The IMF and U.S. International Policy

The case for the massive IMF/​World Bank response to the recent Asian crisis reminds me of an all‐​too‐​frequent proposal to jump‐​start economic growth: the combination of massive demand stimulus and a solemn promise never to do it again. The problem of this type of policy, of course, is that the initial response undermines the credibility of the promise. Secretary Rubin seems to understand the moral hazard problem caused by socializing the losses on international loans, but he claims not to know what to do about it. The young St. Augustine was rather more honest with himself; when faced by a similar problem, he prayed: Lord, make me chaste, but not quite yet.

For there should be no doubt about the nature of the choice that was made by the response to the recent Asian crisis: the international financial establishment committed over $100 billion to reduce the near‐​term contagion effect of the recent Asian crisis without apparent regard for a longer‐​term contagion effect that this bailout will probably increase the number of similar future crises in these and other countries. The historical record is clear: Most of the less‐​developed nations funded by the IMF have later returned for more funds. Mexico, for example, has had a financial crisis in each of the past four presidential‐​election years. A total of 84 nations have been in debt to the IMF for 10 years or more, 43 nations for 20 years or more. And there is little doubt that the massive IMF and U.S. bailout of Mexico in l995 contributed to the near‐​doubling of capital flows to East Asia that same year.

Finance ministers and central bankers will commit almost any amount of our wealth to avoid a major financial crisis on their watch, even when they recognize that the socialization of losses increases the probability of a crisis on some later watch. Rather than resolving the conditions that lead to financial crises, the IMF treats each successive crisis as a new event, indirectly assuring that there will always be a queue of new crises to address. U.S. government membership in the IMF is like being a limited partner in a financial firm that makes high‐​risk loans, pays dividends at a rate lower than that on Treasury bills, and makes large periodic cash calls for additional funds. The current administration campaign to convince Congress to approve more funds for the IMF is also quite deceptive. To some groups, the officials suggest that more funds are necessary to help the poor starving children of Nameyourland. In fact, the IMF bailouts are a form of insurance for the foreign and domestic individuals, firms, and banks that had made high‐​risk investments in the country subject to the crisis du jour. The 1995 Mexican bailout, for example, insured those who had purchased the 28 day government bills, providing little help for the general Mexican population for whom the real per capita income is now less than before the bailout. Similarly, the administration seems to have gained the support of the congressional Democratic leadership for new IMF funds on the premise that such funds would reduce the exchange rate effects and resulting trade effects of future crises. In fact, the exchange rate of an IMF client generally stays weak for some time after a bailout. The dollar value of the Mexican peso, for example, is now less than half that before the l994 crisis, with the effect that Mexico has since had a trade surplus with the United States. Finally, the administration has gone around the world making a series of promises and then asserts that congressional support of these promises is necessary to maintain U.S. leadership. The Clinton administration did not invent this gambit but it has been especially consistent in using this argument to support its position on trade negotiations, global warming, NATO expansion, Iraq, and now the IMF.

For now, it looks like the bailout of Thailand, Indonesia, and South Korea is history, a done deal for which the IMF does not need any more funds. So the current issue is whether the IMF should be refunded to prepare for the next round of financial crises. For now, I suggest, Congress should defer a decision to refund the IMF until it has a better understanding of the conditions that lead to a financial crisis, the moral hazard effects of socializing the losses on international investment, the long‐​term record of the IMF, and the feasible alternatives — including the implications of no multilateral governmental response to a financial crisis in any country.

It is especially important, for example, to understand the reasons why the recent Asian crisis was limited to Thailand, Indonesia, and South Korea but with much less effect, at least so far, in Singapore, Hong Kong, Taiwan, and China. My initial judgement is that two patterns are common to the problem countries in Asia and also in Mexico:

It is also important to understand why the frequency and magnitude of financial crises are increasing. Studies by the IMF and the World Bank have documented some 90 episodes of severe banking crisis over the past 15 years, a period of relatively stable economic growth. For this condition, I suggest, the IMF and the World Bank bear substantial responsibility. When a borrower is illiquid or insolvent, the only way to avoid the moral hazard problem is a financial workout in which both the borrower and the lender take a major hit:

Private bankers have handled such problems for generations, long before the IMF and the World Bank muscled their way into this role with our taxes. I ask you to at least entertain the possibility that private bankers, committing the assets of their own firms, are likely to handle such problems better than do public officials who play this game with other people’s money.

As a rule, however, as documented in an important recent article by Prof. Charles Calomiris of Columbia University, recent IMF assistance has been “designed to absorb the losses of insolvent banks and their borrowers in developing economies, and to insulate international lenders from the losses that they would otherwise suffer.” Calomiris goes on to document three major consequences of this developing policy:

Calomiris concludes that “The principal lesson of the recent bailout programs managed by the IMF and the U.S. government…is for all parties…to find a credible way to commit not to sponsor such counterproductive bailouts.”

The characteristic IMF response to this type of criticism, of course, is that the conditions for receiving IMF credit induce the type of reforms that are necessary to avoid a future crisis. In a few cases, this has been successful. The larger record, however, does not provide a basis for optimism. Most developing country governments, once the recipient of IMF’s subsidized credit, have become loan addicts. As noted earlier, most of these governments have relied on IMF loans for more than two decades, despite the conditions for receiving these loans and the usual two‐​to‐​five year maturity of these loans.

Maybe we don’t need the IMF — that is now the judgement of former Treasury secretaries George Shultz and William Simon and the former chairman of Citicorp Walt Wriston. I am willing to defer judgement on this issue. In the meantime, Congress should not approve any additional funds for the IMF, at least until some of the broader questions are addressed.

Thank you for your attention.