New mandates for the IMF and World Bank.

Meltzer, Allan H.

Much of what the charters of the International Monetary Fund and World Bank say about purposes and objectives is out of date. The current mandate of the IMF should be to reduce global risk to an attainable minimum. The mandate of the World Bank should be to facilitate social and economic development as a means of reducing poverty.

Ending the IMF's Command-and-Control System

How can the IMF reduce risk to an attainable minimum? The IMF has two principal functions that can improve the market's operations in ordinary times and in crises. One function is to increase the quantity and improve the quality of information available to private lenders. The other function is to reduce the risk of financial crises in a given country and the spread of crises to other countries, as in Latin America in the 1980s and Asia in the 1990s.

Under pressure from its critics, the IMF has made much more information available about its activities, recommendations, and assessments. This information can be used by private lenders to improve their assessment of risks in a given country. This is particularly important for making judgments in the ordinary course of lending. Many problems in developing economies arise or are exacerbated by the volume of short-term renewable loans used to finance risky, longer-term assets. Timely release of information about a country's debt structure and performance can reduce this type of lending.

Important as is the improvement of information, the most important function of the IMF is to reduce the risk of severe crises that spread internationally. Improved information contributes to that goal, but reform of IMF procedures is also important. Prodded by its critics and its new management over the last three years, the IMF improved its operations and recommendations in several ways. It now restricts the conditions attached to its loans to a small number of macroeconomic and financial measures or objectives. It appears less willing to make massive loans than in the 1990s. And it pays more attention to avoiding crises and to determinants of debt sustainability in developing countries.

The most important single change remains undone. The IMF should move from its command-and-control approach to one that relies on incentives.

Historically, the IMF has attached conditions to its loans. The country agrees to the conditions to get the loan, but it may be politically unpopular at home to enforce conditions such as expenditure reduction or tax increases. Or, growth may be less than anticipated, requiring additional painful adjustment. The IMF's Independent Evaluation Office (2003: 7-8) found that countries achieved about one-half of the proposed change in fiscal balance on average. About 60 percent of the programs underperformed.

This command-and-control approach has the unfortunate side effect of making the IMF appear responsible for imposing harsh measures under adverse circumstances. The country's government, of course, agrees as a condition of the loan. This does not remove the IMF's responsibility in the minds of the country's electorate, the protestors at international meetings, and much of the public.

I believe that reform occurs when the country's leaders, a majority of its citizens, or both, want reform. Reform cannot be imposed successfully by external technocrats without local support. Local governments can, and do, frustrate reforms or ignore IMF (or World Bank) conditions. The reason Turkey, Brazil, Argentina, Ecuador, and others have repeated crises is that they do not reform enough to avoid them, They promise, but they do not reform. Command and control fails, as we expect it would.

The main reform needed at the IMF is development of an incentive system to replace command and control. Briefly, the IMF should establish a short list of policies or observable standards that countries should adopt to be assured of assistance in a crisis. It should use its surveillance to ensure that a country meets the standards and then publish the list of countries that do--and do not--get a guarantee of crisis assistance. The IMF would not help countries that do not meet the standard. To prevent crises from spreading, the IMF would assist countries that are victims of crises in their neighbors or trading partners.

Countries that adopt the standard would be subject to less risk. Hence, they could borrow more capital at a lower interest-rate spread over U.S. Treasuries. Other countries would get less capital and pay a higher interest rate. This approach would give the government and the public considerable incentive to adopt stabilizing policies. The capital markets, not the IMF, would impose discipline.

The IMF itself is at risk. As my colleague Adam Lerrick has shown, that risk is a cost to the United States (and other countries) but does not appear in our budget. Lerrick estimates the hidden annual cost of the IMF to U.S. taxpayers currently as $1.5 to $2 billion. The principal component is the risk of default by one of the major debtors.

Four countries--Argentina, Brazil, Indonesia, and Turkey--owe about 70 percent of the IMF's outstanding debt. The IMF avoids default by lending more money or, as in the case of Argentina, by extending the maturity of the debt. As in the past, the IMF will eventually come to the Congress for a quota increase either because of a default or because its resources are allocated to unpaid loans.

Reform of this system should be a priority. The Bush administration, to its credit, has made considerable progress in getting collective action clauses into private debt contracts. Reform of debt repayment to international financial institutions and to lenders should be next on the agenda.

Reforming the World Bank

In the past few years, the administration and the Congress have insisted on some of the reforms advocated by the majority report of the International Financial Institution Advisory Commission. Monitored grants replaced some of the lending to the poorest countries. The administration has worked to set explicit conditions that can be monitored and has introduced incentives for countries to meet those conditions. In its most recent budget, Congress required an independent performance audit of some International Development Association (IDA) programs and insisted on greater transparency at the World Bank.

These steps are a good start, but only a start. The central issue about the World Bank with its many programs is: It spends or lends about 820 billion a year but neither we nor they know which programs are effective and warrant expansion or retention, and which are ineffective and inefficient and should be abandoned. The monitoring that Congress insisted upon for some IDA programs should be extended to the entire World Bank and its affiliates.

There are two ways to gain the needed information. One is an independent performance audit by an outside agency. Another is development of an independent, internal group similar to the Government Accountability Office or the IMF's Independent Evaluation Office. The current arrangement does not meet this standard.

An example will illustrate the problem. We have considerable evidence that poverty has declined dramatically as measured by the number of people living on $1 per day or less. The decline is most striking in Asia, especially in China and India. Market opening, private investment, protection of property rights, and the like contributed much to the improvement. Where these spurs to growth and development are largely absent, as in Sub-Saharan Africa, poverty has increased. Did World Bank programs contribute to the reduction of poverty in Asia? Did these programs ameliorate worsening prospects in Africa? The Congress should require answers to these questions.

Further, the World Bank should concentrate on the hard cases--the impoverished countries. The Bank should have an explicit program for graduation. Countries that can borrow in the capital markets with investment grade ratings should not receive subsidized loans. Those loans can be better used to provide potable water, sanitary sewers, disease control in the poorest countries, and to encourage countries to adopt institutional reforms that have been effective in spurring development. These include the rule of law, open trading arrangements, and protection of property rights and individual rights.

Finally, we should insist that the IMF and the Bank eliminate overlapping responsibilities. The World Bank should become a more effective development bank. The Bank has estimated that $1 trillion a year is paid in bribes in all countries. A large part is in the developing countries. Ridding the system of corruption is a major challenge. The IMF's responsibility should remain the maintenance of global financial stability. As a result of experience in the Asian crisis, many Asian countries have accumulated substantial reserves to protect them against crises and to avoid being put under IMF supervision. They have also established a regional lending system outside the IMF. This, too, opens questions about the future role of the IMF.

New leadership at the IMF and the end of James Wolfensohn's term at the World Bank in 2005 provide an opportunity for new approaches and much needed reform.

Reference

International Monetary Fund (2003) "Fiscal Adjustment in IMF-Supported Programs." Washington: Independent Evaluation Office, IMF (September).

Allan H. Meltzer is the Allan H. Meltzer University Professor of Political Economy at Carnegie Mellon University and a Visiting Scholar at the American Enterprise Institute. He was Chairman of the International Financial Institution Advisory Commission, U.S. Congress.