IMF Governance Reform: A "Pretty Good" Step in the Right Direction
Two weeks ago in Gyeongju, Korea, the G-20 finance ministers and central bank governors agreed on the outlines of a shift in the shares (votes) and chairs (representation) in the International Monetary Fund (IMF). On November 5, the IMF executive board fleshed out the agreement in time for the G-20 leaders to bless it when they meet in Seoul on November 11–12.
Do these decisions amount to a historic reform of IMF governance? No, but they are a significant evolution. Once implemented, the decisions will move the Fund a long way from where it was when I wrote about this topic in 2005.1
That the IMF governance package is not revolutionary reflects the challenge of reforming an institution from within when major decisions require 85 percent majority votes. Change can be blocked not only by one country (the United States) but by many other small groups of countries (the European Union, Brazil, Russia, India, and China (BRICs) plus a few similarly minded countries).2 In addition, almost all elements in the package involve zero-sum situations in which some countries are perceived as gaining at the expense of others even if all countries in the end are better off. A reasonably balanced package is required along with a convincing appeal to the collective interest that brings most countries along.
Some will argue that the proposed changes should be blocked because they are insufficiently bold and comprehensive. My view is that they deserve support and should be recognized for what they are: a significant step forward in IMF governance.
My assessment below covers five aspects.
- Size of the Fund
- Distribution of shares
- Distribution of chairs
- Other issues
- The outlook for implementation
Size of the Fund
The IMF governance agreement doubles the total size of IMF members' quotas in the Fund from about $375 billion to $750 billion. IMF quotas set each member's voting power in the Fund, establish the member's commitment to provide financial resources to the Fund to support other members (Greece, Pakistan, Colombia), and provide the metric under which each member can borrow from the Fund. Because their economic and financial positions differ over time, not all members of the IMF are in a position to lend to the Fund at every point in time. A rule of thumb is that about 70 percent of quota subscriptions are available, amounting to about $275 billion currently.
Under the proposed agreement, total IMF quotas will be increased to about $750 billion, of which about $550 billion normally will be lent out or available to lend. This $275 billion boost to the IMF's lending resources is significant. Unfortunately, however, it largely will amount to a repositioning of existing IMF resources from the New Arrangements to Borrow (NAB) to quotas.3 For those IMF members that participate in the NAB, their NAB commitments will be reduced proportionately to fund increases in their IMF quotas. The size of the NAB, which is now about $555 billion, will be reduced proportionately, to about $225 billion, compared with $50 billion prior to the crisis. For the United States, the US Congress will have to authorize a shift of about $65 billion from the US NAB commitment of about $108 billion, which was approved in July 2009, to the US quota subscription.
It is regrettable, in my view, that the agreements over the past two weeks did not include a net increase in financial resources available to the IMF. As I have written elsewhere, the IMF needs substantial additional potential funding if it is going to become a more effective international lender of last resort.4 Nevertheless, the doubling of IMF quotas is a positive achievement.
First, it acknowledges that the IMF is, as it should be, a quota-based institution, relying in the first instance on quota subscriptions from all members to finance its lending operations and only calling upon borrowing arrangements with a more limited number of countries, such as the NAB, in emergency circumstances. The agreement restores this balance, though it may have gone too far in raiding the NAB "reserve tank."
Second, the agreement provides for a substantial increase in the quotas of all members. For most of them, this augmentation, which is expected to take effect in two years at the earliest, will be the first increase since 1998. It is long overdue. Over this 14-year period, global GDP will have increased more than 125 percent, global trade more than 200 percent, and gross international financial flows by substantially more than that. A doubling of IMF quotas was the minimum necessary to support the IMF. It maintains its potential lending capacity at about $750 billion.
Distribution of Shares
IMF quotas determine not only each member's commitment to lend to the Fund to support other members and a metric for each member's rights to borrow from the Fund on appropriate terms, but also each country's voting power in the IMF.5 Because of historical inertia, the IMF for too long has been dominated by the traditional industrial countries, including, for this purpose, Canada, Japan, and Australia. This pattern changed a bit with the agreement in April 2008, and now it will change further. There are many different ways to describe this shift.
First, the IMF reform package overachieves on the commitment to "shift quota share to the dynamic emerging market and developing economies of at least 5 percent from overrepresented to underrepresented countries." The negotiators achieved shifts of 6 percent or more on both bases. If one looks under the hood of these numbers there is plenty of room to quibble, of course, in part because the negotiators were more or less free to define the terms of any comparison.
Second, compared with the situation five years ago, the 26 countries classified as "advanced" for these purposes will have reduced their voting share by 5.2 percentage points to 55.3 percent. Half that reduction is associated with the decisions taken over the past two weeks, but the contribution of the quota-share component provides the entire boost in this agreement. The 2008 agreement included a boost in basic votes that benefitted smaller members on balance.
Third, the corresponding boost in the voting shares of the remaining 162 members of the Fund will have increased since 2005 by the same amount—5.2 percentage points. My preference would have been a shift from the so-called advanced countries to the other members of the IMF on the order of 10 percentage points. By this crude standard, we are slightly more than halfway there.
Fourth, much of the adjustment will come from a reduction in the combined voting share of the advanced countries, including members of the European Union as a group (3.1 percentage points) to 29.4 percent. The US voting share will decline slightly by 0.5 percentage points to 16.5 percent. The adjustments in the combined shares of EU members include increases for some members, such as Spain, Poland, and Ireland, which are currently under-represented by most metrics. Nevertheless, the European Union will remain over-represented in the IMF. Its share of global GDP as used in the quota calculations is 27.8 percent while its voting share will be 29.4 percent.6 The comparable figures for the United States are 21.6 and 16.5 percent.
Fifth, one challenge in reaching agreement on new quota shares has been that the negotiations have been based on a formula that was revised in 2008 but that points in the wrong direction.7 The formula implies that the combined quota share of the advanced countries will be reduced too much, in particular the quota share of the European Union as a group. As a consequence of these flaws, the quota negotiations largely had to set aside the 2008 formula and rely on a bevy of ad hoc metrics to achieve a balanced political result. The good news is that there is a commitment to revise the quota formula further by January 2013—presumptively with a greater weight on GDP and GDP measured in purchasing power parity terms—in time for the next quota review, which has been moved to January 2014. However, one should not hold one's breath about the result from the formula review or expect that an increase in quotas will be agreed in 2014.
Sixth, one positive result of this negotiation is that the oil-producing countries, whose quota shares were doubled in the 1970s, will have their quota and voting shares scaled back because their growth performance as a group over the past three decades has not been sufficient to justify their larger combined share. On the other hand, the combined voting share of the low-income members of the IMF has been protected at 4.5 percent for the group as a whole. While this is politically appropriate, it is unfortunate that the voting share of members of this group has been protected. It would have been more appropriate to adjust relative shares within this group, rewarding better performers.
Seventh, much has been made of the fact that the countries with the 10 largest voting shares in the IMF will be the United States, Japan, Germany, France, the United Kingdom, and Italy plus the four BRIC countries. Saudi Arabia and Canada will be pushed out of the top 10, and India and Brazil will have moved into the top 10 group and ahead of the Netherlands and Belgium as well.8
On the whole, the agreement represents significant positive progress on the shares issue. It might have been better, but it could have been worse. We might have been stuck with the status quo.
Distribution of Chairs
In addition to their outsized voting power in the IMF, the advanced countries of the North Atlantic, Europe in particular, dominate the 24-member IMF executive board. Advanced European countries (including Switzerland for this purpose) hold eight of the 24 seats and two other countries (Spain and Ireland)—as part of constituencies led by non-European countries—might sometimes sit in the chairs as members or alternates. In a body that seeks to reach decision by consensus, this over-representation is a serious problem.9
Until US Treasury Secretary Geithner forced the issue in August 2010 by declining to agree to continue with an executive board of 24 seats rather than the 20 seats provided in the IMF Articles of Agreement, the Europeans had succeed for more than half a decade in keeping the issue of the distribution of chairs on the executive board off the agenda of IMF reform. Secretary Geithner should be congratulated for his bold move. Outsiders can be critical of how much was achieved in the end, but again this agreement involves significant progress, especially when one considers that four months ago it looked as if there would be none.
It has been agreed that the advanced European countries, which include Switzerland for this purpose, will give up two executive director seats on the executive board in late 2012, assuming the overall IMF reform package has been approved by that time. It has also been agreed that at that time all members of the executive board will be elected. (At present the five members with the largest quotas each appoint their executive director, and that person is not allowed to represent any other country. The other 19 executive directors are elected by the remaining 182 member countries.) As part of the package, the size of the executive board will be set permanently at 24 seats. Finally, the composition of the executive board will be reviewed every eight years. The presumption is that these last two elements in the package will lead to further consolidation of European seats in the future and more frequent adjustments in constituencies over time.
What is one to think about this component of the agreement?
First, it is disappointing that there are likely to be no changes in the composition of the executive board seated later this month. Impatient reformers must be patient. Moreover, one cannot be certain what will happen when the agreement to reduce representation by the advanced European members by two seats takes effect. If they are merely replaced by other members of the European Union, e.g., Poland and Hungary, which technically are not classified as advanced countries, that would be an unfortunate breach of faith.
Second, it would have been preferable in my view to return the size of the executive board to 20 seats. This would have increased its collegiality and reduced the considerable cost associated with the maintenance of the offices of executive directors.
Third, it is clear that progress will be slow. In particular, if the package is not agreed by October 2012 as now envisaged, implementation will be delayed another two years because elections of executive directors occur only in even years.
A number of important issues were not included in the announcements of the past several weeks: management selection, voting majorities, reform of the procedures of the executive board, possible establishment of a council, and increasing IMF transparency and accountability.
On management selection, as is well known, the principle of an open, merit-based selection process for the managing director of the IMF and the president of the World Bank has been accepted. However, many observers would like to add the qualification that selection should be without regard for nationality. In other words, the IMF position should not be reserved for a European and the World Bank head should not be reserved for a US citizen. It is important for this "convention," which has prevailed for the past 65 years, to come to an end. I am confident that it will end with the next elections in both institutions. I do not see that there is much to gain from an explicit statement that those individuals not be citizens of Europe or the United States. Much more important is that the principle of open, merit-based selection be applied to all international financial institutions and not merely to their heads.
On voting majorities, I would favor a reduction of the current 85 percent weighted majorities to 80 percent or less. Some would go further. That is not going to happen. Not only does the United States not want to give up its power to block a small set of institutional decisions, but a number of other countries or groups of countries also do not favor such a change. The Europeans say that they favor it. Of course, they would retain the collective power to block any decisions as long as their voting power remains close to where it would be after the current agreements are implemented—29.4 percent—so their views are not to be taken seriously. These voting majorities are a feature of an imperfect world. Some also favor the introduction of double-majority voting, combining the weighted votes of countries with the number of countries. That would be a good idea, in my view, in the selection of the managing director, requiring a majority of members as well as a majority of the weighted votes. De facto, that is what happens now. Moreover, to introduce double majorities into all votes in the IMF, as some advocate, would turn the IMF into the UN General Assembly with all the well-known limitations in the effectiveness of that body.
On reform of the executive board, which is sorely needed, I would say: "Board, reform yourself!"
On the proposed council with formal voting powers to replace the advisory International Monetary and Financial Committee, my view is that it would do no harm, but also be of little benefit. The idea is to force finance ministers to take their responsibilities more seriously by making them formally accountable. I have my doubts.
On transparency and accountability more generally, much progress has been made over the past decade or so. More progress is required. For example, the background papers for this round of IMF reform discussions should have been made public as soon as the discussions were held rather than after their conclusion. In addition, formal statements by executive directors should be routinely circulated to all members and released to the public with a short time lag. Again, I would say, "Board, reform yourself!" In this case, however, reform requires agreement of the member countries that the board members represent. In the end, the IMF is not an abstraction of 19th Street in Washington, DC; it is the member countries.
The Outlook for Implementation
The IMF reform package agreed by the G-20 and by the IMF executive directors is a good piece of work. The package involved many compromises, too many for some. I hoped for greater substantive progress, but it will be difficult enough to obtain final implementation of this package. The much more modest agreements of April 2008 and April 2009 have yet to be implemented. Many groups will oppose this package for a variety of reasons, some better than others. My view, at this stage, is let not the perfect, or even the better, be the enemy of the pretty good.
2. When fully implemented, the BRICs will have 13.45 percent of IMF votes.
3. Technically the April 2009 G-20 agreement to increase the NAB to $555 billion has not yet gone into effect, nor have the adjustments in IMF quotas that were agreed in April 2008 become effective, but the presumption is that this will happen soon and all calculations treat those decisions as implemented.
5. Each member receives a specified number of "basic votes," which account for roughly 5 percent of total votes, and additional votes based on the size of its IMF quota.
6. The IMF quota formula uses for its GDP variable a blend of 60 percent of GDP measured at market prices and exchange rates and 40 of GDP measured on a purchasing-power-parity basis.
7. See pages 26–27 in Truman, Edwin M. 2010. The G-20 and International Financial System Governance. Peterson Institute for International Economics Working Paper 10-13 (September).
8. It is noteworthy that Spain also will have a larger voting share than either the Netherlands or Belgium.
9. See Edwin M. Truman. 2006. Rearranging IMF Chairs and Shares: The Sine Qua Non of IMF Reform. In Reforming the IMF for the 21st Century ed. Edwin M. Truman. Peterson Institute for International Economics Special Report 19.