IMF reform: a marathon, not a sprint

Peter Kenen

30 October 2007



The current reform effort began in earnest 18 months ago, when the Managing Director issued a paper on the implementation of the Fund’s Medium-Term Strategy, in which he proposed numerous changes in the activities of the Fund. Here is a partial list of those changes:

  • Dealing with an impending financial problem confronting the Fund itself, because the recent fall-off of Fund lending has reduced its interest income;
  • Streamlining the Fund’s surveillance of its member countries;
  • Undertaking a new form of multilateral surveillance to confront systemic threats to the stability of the monetary system;
  • Establishing a new Fund facility to provide precautionary financing for countries that follow prudent policies;
  • Reforming the distribution of Fund quotas, which govern its members’ financial contributions to the Fund, their ability to draw on its resources, and their voting power, with the stated aim of recognising the increased economic importance of the emerging-market countries;

What has been achieved thus far with regard to these five matters?

The first item on my list has not yet led to visible reforms, but has led to a set of proposals by a committee appointed by the Managing Director. The committee’s report has proposed, inter alia, that the Fund sell up to 400 tons of gold and use the proceeds to establish an endowment, the income from which would finance some of the Fund’s activities. The staff of the Fund is examining this and other ways to deal with the budgetary problem, but has not yet come forward with its own proposals.

Turning from matters internal to the Fund to those involving relations with its members, it can be said that the Fund has taken important steps to reform bilateral surveillance. Heretofore, the Fund has sent staff missions to most of its members every year, and they have prepared compendious reports on the countries’ economic situations and policies. This practice has placed an enormous burden on the staff and Executive Board, which must read and discuss each report. Therefore, the Managing Director has proposed and the Board endorsed a streamlined approach. The Fund will no longer undertake annual surveillance of many member countries, especially small stable countries; it will do so biennially. It will also focus on matters of central concern to the Fund and pay greater attention to financial and balance-sheet vulnerabilities.

Most importantly, the Managing Director proposed that more emphasis be given to the main aim of surveillance—assessing the consistency of its members’ exchange-rate and macroeconomic policies with national and global stability. Thereafter, the Executive Board has revised its guidelines for surveillance for the first time in three decades. The new guidelines are broader and, importantly, include new language. Henceforth, a country may be deemed to violate its obligations under the Fund’s Articles of Agreement if it is deemed to maintain an undervalued exchange rate in order to increase net exports. (Not surprisingly, the Chinese Executive Director was alone in opposing the addition of this criterion.)

The surveillance of individual economies, appropriately limited in frequency and scope, are useful but cannot readily grapple with global problems, especially those involving the world’s largest countries, and existing multilateral forums such as the G-7 are not well-structured for this purpose. Therefore, the Managing Director proposed a new process of multilateral surveillance aimed at the mutual adjustment of the major countries’ policies.

The process proposed by the Managing Director would begin with bilateral consultations between the Fund’s staff and the countries involved, followed by multilateral meetings with those same countries. The staff of the Fund would then draft a report to be discussed by the Executive Board. The first round of multilateral surveillance involved China, the Euro Area, Japan, Saudi Arabia, and the United States, and it focused on global imbalances. But it did not have the intended effect – achieving agreement on policy changes aimed at reducing global imbalances. Instead, the key participants, China and the United States, reaffirmed commitments they had already made. China promised once again to “improve the exchange-rate formation mechanism in a gradual and controllable manner” with the aim of achieving a gradual increase in exchange-rate flexibility, but said nothing about the time it would take to achieve that objective. And the fiscal commitments of the United States were already contained in President Bush’s budget proposals for fiscal 2008 – proposals that have no chance of passage by the US Congress.

Multilateral surveillance can perhaps contribute to the resolution of global problems, but it is unlikely to produce modifications of national policies unless the staff of the Fund is empowered to put before the participating governments its own views on the requisite policy changes.

Countries normally seek financing from the Fund after they run into trouble and need then to negotiate the policy conditions that will be attached to that financing. The Managing Director has therefore proposed the creation of a new precautionary facility. It would be available to countries that have strong macroeconomic policies and sustainable debt burdens, but are still vulnerable to crises because of balance-sheet weaknesses and vulnerabilities. A country qualifying for access to this new facility would be free to make a very large drawing. Policy conditions might be attached to drawings on the new facility, but they would target policies aimed at maintaining macroeconomic stability and reducing the country’s vulnerabilities, rather than focusing in detail on a country’s immediate problem and the policies required to deal with it.

This appears to be an attractive proposal, especially for countries that fear contamination from a neighbour’s problems or from the effects of a tightening in global credit conditions. Nevertheless, it has one weakness, illustrated by a paper that worked with a simplified version of the plan.1 Compiling data for 34 countries from 1991 through 2002, the authors found that the countries’ ratios of debt to GDP and of budget deficits to GDP would have impaired the eligibility of 11 countries, temporarily or permanently, after they had qualified initially, and it would have precluded altogether the eligibility of 22 other countries. Chile was the only country to qualify continuously. As the data used by these authors would have been readily available to market participants, any deterioration in those data, foretelling a loss of eligibility, would be a very important matter; it could actually expose a country to a severe crisis. This risk and other concerns have led many to conclude that few if any countries would avail themselves of the new facility, and the proposal may be shelved.

We come now to the final item on my list, the question of quotas and voting power in the Fund. There has been progress on this issue, but the most likely outcome will not produce a dramatic change in the governance of the Fund. There would appear to be an emerging consensus on this issue, but one that will surely disappoint those who had hoped that reform would greatly enhance the influence of the developing countries. There will be a new formula for allocating increases in Fund quotas, in which gross national product will be the most important variable, but gross national product at purchasing-power parity will also play a role. There is then likely to be an increase of Fund quotas on the order of ten percent. But a ten percent increase of quotas will not much raise the voting power of the emerging-market and low-income countries. A back-of-the-envelope calculation will show why. At present, 12 countries have 59.4% of total votes in the Fund, leaving 40.6% to all the rest.2 Suppose that there is a ten percent increase of quotas and that the 12 countries with the largest quotas agreed to forgo completely any increase in their quotas, turning their shares of the increase to the rest of the Fund’s members. This would reduce the share of the votes of those 12 countries from 59.4% to 54.0%, and thus raise the share of the rest to 46.0%. Suppose instead that the 12 countries agreed to “disclaim” only half of the ten percent increase in their quotas; their share would fall by less, to 56.7%, and thus raise the share of the rest to 43.3%.3 Without knowing the parameters of the new quota formula, moreover, it is impossible to judge how much a ten percent increase of quotas would narrow the many large gaps between countries’ actual quotas and the new formula-based quotas. It is nevertheless clear that a single ten percent increase in total quotas would not alter substantially the distribution of voting power in the Fund. Several such increases would be needed to have that effect, and that would surely be opposed at a time when so few countries are drawing on the Fund.

This note has not touched on all of the issues involved in reform of the Fund. It has paid no attention to the proposal made by a committee of experts chaired by Pedro Malan -- that the Fund cease to make long-term subsidised loans to low-countries -- or to the proposal strongly endorsed by Tommaso Padoa-Schioppa, the Italian finance minister, that the European Union or, at least, the euro-zone countries should consolidate their representation in the Fund. Nevertheless, this note should suffice to demonstrate that reform of the Fund must continue, not just until next year’s meeting of the Fund’s Board of Governors, but for some time thereafter.

This column draws in part on the author’s monograph Reform of the International Monetary Fund, Council Special Report 29; Council on Foreign Relations, May 2007.




1 Tito Cordella and Eduardo Levy Yeyati, “A (New) Country Insurance Facility,” International Finance 9 (2006), pp. 1-36.
2 The 12 countries are Belgium, Canada, China, France, Germany, Italy, Japan, the Netherlands, Russia, Saudi Arabia, the United Kingdom, and the United States.
3 It should be noted that these calculations take no account of the proposed increase in so-called basic votes, which are distributed uniformly to Fund members, but even a doubling of basic votes would not much alter the outcomes in the text.





Topics:  Institutions and economics International finance

Tags:  IMF, reform

Walker Professor of Economics and International Finance Emeritus at Princeton University