jonathan Ostry, Dani Rodrik, Mark Sobel, Bob Staiger, Alan Sykes, Ted Truman, John Williamson, Martin Wolf, and participants at seminars at the Peterson Institute for International Economics, Planning Commission, New Delhi, Reserve Bank of India, Bombay, and Stanford University for extremely helpful discussions and comments. The findings, interpretations, and conclusions of this paper are entirely those of the authors and should not be attributed to the World Bank. I. INTRODUCTION In the protracted Doha Round of multilateral trade negotiations of the World Trade Organization (WTO), countless negotiating hours have been expended on realizing outcomes in agriculture and manufacturing that would have minimal trade effects. Thus, in the market access negotiations on manufacturing, an optimistic outcome would yield actual tariff cuts by developing countries on average of about 2 to 3 percentage points and only slightly higher cuts by industrial countries (Martin 2007). In the agriculture negotiations, cuts in actual domestic support for the United States would be mostly notional and for the European Union not much greater than those that are already planned to deal with the fiscal consequences of Eastward expansion. And the reductions in applied agricultural tariffs will probably only be 3 to 4 percentage points on average once the likely exemptions for sensitive and special products are taken into account (Anderson, Martin, and van der Mensbrugghe 2006). In contrast, exchange rate changes--in particular persistent undervaluation of major currencies--these last few years have been substantial, leading to one of the most pressing contemporary problems: global imbalances. For example, estimates of China's exchange rate suggest a sizable undervaluation for the 2000­2007 period, ranging from 20 to 60 percent (Cline and Williamson 2007 and Chart 1). Eliminating this undervaluation is estimated to reduce China's current account surplus by between 6 and 12 percentage points of GDP, or between $150 billion and $300 billion (Goldstein and Lardy 2007). But undervaluation is not confined to China. Take the case of eight large oil exporters for which exchange rate data are available. Chart 2 depicts the evolution in their real exchange rates juxtaposed with real oil prices. Between end-2001 and July 2007, real oil prices increased by about 250 percent. During this period, five of the eight countries (the United Arab Emirates, Kuwait, Saudi Arabia, Bahrain, and Venezuela) have seen their real exchange rate depreciate by about 20 percent on average. The remaining three countries (Norway, Russia, and Iran) have seen appreciations, averaging about 24 percent. Exchange rates and exchange rate policies of oil-exporting countries are, of course, shaped by special factors. Even allowing for the differing circumstances of oil- exporting countries in terms of the dominance of the oil sector and the magnitude of proven reserves, however, real depreciations or modest appreciations in the face of a large, and plausibly permanent, positive price shock appear odd (Frankel 2006, Setser 2007). Jaewoo Lee, Gian-Maria Milesi-Ferreti, and Luca Ricci (2007) estimate that a 100 percent increase in real oil prices leads to a long-run real appreciation of the currencies of oil-exporting countries of about 50 percent. By that yardstick, oil-exporting countries' currencies should have appreciated in real terms by 125 percent. Of course, the short-run impacts are likely to be smaller. But even in the short run, the trade consequences need not be small. Under crude assumptions, we estimate that imports of oil-exporting countries could have been greater by about $200 billion or more. Thus, between China and the oil-exporting countries, moving toward more appropriate levels of exchange rates could have increased world trade by about half a trillion dollars. (Obviously, these are rough numbers, intended not as precise estimates but to convey a sense of orders of magnitude.)1 There is also a strong sense in the political arena that the trade consequences of exchange rate misalignments require remedial trade action. This sentiment underlies the numerous legislative initiatives in the US Congress, including the famous Schumer- Graham bill, which called for a countervailing subsidy against alleged Chinese undervaluation of the exchange rate, and the more recent Grassley-Baucus bill, which does not envisage trade retaliation but calls for a greater role for the International Monetary Fund (IMF) and US Treasury in assessing whether exchange rates are misaligned. In the European Union too, there is now a rising chorus of calls for antidumping and countervailing action against Chinese imports. And, increasingly, these concerns are being felt and articulated around the developing world, most recently by Indian Prime Minister Manmohan Singh.2 Another aspect of global imbalances is the accumulation of wealth in sovereign hands and the phenomenon of sovereign wealth funds (SWFs), which have raised a number of concerns. For example, Lawrence Summers worries about the twin dangers of "the pursuit of objectives other than maximizing risk-adjusted returns and the ability to use government status to increase returns."3 SWFs may not, at first blush, appear to have links with the multilateral trading system, but, as we show below, they do. In the face of such new and pressing problems and their large trade consequences, Doha's preoccupation with small and notional tariff cuts seems almost like an indulgence. It is like Hamlet not just without the Prince of Denmark but also without the most important characters. This paper elaborates how the WTO can become reengaged with, and provide a meaningful forum for addressing, these new issues and problems. In doing so, we argue that the WTO could also reenergize multilateral trade negotiations by expanding the bargaining space in a manner that offers opportunities for all groups of countries. Above all, addressing these new issues could rekindle serious private-sector interest in the WTO system, the absence of which has immobilized and ultimately derailed the Doha Round. In sum, this paper makes three contributions: the why and how of WTO involvement, in collaboration with the IMF, in dealing with undervalued exchange rates (section II); the why and how of WTO involvement in regulating SWFs (section III); and how placing these issues on the trade negotiating agenda could help galvanize the Doha Round (section IV). Section V offers some concluding remarks. 1 It is surprising that nearly all the recent concern about undervaluation has focused on China with very little on the oil-exporting countries. We suspect that part of this arises from the asymmetry of impact on trading partners. Chinese undervaluation manifests itself mostly as increased Chinese exports into the markets of trading partners, which is seen as a real, tangible threat. Oil exporters are not major suppliers of goods and services to other countries. Their undervaluation operates by reducing their imports, leading to a loss of export opportunities for trading partners. However, this loss of exports is not an actual loss but a loss of the counterfactual kind (higher than otherwise would be), which limits the sense of harm. 2 "India to Work with China to Remove Trade Barriers," Hindustan Times, January 14, 2008. 3 Lawrence Summers, "Funds that Shake Capitalist Logic," Financial Times, July 29, 2007. 2 II. THE WTO AND UNDERVALUED EXCHANGE RATES In this section, the argument proceeds in three steps: Why is there a role for the WTO in dealing with undervalued exchange rates? Why is the IMF, the natural forum for discussions on exchange rates, not effective on its own? And, finally, what should be the contents of new rules on exchange rates in the WTO? Why Should the WTO Address Undervaluation of Exchange Rates? There are compelling reasons for the WTO to address exchange rate undervaluation. The genius of the General Agreement on Tariffs and Trade (GATT) was to recognize that the politics of trade policy is unavoidably mercantilist and then to harness this very mercantilism to avoid protectionist outcomes. Two types of goods trade policies that the WTO regulates are import protection--through tariffs and quotas--and export support-- through subsidies. An undervalued exchange rate is both an import tax and an export subsidy and is hence the most mercantilist policy imaginable. Yet, exchange rate manipulation remains mostly unregulated in the WTO. The WTO rules on tariffs prohibit the taxation of imports above certain negotiated and legally bound levels. A country is, however, free to impose an implicit import tax by maintaining an undervalued exchange rate--in fact, such a measure is not even considered a tax. Similarly, the WTO rules on export subsidies exclude exchange rates from their scope because of the notion of specificity--i.e., policies that affect a few products are prohibited whereas subsidies that have economywide effects are not. This is like having disarmament negotiations where howitzers are haggled over while nuclear weapons remain beyond the scope of negotiations.4 Moreover, as noted earlier, undervalued exchange rates could have important negative trade consequences for partner countries, and are a classic example of beggar- thy-neighbor policies that both the IMF and WTO were set up to prevent. The implications for partner countries are straightforward if we consider separately the import tax and export subsidy aspects of undervalued exchange rates. A large country that imposes an import tax reduces the world price for its imports and hence worsens terms of trade for countries that export to it, while benefiting countries that import the same products. A large country that grants an export subsidy reduces the world price of its exports, hurting exporters of the same products but benefiting importers. Apart from these static effects, the induced changes in the structure of production in trading partners may have dynamic consequences. In particular, if the undervalued exchange rate causes a shift in resources in partner countries away from sectors (typically tradable goods) that have growth generating characteristics, then the result will be slower growth. To prevent such adverse spillovers was arguably the raison d'être of these institutions. Do current WTO rules already provide for redress against undervalued exchange rates? Potentially, recourse is possible to Article XV (4) of the GATT, which states that 4It is worth noting that in this regard, there is an asymmetry between WTO regulations on imports and exports. On the import side, GATT/WTO rules do not recognize the notion of specificity because across- the-board import measures such as surcharges are regulated and prohibited under Article II:1 (a) of the GATT. 3 "Contracting parties shall not, by exchange action, frustrate the intent of the provisions of [the WTO] Agreement ...." But this is too vague an obligation to provide a basis for effective enforcement (Hufbauer, Wong, and Sheth 2006). Indeed, there is no jurisprudence on this provision of the GATT, and it is highly unlikely that WTO dispute settlement panels would be willing to rule against undervalued exchange rates on this tenuous basis.5 Another possible avenue under current rules is for countries to initiate antidumping and/or countervailing action against products originating in countries with undervalued exchange rates. In our view, countervailing action would not be permissible because undervalued exchange rates cannot be deemed to be subsidies under current rules, which require subsidies to be "specific" to a product rather than broad-based as an undervalued exchange rate would be. Even if antidumping action were permissible, this instrument would be inadequate because it would not allow for effective action by countries that suffer from a loss of potential exports either to the country with the undervalued exchange rate or to third markets. Why Not the IMF? Before we go on to describe the substance of any new WTO rules, we have to take on the important and obvious argument against the WTO regulating exchange rates: namely, that exchange rates fall within the jurisdiction of the Fund. As Michael Mussa (2007) makes clear, the IMF has rules on exchange rates and exchange rate manipulation, creating obligations for individual countries and the membership. But Mussa (2007) makes equally clear that these rules have seldom been used: "...since Article IV was ratified 29 years ago, the number of consultations under its auspices has probably reached something between forty and fifty thousand. In most of these consultations, issues regarding the member's exchange rate and exchange rate policy have received scant attention. In none of these consultations has the Executive Board ever concluded that a member was out of compliance with its obligations regarding its exchange rate policies or any other matter" (emphasis in the original).6 Where exchange rates are overvalued, public determinations tend to be understandably avoided because they risk precipitating the very crisis that the Fund is entrusted with averting. Where pressure has been exerted, it has almost always taken the 5Hufbauer, Wong, and Sheth (2006) also make this point. In addition, they argue that the addenda to the interpretation of Article XV (4) make clear that another "specific GATT article needs to be frustrated in an important way before the strictures of Article XV (4) can be invoked." (p. 19). 6John Maynard Keynes himself recognized the asymmetry of Fund leverage between creditor and debtor countries during the discussions leading up to the creation of the Fund. As Timothy Adams notes: "Keynes worried about the potentially damaging effects of global current account imbalances and the fact that market forces were not very strong in compelling surplus countries to adjust. He believed that the IMF should have the ability to pressure surplus countries to play their part in resolving imbalances and developed what was then known as the "scarce currency clause." That clause [which would permit trade restrictions against surplus countries] faded into history as an unused relic of the Bretton Woods system of fixed exchange rate regimes, but the need for a lever on surplus countries remained. The asymmetric bias of the international monetary system was evident in the late 1960s ahead of the demise of the Bretton Woods system. It is also present today in discussion of the need for more flexible exchange rate regimes-- including in emerging Asia" ("Working with the IMF to Strengthen Exchange Rate Surveillance," remarks at the Cato Institute, February 2, 2006). 4 form of discrete, behind-the-scenes discussion between Fund staff and management, on the one hand, and the country authorities on the other.7 The IMF's role on exchange rates has come to the fore in the current case of China, evoking strong expressions of alleged negligence on the part of the IMF. Mussa's (2007) primary explanation for the Fund's reluctance to be more active on the Chinese exchange rate issue is weak leadership. But it seems to us that IMF ineffectiveness is a proximate manifestation of deeper structural causes related to leverage and legitimacy. It should be remembered that while the Fund has been able to effect changes in member-country policies in the context of financial arrangements, it has not been influential without the leverage of financing. In its key surveillance function (where no financing is involved), there have been relatively few instances where Fund intervention has led to changes in the policies of large creditor countries even when such policies have had significant spillover effects on others. In other words, the Fund has not been able to persuade large creditor countries to sacrifice domestic objectives for systemic ones. There seems to be an implicit "pact of mutual non-aggression," to use Mussa's phrase, in Fund surveillance. Perhaps, as a result, the Fund has had a history and tradition of nonadversarial dialogue between its members in a surveillance context and has not had to develop a real dispute settlement system. The Fund's attitude toward China could then be interpreted as merely a continuation of long- standing policy of being soft on large creditor countries, which China has recently become. Compounding this problem of limited leverage is the Fund's eroding legitimacy. The IMF's role has been diminished, and it has lost some of its status as a trusted interlocutor in emerging-market countries, and Asia in particular, in the aftermath of the Asian financial crisis.8 Fund staff and management have taken great pains to avoid antagonizing country authorities in part as a way of regaining some of this lost trust. There is also the more general perception, evident in discussions of governance reform, that the Fund's governance structure is outdated, reflecting the receded realities of Atlantic-centered 1945 rather than of ascendant-Asia 21st century. Emerging-market countries have expressed skepticism about the role of the Fund and its credibility. Indeed, if the Fund acts or had acted aggressively in the China case, it would lend itself open to the charge of hypocrisy, namely of applying tougher standards toward the newly 7See Goldstein (2006) for a lucid discussion of the IMF's role in exchange rate surveillance. 8Typical of this sentiment is the following editorial from the Business Standard, India: "... the Fund today is not playing its primary role as the lender of last resort to any high- or middle- income country (a category that now includes India), other than Turkey. The Fund had little influence over the rich economies even earlier; now it has little influence over the emerging market economies too, since they have decided (after the experience with the IMF in the 1997 Asian crisis) that they are better off building up their own foreign currency reserves. If the Fund is therefore reduced to a body that gives advice on good macro-economic management, and its opinion is valued most by those in charge of private international capital, then the developing countries are likely to view the body as an agent of such capital (a point that goes back to its handling of the 1997 crisis) and not one that has its member-countries' interests at heart. This could be tackled if the voting issue were to be resolved, because the Fund would then become a more representative body. But such change is blocked because the developed countries seek to use the Fund as a weapon against the rising economic powers" (Business Standard, October 24, 2007). 5 emerging creditors when it has for long been totally permissive when dealing with the industrial-country creditors. The WTO seems to be different on these two counts of legitimacy and leverage. As a political forum for negotiating and resolving conflict, the IMF is losing credibility while the WTO seems to be gaining it. In the Bretton Woods institutions, power was historically determined and has proved immutable, especially since the currently powerful have been reluctant to cede power. In the WTO, power and influence evolve organically because they flow from market size. As China, India, and Brazil have grown rapidly, they have naturally and without any help from other countries become serious players in the trading system. They have commanded power. They have not needed it to be conceded. Moreover, in a major development, the Uruguay Round's "single undertaking" approach, which equally burdened all countries, rich and poor, with obligations, by the same token empowered all countries with symmetric rights (Mattoo and Subramanian 2004). These advantages of the WTO in governance are reflected more specifically in the following. First, there is no legal mechanism in the Fund for affected members to initiate action against the offending party. The burden is entirely on the staff and management of the IMF, which is averse to taking sides--when many already doubt its impartiality. In contrast, the WTO has also developed a reputation amongst its members as a forum for relatively impartial dispute settlement. In our scheme, spelled out below, an affected member would be able to complain to the WTO, bringing into play a well-established, frequently used, and generally respected system of dispute settlement. In fact, disputes are the stuff of WTO proceedings, and it is not as unusual to bring a dispute in the WTO as it would be in the Fund. Moreover, bringing disputes in the WTO has become a much more two-way practice (in the sense of industrial and developing countries being both complainants and defendants). For example, the United States has been a complainant in 88 disputes and a defendant in 99 disputes. The analogous figures for India are 17 and 19; and for Brazil they are 23 and 14. Even China has initiated two disputes in the WTO and both against the United States. Not only are the large developing countries active complainants, so are small countries. Second, even if the IMF could be induced today to make a judgment, there is no mechanism to enforce its judgments. In our proposal, the standard WTO remedies-- multilaterally approved retaliatory action--could be applied if the rulings were not implemented.9 What would the WTO process add to the possibility of unilateral measures, which after all exist even today? Multilateral legitimacy and multilateral restraint. On the one hand, if action is today not being taken by countries for fear of seeming to play "judge and jury," WTO approval could lend to the threat of retaliation the legitimacy that it today lacks and hence also the credibility that may make such action unnecessary. On the other hand, if growing frustration did ultimately provoke unilateral 9Though past experience suggests that the mere prospect of retaliation, as well as the reluctance to be seen as a rule-breaker, are sufficient to ensure compliance, and there is rarely need for action. 6 action, from a systemic perspective such action would be inferior to, and more likely to escalate conflicts than, multilaterally approved sanctions.10 But one important concern arises from our advocacy of the WTO as an effective dispute settlement forum. This might be called the "too big to litigate problem." Since exchange rate undervaluation is potentially an issue of major macroeconomic significance, would it be possible to bring countries to dispute settlement on such an issue and reasonably expect rulings to be implemented? After all, it can be argued the WTO has not entirely been successful in relation to the big disputes: subsidies to Airbus and Boeing, import restrictions on beef containing hormones, and the EU system of preferences for its former colonies. Our reading of WTO dispute settlement is more nuanced. While some big disputes have not been resolved, others have been: for example, the US Foreign Sales Corporation tax, the Byrd amendment, US subsidies on cotton (a major aspect of US agricultural policy), European export subsidies on sugar, US safeguard action on steel, and European rules on geographical indications. Even small countries have had a modicum of success when litigating against their larger trading partners: Costa Rica successfully challenged US restrictions on underwear; and tiny Antigua brought a high- profile and successful case against the United States on online gambling--while the United States has not completely removed the offending action, it is modifying its commitments and compensating trading partners. What determines whether disputes can be successfully litigated and implemented is not necessarily whether they are big or important but whether countries perceive that these policies were part of a previous bargain that was considered fair and mutually beneficial. For our purposes, it is not necessary that WTO dispute settlement be perfect. It is enough if it is an improvement on virtually nonexistent enforcement in the IMF. Content of New WTO Rules If there are plausible push and pull reasons for the WTO to regulate exchange rates, the question is how this should be done. The current provision in Article XV that "Contracting parties shall not, by exchange action, frustrate the intent of the provisions of [the WTO] Agreement ...," is too vague an obligation to provide a basis for effective regulation. Indeed, there is no jurisprudence on this provision of the GATT. Any new rules should not be expedient and ad hoc, targeted at specific countries and tailored just to meet today's problems. Rather they should be such that their rationale and usefulness outlive current circumstances. In terms of content, procedures, and caveats, new rules could draw upon existing ones. First, any new rules would need to stipulate that undervalued or substantially undervalued exchange rates that stem clearly from government action act like import tariffs and export subsidies. The rule would therefore have two conditions: a clear finding of undervaluation and its demonstrable attribution to government action. 10Of course, one way out for the IMF would be to develop its own dispute settlement system. But given its history and the basic asymmetry between creditor and debtor countries, it is difficult to conceive of new IMF dispute settlement as a serious alternative to the WTO. 7 Once these two conditions are established, it would be as if GATT rules that prevent tariffs and other charges beyond previously specified ceilings and export subsidies in any form are violated.11 Is it possible to make pronouncements on the issue of exchange rate misalignment with a high level of confidence? The answer is probably not, but that could be a strength rather than a weakness because the WTO would regulate only egregious cases of misalignment--where the technical determination is relatively robust and criticism- proof--demonstrably caused by government action such as intervention. The high bar would act as a disincentive to frivolous litigation on this issue. Is it possible to attribute undervaluation to government action? Establishing this is important because most WTO obligations relate to policy instruments themselves, but undervalued exchange rates are an outcome rather than a policy instrument. Undervaluation can result from a number of factors, including fiscal and monetary policies, policies related to capital flows, taxes and subsidies, and intervention in foreign exchange markets. A finding that a country has an undervalued exchange rate would not be justiciable unless it translates into some clear remediable policy action that a country can take to change the outcome. In the case of undervalued exchange rates, there is a clear hierarchy of policy actions in terms of proximate causation (see Mussa 2007 for a clear discussion of this issue). Prolonged one-way intervention in foreign exchange markets by the central bank or by government and quasi-government agencies, redenomination of domestic debt into foreign currency, and extensive forward market operations are policy actions that can clearly be identified as causes of undervaluation. The more difficult cases will involve undervaluation caused by fiscal, monetary, or trade policies. Here Mussa's (2007) suggestions for a case-by-case determination seems the most pragmatic way forward. On the one hand, domestic objectives (full employment) will typically drive many of these policies, and the exchange rate consequences will be secondary. In these instances, countries should get the benefit of the doubt even in the event of a finding of undervaluation. But if it could be demonstrated that the mix of policies is clearly aimed at the external objective of gaining a competitive advantage, a country could then be asked to change its policy mix.12 Third, who should determine whether there is undervaluation and what its policy causes are? Recall that the IMF retains jurisdiction over exchange rates, and furthermore, technical expertise on exchange rates still resides with the IMF. Here again we draw on a precedent for joint Fund/WTO oversight over policy instruments. In the GATT, developing countries, for long, used trade restrictions for balance of payments (BOP) purposes. The assignment of responsibility in that instance was for the IMF to determine whether a country did in fact have a BOP problem, and then the WTO took over to appropriately regulate the restrictions. Indeed, two important disputes in the WTO-- 11Relevant WTO provisions are Article II: 1 (b) of the GATT and Article 3 of the Agreement on Subsidies and Countervailing Measures (SCM). The violation of Article II: 1 (b) would occur because undervalued exchange rates act like a charge on imports akin to the "other duties and charges" on imports, which this Article and the associated Understanding on the Interpretation of Article II: 1 (b) of the GATT currently prohibit. The violation of the SCM agreement would occur notwithstanding the requirement that subsidies be "specific" and that they involve a financial contribution. 12One such situation is where a country pursues a tight fiscal policy together with an easy monetary policy in circumstances where its fiscal situation is not a problem. 8 quantitative restrictions on beef imports by Korea (late 1980s) and on consumer goods by India (late 1990s)--involved such restrictions, which eventually had to be eliminated after dispute settlement panels found them to violate GATT/WTO rules. In the Korea beef case, the IMF determination was made in the context of the Fund's Balance-of- Payments Committee deliberations.13 But in the dispute involving India, the Fund's involvement resulted from the dispute settlement panel, in deference to Article XV requiring Fund input on these matters, posing specific questions to the Fund. Responses to these questions were treated as factual inputs, which the panel went on to interpret and use to adjudicate the case.14 We envisage a procedure very similar to Fund-WTO relations on restrictions for BOP reasons. Just as in these cases, where the Fund determines whether there is a BOP problem facing countries, it would be essential for WTO panels to seek the IMF's assessment on whether the member's exchange rate was misaligned and whether it was a consequence of clear government action. So rather than take on the political risks of what might be very one-off and controversial pronouncements, the IMF would respond to the WTO's request by making a more technical determination (based possibly on the results of the IMF research department's multilateral model (CGER) for determining equilibrium exchange rates). To be sure, this determination would have to be made and approved at some high level (either the Fund's management or the Board), but it could still be less controversial than the issue being raised within the Fund itself. Why might it be easier for the Fund to say things in a WTO context that it might be reticent about saying in the IMF? An example might help illustrate the answer. Take the current situation in relation to China's exchange rate. IMF staff reports have typically shied away, for political reasons, from saying whether and how much China's currency is undervalued. In a WTO dispute on the other hand, it would be forced to provide specific answers to the following questions that a panel could well ask: What is the estimate of undervaluation suggested by the CGER model? What is the preponderance of evidence from studies on China's exchange rate misalignment? Is the level of the exchange rate being maintained by intervention by the government of the central bank? These can be framed as purely technical questions to which the Fund will be required to respond. And these are questions to which Fund staff reports have generally avoided providing precise answers. Our proposal raises an important concern: Is it desirable to outlaw undervalued exchange rates in every situation? Increasingly, economic research suggests that the exchange rate can be a tool for economic development--at least, ruling it out as a tool would be difficult to justify on economic grounds (Prasad, Rajan, and Subramanian 2007; Rodrik 2007; Bhalla 2007). Equally, though, undervalued exchange rates should not be used to inflict negative spillovers on trading partners. Thus, exchange rates involve a trade-off--they may confer domestic growth benefits while at the same time inflict costs on other countries. The benefits are potentially greater for poor countries while the costs to trading partners are likely to be related to the size of a country's trade. It should therefore be 13See Republic of Korea--Restrictions on imports of beef--complaint of the United States (BISD 36S/268), available at www.wto.org. 14See India--Quantitative Restrictions on Imports of Agricultural, Textile and Industrial Products, available at www.wto.org. 9 possible to define--through negotiations--a threshold level of development and of country size beyond which the costs to trading partners could be thought to overwhelm domestic benefits. Indeed, there is a highly relevant precedent in the WTO itself. The WTO subsidies code does not prohibit the use of export subsidies for developing countries below a per capita income level of US$1,000 (Annex VII of the WTO Agreement on Subsidies and Countervailing Measures). But developing countries that have become competitive in a product--defined as having a world market share in that product exceeding 3.25 percent for two consecutive calendar years--have to phase out these subsidies (Articles 27.5 and 27.6). Similar income thresholds could be applied to exchange rates, although higher levels could be envisaged if countries feel they need greater policy space in regard to exchange rates. Thus, we argue that (1) exchange rates have serious trade consequences and unlike trade interventions, which are being phased out all over the world, episodes of undervaluation are likely to recur; (2) the Fund, the natural forum for regulating exchange rates, has abdicated its responsibility and is unlikely--for political reasons and its own traditions--to be able to remedy this; (3) the WTO could possibly fill this gap by creating new rules on exchange rates to parallel those on export subsidies and import taxes; and (4) these rules--as many others on trade--could become the subject of disputes in the WTO, with the Fund providing inputs on technical matters as it has in relation to trade restrictions for BOP reasons. Two points bear emphasis. First, our proposals do not involve the WTO getting involved in exchange rate management per se. If there were a distribution of exchange rate outcomes, WTO involvement would only arise in one tail of that distribution where the exchange rate is clearly undervalued. Second, our proposals would not entail the displacement of the IMF by the WTO: rather, it would harness the comparative advantage of the two institutions, with the WTO providing the valuable enforcement mechanism and the IMF providing the essential technical expertise to this enforcement process.15 The final question that needs to be addressed is this: Even if this approach makes sense, what could persuade countries in the WTO to make these changes? Or more starkly, why should China, which would clearly be at the receiving end of this rule in current circumstances, agree to codifying it in the WTO? The answer is that this rule would have to be part of a package of negotiations on other issues in which developing countries and especially China might be a demandeur. We discuss these below. III. THE WTO AND SOVEREIGN WEALTH FUNDS Twenty years ago, all over the world, the state began to retreat from direct ownership of firms, in traditional manufacturing sectors but especially in services from 15Another question that could arise is whether our proposal on exchange rates would entail reallocation of power and responsibility on exchange rate matters away from ministries of finance and/or central banks toward trade ministries? If so, this reallocation would meet with considerable resistance from the former. The fact that undervalued exchange rates could become an issue in the WTO does not entail any necessary change in responsibility within governments. Rather, it would imply a greater degree of involvement of finance ministries and central banks in the WTO. This happened, for example, in the WTO negotiations on the liberalization of trade and investment in financial services, when finance ministries started playing a greater role in the WTO. 10 telecommunications to transportation. This retreat of the state also coincided with an increased willingness to be less fearful of foreign ownership. As a result, flows of foreign direct investment (FDI) have increased sharply: For example, in the last two years, FDI has grown by above 25 percent annually (UNCTAD 2006). Public policy attempts to facilitate and codify these changes, especially at the international level, have, however, been less successful. In the late 1990s, the negotiations on a multilateral agreement on investment (MAI) were aborted. Attempts to facilitate FDI in services in the WTO have, however, led to some modest achievements in providing security of access to foreign investments. A distinctive feature of these flows, and perhaps one reason for the limited success of international cooperation in this area, however, was that they were largely in one direction: from industrial to developing countries. Imbued as international cooperation on access to markets is by mercantilist principles, successful cooperation involving FDI therefore required finding offsetting export benefits for developing countries, either in goods or in exports of labor. Today, that elusive equation--developing countries allow FDI in return for industrial countries easing restrictions on goods and labor--is being fundamentally transformed as a result of global imbalances or the uphill flow of capital. Today, capital flows to a greater extent in both directions. Developing countries are increasingly exporting capital, which creates intriguing and new possibilities for international cooperation in the WTO. These possibilities have been insufficiently explored. Chart 3 depicts an index of two-way flows (the Grubel-Lloyd measure), ranging from zero (complete one-way flow) and 1 (equal outflows and inflows). It shows that in the last few years, two-way flows have increased substantially, with the index close to 0.9 in 2006. In a further twist, this capital is being held in the developing world not in private hands but with the government in the form of foreign exchange reserves.16 First, in the oil-exporting countries, these reserves have built up largely due to high oil prices and the inability of these countries to absorb all the revenues concurrently (Chart 4). Moreover, these reserves are likely to increase: According to the IMF's most recent estimates, oil- exporting countries will accumulate about $300 billion to $400 billion of surpluses every year (this on top of large levels of stocks that are in the trillions of dollars).17 A second reason for capital buildup in public hands--a phenomenon most starkly associated with China--is exchange rate policy (Chart 1). For a variety of reasons--part insurance but larger part mercantilism--China and other East Asian countries have responded to current account surpluses and capital inflows with reserve accumulation by the central bank rather than allowing these surpluses both to be self-corrected and lodged in private hands through currency appreciation. As a result, China has accumulated $1.2 trillion of foreign exchange reserves. Other countries also have substantial holdings 16Martin Wolf, "We Are Living in a Brave New World of State Capitalism," Financial Times, October 17, 2007. 17This buildup of wealth and assets in the hands of oil-exporting countries is not new. Something similar happened in the aftermath of the oil shocks in the 1970s. But then, oil wealth was passively invested--i.e., they were recycled though industrial-country banks, which determined where the money was invested. Thirty years on, however, and with financial globalization and disintermediation having proceeded apace, wealth is now invested actively by its owners rather than passively through banks. Ironically, disintermediation has contributed to the problem of SWFs. 11 (Russia: $330 billion; Saudi Arabia: at least $300 billion; India: $250 billion; Singapore, Hong Kong, Brazil, and Malaysia all have at least $100 billion each). Some of these countries have set up SWFs to manage these reserves. Morgan Stanley has estimated on reasonable assumptions that there is now close to $2,500 billion in SWFs and that this figure will increase to $5,000 billion by 2010 and $12,000 billion by 2015. In turn, these surpluses are being disposed through acquisition of foreign assets-- government bonds and increasingly private-sector assets. But this acquisition is raising concern, even alarm, in the industrial world, which was traditionally on the other side of the capital equation. Summers gives the following examples: In early 2007, government-controlled Chinese entities took the largest external stake (albeit non-voting) in Blackstone, a big private equity group that, indirectly through its holdings, is one of the largest employers in the US. The government of Qatar is seeking to gain control of J. Sainsbury, one of Britain's largest supermarket chains. Gazprom, a Russian conglomerate, in effect controlled by the Kremlin, has strategic interests in the energy sectors of a number of countries and even a stake in Airbus. Entities controlled by the governments of China and Singapore are offering to take a substantial stake in Barclays, giving it more heft in its effort to pull off the world's largest banking merger, with ABN Amro.18 Why should anyone care about ownership by government-controlled entities? One set of concerns is not so much about SWFs per se as about foreign investment and is based on the supposed benefits of national ownership and control. A second more legitimate fear is related to national security and the suspicion that the purpose of the investment might be to secure control of strategically important industries for political rather than financial gain. And the magnitudes are significant: For example, in the United States, the total stock of FDI is $1.5 trillion, and market capitalization is about $25 trillion. These numbers are of the same orders of magnitude as the Morgan Stanley estimate of $12 trillion for SWFs in 2015. While the world is focusing on the concerns, there seems to be an opportunity here that could be seized. The new capital exporters want free and secure access to industrial-country assets to maximize the returns on their holding of capital while diversifying the attendant risks. But capital importers have legitimate concerns about the motivations and consequences of these transactions, especially since the wealth is owned and invested by governments and related entities. A mutually beneficial bargain is there for the taking. And the interesting thing is that there is a well-established legal precedent for regulating similar transactions in the WTO. No radical legal leaps are necessary. First, why do we need multilateral action in this area? From the perspective of countries with SWFs, the attractions of a multilateral approach are more obvious: They do not want to be subject to the whims of unilateral actions by receiving countries. But the real question is why recipient countries should forgo such unilateral action. For example, the United States is in the process of adopting legislation to tighten scrutiny of foreign investments by government entities where they raise security concerns (Jackson 2006). 19 Similarly, the European Commission is investigating whether takeovers by 18Lawrence Summers, "Funds that Shake Capitalist Logic," Financial Times, July 29, 2007. 19Section 5021 of the Omnibus Trade and Competitiveness Act of 1988 amended Section 12 publicly controlled foreign investment funds are a concern and need remedial action. The announcement came after German Chancellor Angela Merkel said that her government was considering setting up a system, similar to that in the United States, where a Committee on Foreign Investment can recommend that the US president block foreign direct investments that are deemed a threat to national security.20 Unilateral action has at least three problems. First, unilateral action could easily acquire a protectionist slant, especially if protectionists articulate their concerns in the language of national security as happened in the aborted acquisition effort by Dubai Ports World and in the case of the Chinese national oil company, China National Offshore Oil Corporation (CNOOC). Second, there could be proliferating and hence highly heterogeneous standards imposed by different capital-receiving governments, which could impose undue costs of compliance on SWFs and hence affect the efficient flow of capital. Third, even where unilateral legislation is enlightened and uniform and takes the form of stipulating reasonable restrictions on SWFs in return for secure access, there are likely to be difficulties in monitoring compliance with these restrictions unilaterally or even bilaterally. A multilateral agreement on SWFs would likely dominate unilateral action. The US Treasury has already asked the IMF to begin work on developing a code of conduct for SWFs. The IMF may be a convenient location for multilateral action on SWFs, but the real concern with SWFs is less the macroeconomic consequences of their activities than the microeconomic consequences of their being able to acquire corporate control. The latter can only be addressed in the context of rules on cross-border flows of direct investment. There are two reasons to believe that a natural home for such an agreement is the WTO. First, the WTO already, albeit somewhat opaquely, covers investments by SWFs in its services agreement--the General Agreement on Trade in Services (GATS). A second argument in favor of WTO regulation is its dispute settlement mechanism (as in the context of exchange rates). Consider a situation where a WTO member felt that a foreign SWF was behaving inconsistently with its obligations. Instead of taking unilateral action based on its own judgment--actions that can provoke retaliatory protection and spiral into a trade or investment war--the member would now have recourse to the WTO dispute settlement mechanism. The well-established mechanism would offer institutionalized consultation and, when necessary, impartial assessment of conformity with mutually agreed conditions. 721 of the Defense Production Act of 1950 to provide authority to the president to suspend or prohibit any foreign acquisition, merger, or takeover of a US corporation that is determined to threaten the national security of the United States. The president can exercise this authority under Section 721 (also known as the "Exon-Florio provision") to block a foreign acquisition of a US corporation only if he finds: (1) there is credible evidence that the foreign entity exercising control might take action that threatens national security and (2) the provisions of law, other than the International Emergency Economic Powers Act, do not provide adequate and appropriate authority to protect national security. 20EU External Trade Commissioner Peter Mandelson has suggested allowing EU governments to use "golden shares" to stop foreign governments from taking control of key industries. EU Internal Market Commissioner Charlie McCreevy was also reportedly looking into whether the European Union should develop a method of handling investments by third countries in EU companies (available at www.euractiv.com). 13 If multilateral action in the WTO is desirable, what should it look like? Two issues seem relevant: the scope and content of rules governing SWFs and the process by which these rules should be determined. Multilateral rules on SWFs would need to have a dual aspect: security of access for investors and a set of conditions that the SWFs would need to satisfy in order to benefit from access. Access for SWF Investors For countries receiving inflows from SWFs, concerns are probably greatest where controlling stakes are acquired in sensitive areas--communications, media, energy, and financial and distribution services--which mostly fall in the services sector. The WTO's GATS already contains a framework of rules governing FDI in services, and so it would seem natural to fold any future rulemaking into this process. Under the GATS, countries negotiate schedules of commitments, which are in effect legally binding promises to allow foreign investment in specific sectors subject to specific conditions. And, as table 1 shows, these schedules already contain restrictions on SWF-related activities. For example, some countries have explicitly excluded foreign government ownership from the scope of their GATS commitments in a few sectors. The United States' commitment states that government-owned or -controlled insurance companies, whether US or foreign, are not authorized to conduct business in a large number of states, and in basic telecommunications and radio or television broadcast services licenses may not be granted to or held by foreign governments or representatives thereof. 21In other words, these countries have retained the right to disallow SWF-controlled acquisitions in these sectors. Can it, therefore, be presumed that the existing commitments under the GATS on other services sectors already grant SWFs the right to invest except where explicitly excluded? Even if the purely legal answer to this question was yes, the problem is that countries could still invoke the GATS security exception to restrict investment by SWFs. So whether it is to clarify the scope of existing commitments, or to induce countries to eliminate existing restrictions, there is value in an agreement to explicitly include investment by SWFs within the scope of GATS commitments. One consequence of dealing with SWFs in the GATS framework is that the scope of the rules would govern only those cases where SWFs owned or acquired a controlling stake in a company.22 While some of the investments by SWFs will involve ownership or corporate control, and hence be covered by GATS, some will take the form of minority portfolio investments. A simple approach in the multilateral arena would be to accept this definition (difficult though it may be to implement in practice) and restrict the scope of future WTO regulation to transactions that clearly involve majority ownership or control. 21Mexico's commitment specifies that neither the concession provision of a public service nor the rights conferred thereby may be "transferred, mortgaged, encumbered or alienated to any foreign government or State, nor may the latter be admitted as a partner in the enterprise holding the concession." And "Foreign governments may not participate in an enterprise set up under Mexican law nor obtain authorization to provide telecommunications services." 22Moreover, given the definition of "government" control, the GATS framework would cover not just the new SWFs but also all government firms (for example, Dubai Ports World) and agencies (for example, state pension funds) making similar financial decisions. 14 Insofar as portfolio investments leading to minority ownership without effective control has raised fewer concerns, their exclusion need not be a significant omission. Moreover, the effective control criterion also serves as a natural delineation of responsibilities between the WTO and IMF, which could regulate aspects of SWFs other than where effective control is involved. Reciprocal Obligations on SWFs The negotiating quid pro quo for commitments to allow access to SWF investors would be the assumption of obligations by these investors and/or by governments controlling these SWFs. While commitments to allow access fit naturally into the GATS framework, the creation of obligations on SWFs would be based on two other precedents in the WTO for international cooperation to deal with state involvement in commerce. First, in Article XVII of the GATT, there are rules designed to ensure that state trading enterprises do not distort the pattern of international trade in goods. These rules oblige a country that creates a state trading enterprise to ensure that it acts solely in accordance with commercial considerations, does not discriminate in any way between or against trading partners, and is transparent in its operations. These principles could be directly applied to SWFs. Given the novelty of the terrain and the complicated nature of the issues, it seems that multilateral rules governing SWFs should be formulated broadly and avoid intrusive detail. Principles could be established relating to objectives and investment strategy, corporate governance, transparency, and behavior (Truman 2007). For example, if SWFs invested clearly through intermediary asset managers, as is the case with most institutional pools of capital such as endowments and pension funds, this could be prima facie evidence that the role of the government is indeed desirably minimal. More generally, a WTO member that established an SWF would be required to ensure that it was motivated solely by the objective of maximizing returns and did not make choices based on noncommercial grounds. Similarly, a model of transparency could be Norway's Government Pension Fund­Global. Clearly, even these principles would need to be fleshed out. And this effort could be undertaken in different ways, between different groupings, and with inputs from technically competent bodies and organizations. For example, the disclosure requirements of the Organization for Economic Cooperation and Development's principles of corporate governance could be the model for the transparency requirement. If these principles were not deemed sufficient to credibly establish an arm's length relation between the SWF management and their owner states, then rules governing specific actions of SWFs may also be needed. Thus, the second useful precedent is offered by the WTO's Government Procurement Agreement (GPA), which was a response to the concern that state purchases could be used for protectionist ends. In addition to the basic nondiscrimination obligation, the GPA requires a high level of transparency with respect to specific procurements and limits the scope for discretionary departures from commercial considerations. These rules could provide inspiration for disciplines on specific investments (or disinvestments) made by SWFs.23 23At first blush, imposing obligations on source-country entities might seem a legal innovation for the WTO since most rules cover the behavior of importing or host-country governments. However, regulations 15 One concern is that trade negotiators may not today have the capacity to establish norms on SWFs, and the WTO may not have the necessary expertise to adjudicate disputes pertaining to investments by SWFs. Fortunately, there is already a precedent in the services negotiations under the GATS of trade negotiators drawing upon expertise from different fields: for example, from sectoral regulators in the negotiations on telecommunications and finance and from national investment authorities. We suggest that the norms be negotiated with the involvement of all the relevant experts and regulators whom national governments will have to call upon as appropriate. Similarly on dispute settlement, WTO panels can draw upon external expertise from either institutions or individuals. For example, in the BOP cases, the panels sought the expertise of the IMF, and in intellectual property cases, external expertise from the World Intellectual Property Organization (WIPO) could be called upon. In the case of disputes relating to SWFs, it may be possible to turn to other fora such as the International Centre for Settlement of Investment Disputes (ICSID), which was established under the Convention on the Settlement of Investment Disputes between States and Nationals of Other States. Governments have given advance consent to submit investment disputes to ICSID arbitration in over 900 bilateral investment treaties.24 WTO members would, of course, need to decide how far to use the ICSID as a means of making a technical determination and how far to delegate to it some elements of the dispute settlement function. Finally, can negotiations on SWFs be decoupled from broader discussions and agreement on investment? It is true that comprehensiveness and legal cogency would dictate subsuming discussions on SWFs into such broader discussions. But as a practical matter, most of the concerns about SWFs really apply to investments in sectors that are considered sensitive--telecommunications, transportation, media etc. Most of these are in the services sector and hence covered by GATS. Thus, encompassing SWFs in the GATS should address most of the problems that might arise in this area. IV. REVIVING THE WTO: BROADENING SUPPORT BY EXPANDING BARGAINING POSSIBILITIES The Doha Round has always been afflicted by a private-sector interest deficit, which has been responsible for its modest ambition and desultory progress. The corporate demandeurs--the traditional protagonists--of the North have been conspicuous by their absence. All the focus on developing-country discontent with globalization and the WTO, and attempts to address it by making the Doha Round a "development" round, have obscured this fundamental problem. There are many reasons for this lack of private- sector interest (see Mattoo and Subramanian, 2005), including unilateral liberalization by developing countries and regional liberalization in the context of free trade agreements. on exporters (e.g., rules on export subsidies and quantitative restrictions on exports) suggest that this is not unprecedented in the WTO. 24Arbitration under the auspices of the ICSID is also one of the main mechanisms for the settlement of investment disputes under four recent multilateral trade and investment treaties: the North American Free Trade Agreement, the Energy Charter Treaty, the Cartagena Free Trade Agreement, and the Colonia Investment Protocol of Mercosur. 16 The heart of the problem seems to be the diminished bargaining possibilities between the different groups of countries (Bergsten 1998). The case of the European Union most clearly illustrates this problem. In purely mercantilist terms, it was not clear what the European Union stood to gain especially since Doha would have forced it to liberalize its sensitive agricultural sector. The offsetting gains in terms of access in manufacturing and services were not considered enough.25 If, however, exchange rates were added to the bargaining agenda, the incentives for governments and private sectors in the industrial countries could change appreciably. A number of actors--the European Union, a wide group of developing countries, private-sector interests in the United States and the European Union, and labor constituencies--can be counted on to be enthusiastic about placing the exchange rate on the trade agenda. For the European Union, in particular, which has seen the euro rise dramatically as the dollar has fallen, with no relief because the renminbi has also fallen with the dollar, addressing the exchange rate issue could be particularly attractive.26 There would be real money and market access at stake. One sign of the European Union's potential interest in the exchange rate is the threat of increased countervailing and antidumping action signaled by EU Commissioner for External Trade Peter Mandelson recently. For the United States, too, the prospect of addressing exchange rates multilaterally would be attractive to the private sector. Some of this attractiveness has diminished because of the decline in the dollar over the last year, and some parts of the American corporate sector, with large interests and investments in China, remain ambivalent about an aggressive approach toward the Chinese exchange rate. Nevertheless, on balance, private-sector interest can be expected to be significant. The ambivalence of capital relates to the fact that it is a mobile factor of production. Labor, however, is far less mobile. Hence labor constituencies in the United States and the European Union can be expected to be enthusiastic supporters of trade negotiations that include the exchange rate issue: For example, the AFL-CIO has always been a strong advocate of renminbi appreciation. An exchange rate initiative that could command the support of labor unions would make a Doha deal attractive even for a future Democratic president and Congress. Developing countries as a group should actively support multilateralizing the exchange rate issue. The fixed renminbi has affected nearly all emerging-market countries. For example, a number of emerging-market currencies--the Brazilian real, the Korean won, the South African rand, and the Indian rupee, among many others--have appreciated significantly in effective terms. Had the renminbi also appreciated commensurately, the loss of relative competitiveness and export growth would have been significantly reduced. In the case of India, for example, the elimination of Chinese undervaluation would amount to about a 10 percent effective depreciation of the rupee. Even manufactured exports from and industrialization in Sub-Saharan Africa may be a 25The short-lived "Singapore issues"--competition, trade facilitation, government procurement, and investment--represented an attempt to make the trade agenda attractive for the European Union. But these issues never gained any traction domestically within the European Union because it was not clear which private-sector interests an agreement would significantly benefit. 26C. Fred Bergsten, "Europe Must Look East to Deal with the Euro," Financial Times, October 11, 2007. 17 victim of Chinese undervaluation, which may have accentuated the "Dutch disease" effects of higher commodity prices. A key question that needs to be addressed is: What's in all of this for China? In other words, how can China be made to come to the negotiating table with exchange rates high on the agenda? There are likely to be both sticks and carrots. It may be easier for the wide range of affected countries to bring pressure upon China by forging a collective response. It will be easier for countries to come together in the WTO context because the Chinese exchange rate inflicts clear and identifiable harm on trade interests and as it happens in this case on trade interests in a large part of the world, developed and developing. It is difficult to envisage such coalitions amongst finance officials and bureaucracies, which makes action in the IMF less likely. In other words, the line-up against China (the United States, the European Union, and all the important developing countries) will be more solid and greater in a trade forum. China cannot easily dismiss this. But it is true that China is mighty, and sticks on their own will not work, so we need carrots. Carrots could take the form of securing investment opportunities for its SWF in an environment where Chinese investments could increasingly be subject to national regulations with a protectionist slant. Clear rules on SWF-related investments could thus be one of the inducements for China to cooperate. If this combination of carrots and sticks is insufficient, consideration could also be given to linking a more market-based exchange rate to eventually granting China the status of a market economy, which would make it less vulnerable to arbitrary unilateral action--especially antidumping duties--by its trading partners (Messerlin 2004). It is not certain that this combination of carrots and sticks will adequately induce China to cooperate, but the significant incentive for China to cooperate can be provided only in a WTO context rather than in the IMF. The combination of exchange rates and SWFs also offers a unique opportunity for oil-exporting countries, including Saudi Arabia, that have for long been peripheral players in the WTO to become more active and engaged in the multilateral trading system. The broader question is how these two new issues we have identified should become part of the new agenda for the WTO. If the Doha negotiations are concluded in their current form, WTO members would need to start quickly on a new agenda. On the other hand, if Doha languishes because of the lack of trade promotion authority in the United States and the unsettled broader climate for trade liberalization, an alternative approach would be for the WTO to receive a new negotiating mandate that would include these issues in addition to the existing ones. In either case, for negotiations to proceed expeditiously, they would have to be amongst a select few with the major stake both in relation to exchange rates and SWFs. In other words, negotiations should not be held up by the need to establish consensus amongst the full WTO membership, most of whom will only have a peripheral interest in the issues. Of course, the benefits of any disciplines negotiated, such as transparency of SWF operations, would extend even to those who do not actively participate in the negotiations. For example, African countries, which are increasingly a destination for investments by SWFs, stand to gain from disciplines on SWFs that they may not have been able to negotiate on their own. 18 Even if our proposals make systemic sense, the process of negotiating new rules would take too long for it to be effectively applied to remedying China's current undervaluation. Our response is two-fold. While there could indeed be considerable time lags before new rules are negotiated, duration is directly related to political will. If the international community is willing to act expeditiously--and given the broad consensus that possibility should not be ruled out--changes can be effected quickly. Further, we contend that the very fact of mobilizing to act in the WTO could add to the other pressures--internal and external--on China to change its exchange rate policy. Before we conclude, we need to address one last concern that might arise from our proposals, namely, that there is a danger of "overloading the WTO." If exchange rates and SWFs are included, some may ask, why not add labor and environmental standards also to the WTO? "Overload" is a legitimate concern, and keeping the focus on trade is necessary to keep the WTO manageable and effective. But exchange rate policy has such direct and large trade consequences that including it in the WTO would seem essential for a trade-focused organization. Including exchange rate policy would be very different from including labor and environmental standards, which would really amount to using trade as a means to advance nontrade (social) objectives. Moreover, the existing Doha agenda does not have enough enthusiastic backers; adding exchange rate issues broadens the base of support for the round. V. CONCLUDING REMARKS This paper has argued that two aspects of global imbalances--undervalued exchange rates and SWFs--require a multilateral response. For reasons of inadequate leverage and eroding legitimacy, the IMF has not been effective on its own in addressing undervalued exchange rates. In contrast, the WTO is more credible and effective in resolving disputes, and moreover, rules on undervalued exchange rates are necessary complements to existing rules on tariffs and export subsidies. As it turns out, our proposal rehabilitates the "scarce currency clause," suggested for the IMF by John Maynard Keynes more than 60 years ago, which failed because it was the right idea for the wrong institution. Our proposal harks back to Keynes' idea but with the difference that it delineates afresh the responsibility of the two institutions consistent with their comparative advantage: The IMF would continue to play a technical role in assessing when a country's exchange rate was undervalued, and the WTO would assume the enforcement role. Our proposals on undervalued exchange rates clearly need refinement. But any specific limitations in them should not deflect attention from three central points: that undervaluation is a serious systemic problem; that in the status quo there is a complete absence of effective remedies against it; and that harnessing the WTO's enforcement mechanism affords the prospect of providing some redress, at least for egregious cases of undervaluation clearly attributable to government action. On SWFs, there is a natural bargain to be struck between countries with SWFs, which want secure and liberal access for their capital, and capital-importing countries that have legitimate concerns about the objectives and operations of SWFs. The WTO is the natural place to strike this bargain, not least because its services agreement, GATS, already covers investments by SWFs. Here too, a clear delineation of responsibility 19 between the IMF and WTO is possible, with the WTO involved in those cases where SWF investments are like FDI, entailing effective corporate control, and the IMF in all other cases. Our proposals may have a collateral benefit. The WTO has been treading water since the conclusion of the Uruguay Round and has suffered a string of aborted efforts to advance the liberalization agenda: Seattle, Cancun, and Potsdam are now as much metaphors for trade failures as names of places. The launching of the Doha Round was an aberration, made possible by the fragile, and as it has subsequently proven, temporary, global solidarity engendered by 9/11. A deeper reason for minimal progress in the Doha Round is the lack of private-sector interest because it is not seen to be addressing issues of real global significance. "Fail again, fail better," exhorts a Samuel Beckett character. Multilateral trade negotiations have been failing with such metronomic regularity that it is not clear that it is even failing better. By addressing global imbalances, the WTO could reclaim relevance and reignite the interest of a number of actors--the private sector, governments, and labor interests. While more work needs to be done in fleshing out the proposals made in this paper, they offer the WTO system the opportunity not just to fail better but perhaps even to succeed. 20 REFERENCES Anderson, Kym, Will Martin, and Dominique van der Mensbrugghe. 2006. Distortions to World Trade: Impacts on Agricultural Markets and Farm Incomes. Review of Agricultural Economics 28: 168­94. Bergsten, Fred, C. 1998. Fifty Years of the GATT/WTO: Lessons from the Past for Strategies for the Future. Policy Briefs in International Economics 98-3. Washington: Institute for International Economics. Bhalla, Surjit. 2007.Second Among Equals: The Middle Class Kingdoms of India and China. Draft manuscript. Peterson Institute of International Economics, Washington. Cline, William, R., and John Williamson. 2007. Estimates of the Equilibrium Exchange Rate of the Renminbi: Is There a Consensus and, If Not, Why Not? Paper presented at the conference on Debating China's Exchange Rate Policy, October 19, Peterson Institute for International Economics, Washington. Goldstein, Morris. 2006. The IMF as Global Umpire for Exchange Rate Policies. In C. Fred Bergsten and the World Economy, ed. Michael Mussa. Washington: Peterson Institute for International Economics. Goldstein, Morris, and Nicholas Lardy. 2007. China's Exchange Rate Policies: An Overview of Key Issues. Paper presented at the conference on Debating China's Exchange Rate Policy, October 19, Peterson Institute for International Economics, Washington. Frankel, Jeffrey. 2006. Commodity Prices and Monetary Policy. In Asset Prices and Monetary Policy, ed. John Campbell. Chicago: University of Chicago Press (forthcoming). Hufbauer, Gary, Yee Wong, and Ketki Sheth. 2006. US-China Trade Disputes: Rising Tide, Rising Stakes. Policy Analyses in International Economics 78. Washington: Institute for International Economics. Jackson, A. K. 2006. The Exon-Florio Test for Foreign Investment. CRS Report for Congress. Order code RL33312. Washington (March). Lee, Jaewoo, Gian-Maria Milesi-Ferreti, and Luca Ricci. 2007. Methodology of CGER Exchange Rate Assessments. Paper presented at the workshop on global imbalances, sponsored by Brussels-based BRUEGEL, Seoul-based Korea Institute for International Economic Policy, and Peterson Institute for International Economics, February 8­9, Washington. Martin, Will. 2007. Draft Modalities for Non-Agricultural Market Access: A Quick Assessment. World Bank, Washington. Photocopy. 21 Mattoo, Aaditya, and Arvind Subramanian. 2004. The WTO and the Poorest Countries: The Stark Reality. World Trade Review 3: 385­407. Mattoo, Aaditya, and Arvind Subramanian. 2005. Why Prospects for Trade Talks Are Not Bright. Finance and Development 42: 19­21. Messerlin, Patrick. 2004. China in the WTO: Antidumping and Safeguards. In China and the WTO: Accession, Policy Reform, and Poverty Reduction Strategies, eds. Deepak Bhattasali, Shantong Li, and Will Martin. Washington: World Bank and Oxford University Press. Mussa, Michael. 2007. IMF Surveillance over China's Exchange Rate Policy. Paper presented at the conference on Debating China's Exchange Rate Policy, October 19, Peterson Institute for International Economics, Washington. Prasad, Eswar, Raghuram Rajan, and Arvind Subramanian. 2007. Foreign Capital and Economic Development. Brookings Papers on Economic Activity 1 (March). Rodrik, Dani. 2007. Exchange Rates and Development. Harvard University. Photocopy. Setser, Brad. 2007. The Case for Exchange Rate Flexibility in Oil-Exporting Economies. Policy Briefs in International Economics 07-8. Washington: Peterson Institute for International Economics. Truman, Edwin, M. 2007. Sovereign Wealth Funds: The Need for Greater Transparency and Accountability. Policy Briefs in International Economics 07-6. Washington: Peterson Institute for International Economics. UNCTAD (United Nations Conference on Trade and Development). 2006. World Investment Report. Geneva. 22 Chart 1: Chinese Undervaluation and Current Account, 2000-2007 70 18 16 60 tnec 14 per 50 nir 12 la dol 40 to 10 vetialer PDGfo 30 8 % on uatilav 6 20 nderU 4 Average estimates of undervaluation (left axis) 10 Current account in % of GDP (right axis) 2 0 0 2000 2001 2002 2003 2004 2005 2006 2007* Source: Cline and Williamson, 2007, Table 1 For each year, the average of the estimates across studies is computed. Number of studies per year varies from 1 (2000) to 7 (2007) *Current account figure for 2007 is an extrapolation 23 Chart 2: Real Oil Price and Real Effective Exchange Rates (REER), January 2000-July 2007 (Decline in the REER index denotes depreciation) 350 Saudi Arabia Iran Bahrain Venezuela Kuwait UAE Russia Norway 300 oil price (right axis) 00)1 250 2001= ber 200 ecemD( ex Ind 150 100 50 Jan-00Apr-00Jul-00Oct-00Jan-01 03 05 06 t-06 07 Apr-01Jul-01Oct-01Jan-02Apr-02Jul-02Oct-02Jan- Apr-03Jul-03Oct-03Jan-04Apr-04Jul-04Oct-04Jan- Apr-05Jul-05Oct-05Jan- Apr-06Jul-06Oc Jan- Apr-07Jul-07 24 Chart 3: Emerging Markets Gross Private Capital Flows 2002-2006 1200 0.95 Gross inflows, US$ 1000 0.9 Grubel-Lloyd measure of bn.(left scale) two-way flows, (right 0.85 800 scale) 0.8 600 0.75 400 0.7 200 Gross outflows, US$ bn. (left scale) 0.65 0 0.6 2002 2003 2004 2005 2006 25 Chart 4: Oil Exporting Countries: Price and Reserves, 1999-2006 1000 70 900 60 800 Real oil price, 2006US$ (right scale) 700 50 600 40 500 400 30 Foreign exchange 300 reserves, US$billion, 20 200 (left scale) 10 100 0 0 1999 2001 2002 2003 2004 2005 2006 26 Table 1. Examples of restrictions on foreign government ownership in the GATS Schedules of the US, EU and Mexico Countries Sector GATS Commitment and Offer USA All types of Insurance services including reinsurance Government-owned or government-controlled insurance companies, whether US or foreign, are not authorized to conduct business in: Alabama, Alaska, Arkansas, California, Colorado, Connecticut, Delaware, Georgia, Hawaii, Idaho, Kansas, Kentucky, Maine, Maryland, Montana, Nevada, New Jersey (only with respect to surplus lines), New York (non-life companies are authorized; life and health companies are not), North Carolina, North Dakota, Oklahoma, Oregon, Rhode Island, South Dakota, Tennessee, Washington, Wyoming. Basic Telecommunications services Ownership of a common carrier radio license: Indirect: None Direct: May not be granted to or held by (a) foreign government or the representative thereof... (d) U.S. corporation of which more than 20% of the capital stock is owned or voted by a foreign government or its representative... Other Communication Services --Cable services provided over cable Radio and television broadcast licenses may not be held by: a foreign government;... a corporation systems chartered under the laws of the United States that is directly or indirectly controlled by a corporation --One-way satellite transmission of more than 25 per cent of whose capital stock is owned by non-US citizens or a foreign government DTH and DBS television services ... and of digital audio services. -- Programme Transmission Services. --Television Broadcast Transmission Services ) -- Radio Broadcast Transmission Services --Radio and Television combined program making and broadcasting services. EU-Spain Horizontal commitment: Applies to all Investment in Spain by foreign government and foreign public entities (which tends to imply, sectors in the schedule. besides economic, also non-economic interests to entity's part), directly or through companies or other entities controlled directly or indirectly by foreign governments, need prior authorization by the government. Mexico Horizontal Commitment 27 Mexicans have priority over foreigners for concession-granting purposes, other things being equal. A concession represents discretional authorization by the competent authorities for the provision of a public service. Neither the concession nor the rights conferred thereby may be transferred, mortgaged, encumbered or alienated to any foreign government or State, nor may the latter be admitted as a partner in the enterprise holding the concession. Telecommunications services ...Foreign governments may not participate in an enterprise set up under Mexican law nor obtain authorization to provide telecommunications services. All insurance and insurance-related services: Foreign investment by governments, and official agencies or legal persons exercising governmental functions is not allowed. -Life -Non-life -Reinsurance -Services auxiliary to insurance Banking Foreign investment by governments, and official agencies or legal persons exercising governmental functions is not allowed. Air Transport Services: Mexicans have priority over foreigners for concession-granting purposes, other things being - Supporting services for air transport equal. A concession represents discretional authorization by the competent authorities for the provision of a public service. Neither the concession nor the rights conferred thereby may be transferred, mortgaged, encumbered or alienated to any foreign government or State, nor may the latter be admitted as a partner in the enterprise holding the concession. 28