Despite such arguments, governments consistently react in the same way to crises: they want someone to seem to be in charge. If the IMF were dissolved, it would almost surely be re-created when the next crisis occurs. Thus, given that the IMF exists and will certainly continue to do so, how should it be reformed?

There is now a widespread consensus about one key aspect of this reform: the IMF should be restricted to crisis management in order to limit the damage that its mistaken policies might impose. Its other functions should be given to other institutions. But limiting the IMF to crisis management clearly does not solve the problem. After all, problems in managing the recent global crisis were what precipitated the debate over reforming the international financial architecture. Thus reforms must focus on what the IMF does in response to a crisis. Three sets of reforms are crucial.

First, the IMF must become more transparent. This increased transparency would subject its policies to more critical scrutiny; it would hopefully make it more difficult for it to use outdated and inappropriate models. Part of this increased transparency would require the IMF to disclose the models it used and to predict the consequences of its policies. Outsiders could then independently ascertain the accuracy of these forecasts. In addition, if the IMF paid insufficient attention to the consequences of its programs on poverty, there would likely be a significant outcry before the damage was done.

Second, the IMF’s interventions need to be limited. Huge bailouts should be a thing of the past. In effect, the advanced industrial countries provide the funds for the developing countries to repay the developed countries’ banks, but the real burden is borne by taxpayers in the developing countries, since the IMF is almost always repaid. There should be a prima facie case that if a huge bailout is required, the exchange rate should not be at an equilibrium level.

Third, the conditions imposed by the IMF need to be reformed. Supporters of the IMF often point out that all lenders impose conditions to make sure that the funds are used as promised by borrowers. But the IMF’s conditions are different. They are not necessarily designed to enhance the likelihood that the loan will be repaid; in some cases the conditions actually have the opposite effect. In other cases, the conditions have little or nothing to do with the crisis. For example, South Korea’s crisis had nothing to do with loose monetary policy leading to excessive inflation, yet the IMF entered into the political sphere demanding reforms of Korea’s central bank to make it focus exclusively on reducing inflation and becoming more independent. Forcing Korea to move up the timing of some trade liberalization measures to which it had previously agreed had absolutely nothing to do with the crisis; it was simply a crude political power play.

Even when there are particular structural weaknesses that can be linked to a crisis, reforms must be carefully timed and paced. What would have happened had the United States gone to the IMF in the midst of its 1989 financial crisis? The IMF would have insisted on abolishing the special tax treatment of real estate and agricultural subsidies, which serve to inflate the price of land. Had it done so in the midst of the Savings and Loan debacle, the United States would have had a full-blown crisis.

The rapid abolition of tax preferences would have had a devastating effect on real-estate prices, and thereby on the entire banking system. Eventually the United States should eliminate the real estate preferences as they interfere with the productivity of the US economy, but adverse effects from the rapid elimination of these preferences would have more than outweighed any gains from improved efficiency. IMF conditions thus need to be greatly circumscribed, limited only to actions that are absolutely essential to ensure the repayment of the loan and/or to mitigate externalities, such as those associated with excessively contractionary policies.

Some reformers have argued that the IMF should be de-politicized. They observe that some of its worst lending practices, such as the loan to Russia in 1998, are based not on economic analyses (which showed that Russia had unsustainable debt dynamics and an overvalued exchange rate), but on political motivations. Such an argument, however, runs contrary to the idea that one of the key problems is the IMF’s lack of political accountability. It would be one thing if the IMF’s bureaucrats were engaged in totally technical operations. But as we have repeatedly seen, they are engaged in actions that are inherently political, whether it involves a decision to bail out international creditors at the expense of the domestic economy, or designing bankruptcy laws that are more creditor-friendly than debtor-friendly.

Beyond the Rhetoric

While markets are by and large the most effective way of increasing output and promoting growth, they often fail. Thus, within national economies, the government must often assume some role. But just at the time when the need for international economic institutions has increased, confidence in global institutions has eroded, and for good reason. Reforms must focus more on what the IMF does and how it does it. But we should also be aware that the IMF is a political organization that has survived and expanded over the past 50 years; while its adaptation may not have enabled it to better stabilize the world economy, it has enabled the IMF to survive—some say even to prosper. Thus, we should expect the rhetoric of the IMF to seriously consider such critiques. True reform needs to look beneath the surface, beneath the rhetorical veneer.