Journal Issue

A Developing Country Viewpoint: A New Way Forward on IMF Quota Reform

International Monetary Fund. External Relations Dept.
Published Date:
February 2007
  • ShareShare
Show Summary Details

At the UN’s Monterrey Summit in March 2002, heads of state and government agreed to broaden and strengthen the participation of developing countries and economies in transition in international decision making and norm setting. But five years later, it is clear that the reform of the IMF’s governance—which encompasses, among other things, quotas, voting rights, and voice—has progressed rather slowly. This situation urgently needs to be turned around, as recognized by the International Monetary and Financial Committee, which regularly reviews progress on the Monterrey consensus. Its April 2006 communiqué stated that the IMF’s effectiveness and credibility as a cooperative institution must be safeguarded and its governance further enhanced, emphasizing the importance of fair voice and representation for all members.

In an effort to further the debate on how best to formulate a proposal for quotas and voice reform in the IMF, we have looked to John Rawls, arguably the greatest political philosopher of the 20th century, for guidance. We feel that his theory of justice provides an appropriate method for understanding the core issues in this debate. We use the Rawlsian notion of “justice as fairness” to demonstrate that justice in the governance structure requires a distribution of voting power that participants accept as the end result of a fair process—and that a major revision of the quota formulas is long overdue.

Problems with current governance

So how are quotas calculated? When the IMF was established in 1944, a formula—which would become known as the Bretton Woods formula—was developed. It contained five variables: national income, official reserves, imports, export variability, and the ratio of exports to national income. In the early 1960s, this formula was supplemented with four more formulas (which contained the same basic variables but with larger weights for external trade and external variability). But two different data sets were used, making, in effect 10 formulas. In 1981-82, the number of formulas was reduced and the variables used were simplified.

What we are left with today are five formulas that try to capture economic size, openness, and the demand for and supply of IMF resources. The calculated quota serves multiple purposes—including voting power and a country’s access to financing—meaning that it has to balance competing considerations about what variables to include in the formulas and the weight to attach to each variable. Over time, members’ quotas have become increasingly out of line with their economic weight in the global economy. In addition, the current quota formulas do not capture some important aspects of members’ economic situation and other variables that should have a bearing on voting rights.

Another problem is that the Bretton Woods founding fathers of the IMF and the World Bank reached a compromise to balance the Westphalian principle of the legal equality of states (which called for one country, one vote) and the economic argument for basing votes on capital contributions—but the balance has been eroded over time. In 1944, they decided to allocate 250 “basic votes” to each member country, ensuring that each country would have a voice. However, since then, IMF quotas have increased 37-fold, with basic votes remaining unchanged. The result has been a drastic reduction in the participation of small countries in decision making.

What Rawls would say

As we weigh how to improve IMF governance, Rawls helps us by stressing in his theory of justice that we should imagine a situation in which a group of individuals are brought together to agree on the basic constitution of an international institution that they are about to enter, but in which, to ensure their impartiality, they are placed behind a veil of ignorance—a device that screens out information about, among other things, population, national output, and level of development. In deciding what the quota formula should be, the representatives must make a choice without knowing what type of country they would represent. With the expulsion of bias-inducing knowledge, the participants are forced into the moral point of view, which allows Rawls to claim that he has set up an inherently fair procedure.

We cannot know for sure what quota formulas would be chosen by rational actors in the hypothetical situation postulated by Rawls. But this should not stop us from doing our best to imagine what that outcome might be, even while recognizing that reasonable people can disagree. That said, it is much easier to demonstrate what does not satisfy Rawlsian principles than what does, and the quota formulas and the IMF’s current governance structure do not even come close to justice as fairness. Thus, it is clear that a rational representative considering the governance structure behind the Rawlsian veil of ignorance would not support the current voting distribution that gives almost no weight to the Westphalian principle of “one nation, one vote” and an exact zero weight to the democratic principle of “one person, one vote.”

Devise a new quota formula

So what should a new quota formula include? We would propose the following changes:

Add population. Some observers argue that even though population is not a variable in the quota formulas, it is in effect included—and thus not explicitly needed—because there is a correlation between population and other variables, such as GDP. We disagree and call for the inclusion of population as a variable for several reasons. First, because population data are available, there is no need for a proxy variable. Second, the correlation coefficient between population and other variables would be higher than that between population and GDP—especially if a purchasing power parity (PPP)-based measure of GDP, rather than market exchange rates, were used. Third, GDP is meant to capture a country’s economic size and ability to contribute resources, not the worth of human beings as human beings, which cannot be measured by some aggregate economic concept.

Use PPP, not market rates. Our analysis shows that calculated quotas would be sharply different if appropriate variables for the potential supply of and demand for IMF resources were used. Regarding an economy’s capacity to contribute financial resources, the formulas have always used GDP converted at market exchange rates, but PPP-based GDP is the more relevant indicator for measuring potential contributions. The latter variable correctly shows that the larger the volume of goods and services an economy produces, the greater its size in the world economy. The argument for using the former variable misses the point that there is no need to convert nontraded goods and services at market exchange rates to pay for quotas. Quotas are a small fraction of GDP or exports, and one cannot imagine an IMF so large that countries will have to sell nontradables to pay for their quotas. Moreover, the variable chosen makes a big difference in the calculations of quotas. Indeed, advanced economies’ share of global GDP at market exchange rates is more than 75 percent but only about 50 percent when PPP is used.

“The solution thus lies in a sharp increase in basic votes—which would automatically give low-income countries a bigger voice—on top of reforming the quota formula, which would help all developing countries.”

Capture demand for IMF resources better. The need to include an openness variable—the first demand variable—is based on the argument that more open economies are more vulnerable to external shocks and, therefore, will be more likely to use IMF resources. The irony is that advanced economies have 70 percent of the share of current payments and receipts in global totals, which means that the bulk of the share in calculated quotas from this variable is given to countries that have not borrowed from the IMF in decades and are not expected to borrow in the foreseeable future. Variability of current receipts—the second demand variable—has the same problem because the advanced economies’ share exceeds 60 percent. If these variables are capturing the demand for IMF resources, why is their correlation with actual use close to zero?

The traditional openness and variability variables in the quota formulas need to be replaced with new ones that have a high correlation with the actual use of IMF resources. One promising approach would be to include those variables that are good predictors of demand for IMF resources. They could include past use of IMF resources, subinvestment-grade credit rating, sovereign bond spreads, and reserves and short-term debt. There are good reasons for including the capital flow volatility variable in the quota formula but for normalizing it to reflect major differences across countries. If two countries experience the same capital account shock in absolute terms, the smaller one will face a greater burden, which suggests that the volatility of capital flows should be measured as a proportion of GDP. Several important differences between the capital flows of industrial versus developing countries should also be taken into account.

Boost basic votes

Finally, the Rawlsian approach underscores that enhancing the IMF’s legitimacy means giving top priority to ensuring that lowincome countries have an adequate opportunity to participate in the institution’s governance. Their share in the world economy is small, but the IMF has a disproportionately large role in these countries in terms of policy advice, financing, and conditions attached to IMF-supported reform programs. Correcting this underrepresentation would also help address the problem of inadequate country ownership of reform programs. The solution thus lies in a sharp increase in basic votes—which would automatically give low-income countries a bigger voice—on top of reforming the quota formula, which would help all developing countries.

Abbas Mirakhor and Iqbal Zaidi

IMF Executive Director and Senior Advisor, respectively, for a constituency

covering the Islamic Republic of Afghanistan, Algeria, Ghana,

Islamic Republic of Iran, Morocco, Pakistan, and Tunisia

This article is based on IMF Working Paper No. 06/273, “Rethinking the Governance of the International Monetary Fund,” by Abbas Mirakhor and Iqbal Zaidi. Copies are available for $18.00 each from IMF Publication Services. Please see page 64 for ordering details. The full text is also available on the IMF’s website (

Other Resources Citing This Publication