Journal Issue

Interview with Clark and Polak: SDRs could meet growing demand for reserves at no cost while reducing systemic risk

International Monetary Fund. External Relations Dept.
Published Date:
February 2003
  • ShareShare
Show Summary Details

Clark and Polak on SDRs

IMF Survey: How did the concept of a global stock of international liquidity become irrelevant?

Polak: Over the past century, people interested in international financial policy—those working in universities, ministries of finance, and central banks—have focused a great deal of attention on one question: Will the stock of international liquidity enable countries to accumulate a level of reserves that can support continued growth in the world economy? Several conferences have dealt with this subject, including two that were held immediately after World War I in Brussels and Genoa, the 1933 World Economic Conference held in London, and the 1944 Bretton Woods Conference.

But it was not until 1969 that an apparent resolution of the question was found with the introduction of the SDR—a new form of international liquidity that would allow the international community to create the amount of reserves that the world needed. At that time, reserve assets were mainly in the form of U.S. dollars, whose supplies were constrained by the Bretton Woods system of fixed exchange rates, and in gold. Increases in the supply of dollars to foreign holders threatened U.S. official gold holdings—a problem known as the Triffin dilemma.

Not much more than a decade later, the problem evaporated. It wasn’t that the SDR solved it. Instead, the problem itself went away when the United States gave up its commitment of gold convertibility of the dollar in 1971. All of a sudden, there proved to be an almost limitless stock of dollars and other major currencies, which countries in need of reserves could either earn or borrow, thus taking away the very basis of the SDR.

IMF Survey: What influences the demand for reserves?

Clark: A country’s demand for reserves is related to the size of changes in its balance of payments, typically in the current account. In fact, there is a fairly close relationship between a country’s level of reserves and its imports of goods and services. Over time, fluctuations in the capital account have also assumed increasing importance. During recent crises—in Asia, Russia, and Latin America—massive fluctuations in capital flows boosted the need for reserves to help ward off a crisis and moderate the impact of capital outflows on exchange rates. Research at the IMF and elsewhere suggests that the ratio of reserves to short-term debt may be a key indicator of reserve adequacy in countries with substantial, but uncertain, access to capital markets. Greater capital market integration tends to increase the demand for reserves.

There are also situations in which countries’ demand for reserves may decline. For example, to the extent that countries move from relatively fixed to flexible exchange rate arrangements—that is, they respond to external imbalances by allowing the price, rather than the quantity, of foreign exchange to adjust—the need for reserves to intervene in the foreign exchange market would be expected to diminish. Many observers thought this would occur with the demise of the Bretton Woods par value system, but reserves have continued to grow significantly over the past 20 years. Even if a country only lightly manages its exchange rate, with a relatively closed capital account it would still want to hold reserves—and probably increase them over time—to help smooth output fluctuations arising, for example, from large movements in its terms of trade. Moreover, research indicates that the shift to floating rates has not been as dramatic as popularly believed. Ken Rogoff and Carmen Reinhart have found that there is little evidence of countries—other than the major industrial countries—moving significantly toward floating exchange rates. So, the shift to floating is perhaps more apparent than real, and it hasn’t had any clear impact on the world’s long-term growth in reserves.

IMF Survey: What options are open to countries in acquiring reserve assets?

Clark: One way is for a country to improve its current account. But that requires it to reduce domestic demand—give up the use of resources for consumption or investment. Obviously, this involves a real resource cost, which is measured by the country’s rate of return on investment and is typically manifested in a market-determined interest rate in that country. The rate is fairly low for some emerging market economies and industrial countries, but for many other countries it is quite high. A country can also obtain reserves by issuing foreign-currency-denominated bonds in international capital markets or by borrowing from banks at market interest rates. It could then hold these reserves in the form of deposits or short-term assets in the major reserve currency countries to earn interest.

IMF Survey: How about the terms?

Clark: These can vary widely. Most advanced countries can borrow at interest rates that are only slightly higher than the return on reserve assets. They can satisfactorily finance increases in reserve holdings by borrowing in international capital markets and therefore really have no need for an SDR allocation to supplement reserves. However, for other countries—largely emerging market economies—the difference between the interest rate on their sovereign bonds and the return on reserve assets is typically much higher. They must also deal with dramatic changes in the availability of funds and the terms on which they can borrow, as we have seen in times of crises. For these countries, borrowed reserves are a useful supplement to owned reserves acquired through current account surpluses, but they are often unreliable and their cost can vary significantly, particularly when countries need them most.

IMF Survey: Despite changes in the international monetary system, you still make a case for SDR allocations to improve the system’s operation and to meet most countries’ need for reserves. Why?

Clark: First, most countries wish to add to their reserves over time. Second, adding to reserves is costly through both current account adjustments and borrowing. The main argument for SDR allocations is that SDRs can be created essentially without cost—other than a very small administrative cost. This argument is analogous to substituting fiat currency for commodity currency—that is, the substitution of currencies for gold clearly saves society the resource cost of digging gold out of the ground and storing it. Third, SDRs can reduce systemic risk. Rather than depend on borrowed reserves, countries would have access to more owned reserves, which would reduce their vulnerability to changing conditions in international capital markets. They also wouldn’t have to pay interest to borrow reserves, making them a better credit risk.

IMF Survey: Why can’t the IMF simply provide more credit, rather than distribute new reserve assets?

Clark: Conditional credit is, of course, very important for the IMF. But there is no reason the IMF’s activities should be restricted to conditional financing—that is, under IMF programs. The IMF itself has always recommended that members not depend exclusively on conditional reserves. Indeed, an adequate stock of owned reserves is, in almost all cases, a condition in IMF programs. Since the recent balance of payments crises, the IMF has urged members to build up their stock of owned reserves and has issued a number of papers on reserve management emphasizing the role of reserves in reducing a country’s vulnerability to crises.

IMF Survey: If the IMF’s membership were to back a resumption of regular SDR allocations, what would be their optimal timing, frequency, and size?

Polak: Much of the criticism surrounding the idea of SDR allocations is based on the misconception that countries would suddenly receive a large amount of money that they would be very tempted to spend instead of hold as reserves. That was never the idea of the SDR mechanism. On the contrary, given that countries wanted to consistently build up their reserves in line with the increase in their international transactions, the idea was that allocations would be made annually to take care of this slowly growing need. Allocations shouldn’t be a onetime event. They should take place regularly, annually or even quarterly, in modest amounts.

Clark: The question of the optimal size of an SDR allocation is a difficult one, even for a single country. The answer would require consideration of a range of variables, including the cost of acquiring reserves, the cost of adjustment in the absence of reserves to finance payments imbalances, and a country’s risk preferences. To make estimates for all members would be a daunting task. The past approach to SDR allocations involved estimating growth in demand for reserves and then judging what fraction of that growth should be satisfied by SDRs. It is important not to make the allocation so small as to be inconsequential or so large as to cause concerns about inflation.

Polak: A fundamental element of the design of the SDR system was that the size of the allocation to a country would be based not on its need for reserves at a certain time but on an objective measure that would reflect the country’s relative need for reserves over the long term. After much discussion, it was decided that, in the absence of a better alternative, IMF quotas would serve as that measure.

IMF Survey: Since the last allocation of SDRs in 1981, there has been considerable resistance to additional allocations. Why?

Industrial countries have easy access to capital markets, so they don’t need the SDR system. That wasn’t so at the time the SDR was created in the 1960s.

—Jacques J.Polak

Polak: This resistance, which dates back to shortly after the first allocation in 1970-72, comes from industrial countries. Some of them considered the last allocation to be a political arrangement and not a logical extension of the SDR system because, by then, many of the changes in the international financial system that we talked about earlier had already become evident. Industrial countries have easy access to capital markets, so they don’t need the SDR system. That wasn’t so at the time the SDR was created in the 1960s.

In addition, some may find awkward the idea of a single institution—the IMF—providing both conditional and unconditional credit in the form of SDRs. Industrial countries, in particular, have difficulty with this concept. But their objection to SDR allocations—helping debtor countries maintain an adequate level of reserves at no cost to the creditor countries—is curiously inconsistent, because these same countries are bearing the considerable cost of providing very low interest loans to finance other capital needs of the developing countries through IDA [World Bank’s International Development Association] loans and the IMF’s concessional loan facility.

IMF Survey: Do you have any reason to be optimistic that this resistance can be overcome any time soon?

Polak: I’m not at all optimistic that it can be overcome in the near future. In the mid-1990s, industrial countries favored a special SDR allocation to help channel resources to Russia and other countries that had been members of the former Soviet Union. And in 1997, at the [World Bank-IMF] Annual Meetings in Hong Kong, the IMF’s Board of Governors finally endorsed this allocation, but it hasn’t become a reality because the United States has not passed the necessary legislation.

While SDR allocations are no longer favored by most industrial countries, they have recently received attention in nonofficial circles. For example, the Zedillo Report—prepared for the UN International Conference on Financing for Development held in Monterrey, Mexico, in March 2002—advocates a resumption of SDR allocations, and George Soros has put forward a proposal to use SDRs to finance expanded foreign aid. Although these proposals were not endorsed in the Monterrey Consensus, at least they were given consideration.

Copies of IMF Working Paper No. 02/217, International Liquidity and the Role of the SDR in the International Monetary System, by Peter B. Clark and Jacques J. Polak, are available for $15.00 each from IMF Publication Services. For ordering information, see page 21.

Other Resources Citing This Publication