What Should the Fund's Role Be Now?

Remarks to the Bretton Woods Committee Annual Meeting

June 12, 2007

Steven Weisman, summarizing the views of pundits, recently reported, "The entire international financial architecture established after World War II-the World Bank, the International Monetary Fund, and what is now called the World Trade Organization-is buckling under the weight of globalization, trade disputes and the ambitions of rising economic powers in Asia and elsewhere."1 This statement was puzzling. One would think that the weight of globalization, trade disputes, shifts in economic power, as well as unprecedented global imbalances would imply that international institutions are all the more important to provide a framework and forum for managing the associated challenges.

In my remarks this morning, I first briefly review the role of the International Monetary Fund (IMF). Second, I touch on the IMF's lending activities, which absorb less than 25 percent of the IMF's administrative budget. Finally, I address four aspects of the Fund's nonlending activities where it could be doing a better job.

I strongly support the Bretton Woods Committee's constructive criticism and support for the Bretton Woods Institutions. The IMF is a central institution of global governance. I count myself as a long-time supporter of the Fund and its many accomplishments. The Fund's purpose is to promote and to sustain global growth and financial stability. These are public goods. On any objective basis, the excess demand for international monetary cooperation is substantial. As is the case for most public goods, those produced by the IMF are generally undersupplied. Recently the shortfall has increased. In my view, the Fund-members, executive board, management, and staff-has been falling down on the job.

If the public goods produced by the Fund are not supplied by global institutions, they will be supplied in some other manner, for example, regionally, as was recently advocated by Russian President Vladimir Putin. Is this likely? Would an increasingly integrated global economy and financial system function better with one institution of monetary cooperation rather than several, each playing by its own rules? Should we worry that some countries would be left out? My answer is "yes" to all three questions.

Nevertheless, the IMF's central role is under stress. It faces an existential crisis, an identity crisis, reduced demand for its lending services, and a lack of consensus about what its role should be. The institution's past and present contributions are badly misunderstood. Over the 60-plus years of its existence, the Fund has evolved constructively to deal with new problems that, in effect, have been assigned to the IMF because it was there. I would submit that this approach is more efficient than creating an additional institution and bureaucracy every time a new problem comes up, especially since many problems are not truly new. As the IMF and the global economy and financial system evolve, a central issue is the Fund's legitimacy and governance structure. They are in desperate need of reconstruction. However, that is not today's topic. Today's topic is, what should we expect from the Fund? What should the Fund be doing or doing better?

The recent Crockett Report on the Sustainable Long-Term Financing of the IMF provided a useful classification of the Fund's administrative expenditures by output.2 The four identified key output areas are (1) credit intermediation (for low-income members and other borrowers); (2) public goods (surveillance, research, financial-sector assessments, and statistical work); (3) bilateral services (technical assistance and training); and (4) governance (the Board of Governors, the executive directors, the managing director and his three deputies, the Independent Evaluation Office, and the secretary's department).

Out of an administrative budget of almost $930 million in FY2006, less that 25 percent was spent on credit intermediation-10 percent on special facilities for low-income members and 14 percent on generally available facilities. 3 Thus, if the Fund got out of the lending business, its expenditures would drop by only 25 percent, which is not a great deal even if one can argue plausibly that expenditures in other areas should be reduced at the same time.

One basic issue is whether the Fund should get out of the lending business. My answer is "no." IMF lending itself is a public good; its purpose is to mitigate the costs of crises and policy mistakes for the residents of the particular member country that is borrowing and for other members who would receive the adverse spillover effects if that particular member country had to adjust without the availability of external financial assistance. More than half the IMF's 185 members do not have meaningful access to international capital markets. For about half of the remaining countries, access is intermittent, in particular for countercyclical borrowing. Although healthy on balance, the increased scope and scale of private financial markets means that the potential adverse economic and financial effects of inherently volatile private capital flows have increased not declined.

Demand for IMF financing is minimal today because of a remarkable sustained period of benign economic and financial conditions, but such demands are cyclical, and all good things will come to an end. The only issue is the nature of the ending. This is the basis for a continuing, if reduced, need for an IMF lending capacity. It is also the basis for giving favorable consideration to the creation of a lending facility that would offer high-access short-term financing to countries that have records of good economic and financial policies and performance. And it is the basis for the IMF to stand ready to provide to its low-income members short-term loans to meet balance of payments needs, but in my view it does not provide the basis for long-term, subsidized (both in terms and administrative expenses) development lending to these members. It is useful to remember, as well, that initially IMF credit intermediation was differentiated by need not by the income level of the borrower.

The IMF's lending operations do not supply the overriding rationale for the Fund's existence as an institution for international monetary cooperation. That rationale lies in the almost 50 percent of the administrative budget directed at surveillance and related activities. The role of the IMF, aside from the provision of short-term financing, is to monitor developments affecting the health of the global economy and financial system, to articulate forcefully worries about those developments even if they have low probabilities attached to them, and at a minimum to prod members to take preemptive action.

It is not the Fund's role, in my view, to act as a cheerleader for globalization, which is how I interpreted First Deputy Managing Director John Lipsky's recent remark that the "accelerating pace of globalization" is responsible for the simultaneous global expansion of the past five years that has surprised the pundits.4 I think that the Fund should be spending more time worrying about the dark side of globalization in the form of financial-sector abuses (including the facilitation of tax evasion), the potential for economic and financial volatility, and how to deal substantively with criticisms that too many lose from globalization.

This is the topic for another set of remarks. Today, in the balance of my comments, I address four aspects of IMF nonlending activities where the Fund could enhance its performance: exchange rate policies, multilateral surveillance, financial stability and standards and codes, and the accumulation and management of reserves and other official holdings of cross-border assets.

All members of the IMF are obligated "to collaborate with the Fund and other members to assure orderly exchange arrangements and to promote a stable system of exchange rates." In addition, each member shall "avoid manipulating exchange rates or the international monetary system in order to prevent effective balance of payments adjustment or to gain unfair comparative advantage over other members." The Fund itself "shall oversee the international monetary system in order to ensure its effective operation, and shall oversee the compliance of each member with its obligations" and "shall exercise firm surveillance over the exchange rate policies of all members, and shall adopt specific principles for the guidance of all members with respect to those policies."

These are the familiar words from Article IV of the IMF Articles of Agreement as amended in 1978. The problem the Fund and its membership face today is that those words were a 1975 compromise between the US and the French positions on the exchange rate regime. This framework and the associated principles for the surveillance of exchange rate policies adopted in 1977 were never fully embraced by the membership of the Fund. One consequence, as detailed in the recent report of the Independent Evaluation Office (IEO), is that the Fund's management, staff, and executive board have been largely ignoring the obligations of members and are not using the 1977 principles in the Fund's surveillance activities.5 In effect, the management of the Fund has downplayed substantially its umpire or regulatory role with respect to the exchange rate policies of members. It has failed Mervyn King's test of "ruthless truth telling."6

Moreover, Managing Director Rodrigo de Rato's response to the IEO report, based on a more limited parallel internal review of exchange rate surveillance, was to argue that the IEO report "presents a narrow interpretation of the evidence, does not recognize progress made over time, and therefore does not offer a balanced perspective identifying remaining weaknesses and their relative importance." As outsiders, why should we trust an internal audit more than an external audit?

This is a serious challenge for the Fund. The management of the IMF has determined that it is playing its appropriate role with respect to its responsibilities under the IMF Articles of Agreement. I disagree. The Fund management's inaction on exchange rate policies risks serious adverse consequences for the international economic and financial system and international monetary cooperation going forward. The management of the Fund has declined to become prominently involved in the exchange rate policies of countries like China, Malaysia, and Japan, favoring a posture of trusted advisor rather than regulator. As a consequence, disputes about those countries' exchange rate policies have been redirected into bilateral, rather than multilateral, channels. The potential consequences are adverse for the system as a whole, for example in feeding protectionist sentiments affecting international flows of goods, services, and finance. The role of the Fund is to be a proactive participant not an interested bystander.

Since the end of 2002, when concerns about these issues began to surface, China's reserves and current account surplus have quadrupled to more than $1 trillion and more than 10 percent of GDP, respectively. Malaysia's reserves have more than doubled, and its surplus has also doubled to more than 15 percent of GDP. Since 2003, Japanese economic growth has averaged 2¼ percent a year, and the yen has weakened in real effective terms to below its level in June 1998, when concerted intervention was thought to be appropriate to support the yen even though the Japanese economy then was much weaker than it is today. Through all this the management of the Fund has not substantially altered its advice to these members.

Managing Director de Rato did finally initiate in April 2006 a multilateral consultation process. This is the IMF management's much ballyhooed effort to bring together representatives of China, the euro area, Japan, Saudi Arabia, and the United States to address global imbalances. The initiative was late but welcome. Its accomplishments have been limited because of excessive timidity, unsound analysis, or lack of cooperation by the participants.

A year later, the only positive result for the Fund was one of process: The Fund dealt itself into the policy coordination business essentially for the first time since the collapse of Bretton Woods. However, as far as one can tell, the management of the Fund exerted no pressure on the participants to make new or more specific policy commitments. The statements of policy intention were not new and not news; in some respects they were less explicit than those contained in the G-20 Accord for Sustained Growth issued in Melbourne, Australia in November 2006. They envisage a process of "immaculate adjustment," in other words adjustment without significant exchange rate changes. The only mention of exchange rates was by Saudi Arabia, which said that it would not alter its peg to the US dollar, and by China, which again said that its "exchange rate flexibility will gradually increase." For the United States, Japan, and the euro area, there was no mention of exchange rate adjustment. This is not Hamlet without the Prince, it is Hamlet set in never-never land.

It appears that the IMF's earlier warnings about the risks of external adjustment have been turned upside down. In place of concern that there will be too little in the way of preemptive policy actions, John Lipsky has declared, "excessively precipitous policy actions undertaken with the sole aim of immediate and substantial reductions in imbalances could be unnecessarily disruptive to global growth and could even undermine financial market stability."7 It is rare to hear a responsible official charged with promoting policy adjustment worrying about a decline in the US budget deficit that is too large, an appreciation of the Chinese yuan that overshoots, or economic reforms in the euro area or Japan that are too rapid. If this is the future of multilateral surveillance in the Fund, then my view is that the Fund is not discharging its responsibilities to promote global growth and financial stability.

One area where the Fund has been doing an impressive, if thankless, job is in the promotion of participation in its Special Data Dissemination Standard (SDDS) and of adherence to internationally agreed standards and codes to support financial stability. The SDDS now has 64 participants, principally the traditional industrial countries and emerging-market countries with, or seeking, access to international capital markets. In addition, a number of other countries mimic large parts of the SDDS in the preparation and release of statistical information, and 88 members participate in the General Data Dissemination System (GDDS), which is less demanding.

The Fund also reviews compliance with 12 international standards and codes ranging from data transparency to financial-sector standards, including standards for anti-money laundering and combating the financing of terrorism. Its reports on standards and codes (ROSCs), in principle, are integrated with the Fund's Financial Sector Assessment Program (FSAP), which it conducts in collaboration with the World Bank, and with the Fund's financial system stability assessments (FSSAs). The integration with the rest of the IMF's surveillance activities is less complete than it should be; this is one reason why the Fund recently underwent the third major reorganization this decade of its work in this area.

This is important work. The SDDS and the standards and codes have now been in place long enough that systematic statistical evidence has accumulated that participation in the SDDS and compliance with the standards and codes yields dividends to the country involved, for example, in the form of lower spreads on international borrowing and increased financial stability, which has positive spillover effects for the system as a whole.8 In the general area of transparency, Public Information Notices (PINs) have been published for 97 percent of members following their Article IV consultations. Ninety percent of members have consented to the publication of the underlying staff reports on those consultations. (The principal shortfalls are in developing Asia (79 percent) and the Middle East (71 percent).) Sixty-two percent of members have completed FSAPs, and 76 percent of the completed ROSC modules have been published.

The unfortunate reality is that this area of IMF activity is under the radar, receiving little credit. Moreover, not all members of the Fund are comfortable with the IMF's role; some want to scale it back on grounds of excessive expense or intrusion. The recent IEO report on IMF Exchange Rate Policy Advice notes a number of instances of noncompliance by members with their obligations in the statistical area. This is a serious problem, which IMF staff, management, and the executive board have tended to sweep under the rug.

The impetus for IMF engagement in these areas came from the external financial crises of the 1990s, starting with the Mexican crisis in 1995. Most of the standards and codes and statistical templates were developed by specialized bodies. However, standards and codes are useful only if they are implemented, and the task of monitoring implementation has been assigned by the international community to the IMF, in some cases in cooperation with the World Bank. Going forward, the issue is what role the Fund should play in developing and implementing new standards, for example, in the area of hedge funds or so-called private pools of capital, as well as in updating existing standards. Fund management should, in my view, be playing a proactive role. The IMF staff's recent criticism of the euro area's financial stability framework is an example of where the Fund is doing its job of "worrying" in situations where the probability of crisis is low but the expected value of the damage to the euro area's economy and the global financial system is significant.9

A fourth topic that is relevant to the Fund's role today is the rapid accumulation of official holdings of cross-border assets. The facts are familiar. Total official foreign exchange reserves now exceed $5 trillion. They have increased 150 percent since 2001. The nonindustrial countries account for 80 percent of the overall increase. In addition to official holdings of foreign exchange reserves, government-controlled cross-border financial assets have reached at least $1.5 trillion, from at most a few hundred billion US dollars five years ago; we don't know the precise figures because many countries are not transparent about their management of these assets. That is a crucial dimension of the problem. (The assets are in stabilization funds, sovereign wealth funds, government-owned holding companies, and a variety of other instrumentalities.) These trends suggest a potential tectonic shift in the structure of global capital flows away from traditional intermediation by the private sector toward a major role for governments in managing large portions of national wealth, in particular in the form of cross-border assets.

Foreign exchange reserves in Asia average 36 percent of GDP. Total Chinese holdings of international assets exceed 60 percent of GDP. For a developing country, this is a large fraction of its wealth. The comparable figure for the United States is 90 percent; excluding foreign direct investments the ratios are very similar. From a global perspective this probable misallocation of wealth is not the principal concern. The concern is about the management of these assets. On conservative estimate, at least 70 percent of China's cross-border assets are directly managed by the government. The comparable figure for the United States is less than 2 percent.10 Consequently, the management by the governments of China and a number of other countries of their external assets raises potentially serious political as well as financial issues.

Where have the Fund management and staff been as these developments have unfolded? Essentially they have been on the sidelines. In September 2003, the IMF's World Economic Outlook devoted 14 pages to the question "Are Foreign Exchange Reserves in Asia Too High?" The understated conclusion was "reserves in emerging economies in Asia are now at the point where some slowdown in the rate of accumulation is desirable from both domestic and multilateral [global adjustment] perspectives." Since that time Asian reserve holdings have continued to accumulate rapidly.

As detailed in the recent IEO report on the IMF's policy advice on exchange rates, the staff and management of the Fund have essentially ignored the topic of the implications of the build-up of reserves, first, for the stability of the economies with the large accumulations of foreign assets in official hands and, second, for the international monetary system.

The rapid accumulation of reserves distorts domestic monetary and financial policies leading to potential asset price bubbles and banking system problems down the road. At the same time, the global adjustment process is distorted by exchange rate pressures that are short circuited. Finally and most ominously in my view, it is highly likely that government management of huge stocks of reserves and other cross-border official assets will exacerbate political disputes and contribute to a further rise in financial protectionism.

The Fund's neglect of these issues stands in sharp contrast with the role of the IMF management and staff less that a decade ago when they strongly supported the initiative to include the Data Template for International Reserves and Liquidity (reserves template) in the SDDS. One consequence of that action is that at least 26 holders report at least annually on a voluntary basis, as suggested in the reserves template, the currency composition of their reserve holdings. This increase in the transparency should be applauded, including by the Fund.11 The Fund should be paying substantial attention to transparency and accountability of members in their management of their external assets.

In conclusion, today's increasingly integrated and complex global economy needs an effective Fund. To be more effective, the Fund needs to be reformed, starting with its governance. It continues to have an important role to play in short-term lending. However, more important is its role in providing other international public goods. I have highlighted four such areas. In all of them, the Fund should do more. In several, in my view, the Fund is falling down on the job. That strengthens the case for an IMF reform that goes beyond the modest agenda in Managing Director de Rato's medium-term strategy.


1. Steven R. Weisman, New YorkTimes, May 22, 2007.

2. International Monetary Fund, Final Report of the Committee to Study Sustainable Long-Term Financing of the IMF, January 31, 2007.

3. Ibid., table 3. The share of expenditures on public goods was 44 percent, on bilateral services was 23 percent, and on governance 9 percent.

4. John Lipsky, The Multilateral Approach to Global Imbalances, International Monetary Fund, May 31, 2007.

5. Independent Evaluation Office of the International Monetary Fund, An IOE Evaluation of IMF Exchange Rate Policy Advice, 1999-2005, International Monetary Fund, May 17, 2007.

6. Mervyn King, Reform of the International Monetary Fund, Speech at the Indian Council for Research on International Economic Relations, New Delhi, India, February 20, 2006.

7. John Lipsky, The Multilateral Approach to Global Imbalances, International Monetary Fund, May 31, 2007.

8. A useful summary of some of the positive results is John Cady, 2005, Does SDDS Subscription Reduce Borrowing Costs for Emerging Market Economies? IMF Staff Papers 52, no. 3: 503-17.

9. Concluding Statement of the IMF Mission on Euro-Area Policies, May 30, 2007.

10. The US estimate based on data on the US international investment position at the end of 2005.

11. Edwin M. Truman and Anna Wong (2006), in "The Case for an International Reserve Diversification Standard," Peterson Institute for International Economics Working Paper 06-2, advocate an expansion of this practice to include all countries with large foreign exchange holdings.