Institutional reform and sovereign debt crises.

Rajan, Raghuram G.

In this brief article, I want to offer some ideas for reforms of the International Monetary Fund (IMF). This subject is complicated, and it is very difficult to do justice to in the space available, but I will try. Let us start with a brief review of what the IMF does.

Surveillance, Lending, and Technical Assistance

The IMF is mainly concerned with three activities: surveillance, crisis prevention/crisis resolution, and technical assistance. I will leave out technical assistance, as that speaks for itself, and focus on the other two functions. Surveillance has always been an important motivation for the IMF--even if a country has the best economic expertise at its command, its policies affect other countries, and there is a need for someone to speak out if these policies create problems for other countries. Traditionally, the Fund's policy focus, in the Bretton Woods era, was in the area of exchange rates and trade, primarily because of the destructive consequences of the policies followed during the 1930s and the attempt to avoid repeating those mistakes. Today, even though many of our members have become sophisticated and do not depend on our macroeconomic policy advice to the extent they did in the past, surveillance continues to be an important function of the Fund because the world is so much more interconnected through trade and capital flows.

Even the domestic policies of countries have large international effects, especially if the countries are important players in the global economy. Developed countries have a big responsibility, not just in helping to solve crises, but also in helping to prevent them. After all, movements in developed countries' interest rates, and the attendant movements of capital, are often the starting point for crises in developing countries. The IMF can play a role in drawing attention to the international consequences of developed country policies, and in helping to improve them.

A big problem, of course, is that the IMF can speak, but others need not listen. When we go to large developed countries and tell them, "Your fiscal deficit is 5 percent of GDP and increasing, and it needs fixing," they are apt to say, "Thank you very much. We know it. What's new?" Certainly, for developed countries, the Fund has minor influence unless it can use leverage. One source of leverage is peer pressure--being an impartial arbiter, the Fund can point fingers without appearing biased, and it has the weight of the community of nations behind it in pressing upon a country to alter its policies. A second potential source of leverage is the cross-country expertise the Fund brings to the table. A few countries have sophisticated analysts and researchers who can draw lessons from international experience for themselves. For the rest, the Fund can distill the lessons from its varied experience and still add value through its advice.

That said, both these sources of leverage are increasingly threatened. Private analysts are quite capable of reaching impartial assessments of a country's macroeconomic policies. Also, the number of bilateral, regional, and multilateral venues in which experience is shared is multiplying. It is therefore essential for the IMF to stay ahead of the curve by being perceived as being more impartial and more expert than the alternatives. Going forward, it is important that the Fund continue to recruit high quality economists, maintain a high degree of independence, as well as improve the quality of its analysis and its information base, so that it continue to remain relevant.

Surveillance, of course, is closely linked to the other main function we perform, which is crisis prevention and crisis resolution. (Indeed, one of the aims of surveillance is crisis prevention.) A key purpose of the IMF in the era when private capital flows were limited was essentially to assist countries that did not have market access to weather temporary adverse shocks without the need to incur tremendous costs in adjustment. Countries essentially faced a flow problem--a temporary fall in a country's exports and a deterioration in its balance of payments, for example, compounded by profligate budgets, created an excessive current account deficit. The IMF came in with an adjustment program that would help curtail the current account imbalance, and with financing to tide the country over the adjustment process. The IMF was ideal for such temporary finance--a common pool of reserves that could be used by countries to finance short-term needs on a revolving basis.

As countries have started being able to access private capital markets, a new form of crisis has started manifesting itself--a complete reversal of private capital flows into a country (this is a slight exaggeration-even the old balance-of-payment problems were associated with capital outflows, but not to the extent we now see). Overnight, a country becomes uncreditworthy. This is a stock problem. The cumulative stock of external capital that has built up over many years suddenly takes fright, and often flight. Not only does the Fund have to make up the flow shortfall caused by a temporary need, it also is called upon to finance the exit of the entire stock. The scale of the financing that the Fund can be called on to do nowadays is far larger than what it used to be. The size of the Fund, determined by its quotas, is more a reflection of the old quotas that were set up when countries primarily faced a flow problem and is now inadequate in many situations.

Dealing with Crises

One way countries might think of bulletproofing themselves against crises is to build foreign exchange reserves. Reserves are expensive to maintain. Furthermore, to the extent that reserves are meant to guard against individual shocks, it would take far less to maintain a pooled reserve across countries than for each country to build its own massive individual stock. The IMF is meant to be the pooled reserve, available for the truly needy on demand, and thus obviating the need for individual stockpiles. That some of the poorer countries of the world are among those building enormous international reserves is an indictment of the international financial system. It is also against the natural scheme of things, with poor countries financing rich countries.

Be that as it may, it simply is not feasible or cost-effective for emerging markets to build up enough reserves to vanquish crisis. What about creating a debt structure that poses little risk--for example, borrowing long term in their own currency? This approach also is infeasible in the near term. Lenders like foreign currency-denominated short-term claims on a country because they do not want to be entirely at the mercy of its policies. Lenders are given a tremendous amount of ability to inflict damage not just because the country is gaming the system or the finance minister is irresponsible, but also because the country itself has institutional fragilities and infirmities that force it to borrow in precarious ways.

All this means that emerging markets cannot simply banish crises through sensible policies. Ultimately, they have to undertake institutional reform, but that can take time. In the interim, when the country faces a shock, and if it is unaided, it could face tremendous destruction in value, closure of firms, loss of jobs, a deep recession, and so on, and, essentially, the IMF will be called on to fill in the gap.

Three Kinds of Crises: But Which Is Which?

It is possible to identify at least three kinds of capital account crises. The first is a pure liquidity shock--when the government needs to be tided over a period after it has lost the confidence of international capital markets, perhaps as a result of contagion. The second is a solvency shock--when something hits a country making its level of debt simply unsustainable. The third, and most intriguing, is a conditional solvency shock--when the country needs to undertake structural reforms to maintain its solvency in the face of the shock, but would be solvent, and would regain the confidence of international capital markets, conditional on undertaking those reforms.

Most economists would agree there is a role for Fund intervention in liquidity shocks, and a need for the country to restructure its debts when it faces a solvency shock. There is room for debate on what the Fund should do in the case of a conditional solvency shock. To the extent that default is extremely costly for a country, and to the extent that the Fund has some ability to help a process of structural reform, there is a greater case for Fund intervention in those cases.

The problem, however, is that we have spoken of liquidity versus solvency as if it is easy to tell one apart from the other. In truth, it is very difficult except in the truly extreme cases. In fact, I will venture to say that you only know after the fact what kind of crisis you were dealing with.

For example, South Korea in 1997-98, and Thailand to a lesser extent, turned out to be liquidity crises. Lenders decided to take flight--they evaporated--and the government put together a program along with international financial institutions. Confidence was restored within a relatively short period. These were crises, most people would agree, where it made sense for international lenders to come in to tide over the government, in order to prevent tremendous destruction.

Yet Brazil and Turkey in recent times did not turn out to be mere liquidity crises--they have been more prolonged, and it has taken substantial reform for markets to gain confidence. Yet during those crises, it is not obvious that the signs distinguishing them from South Korea or Thailand were flashing boldly for all to see. And the examples of the Fund lending in cases when a country should, in hindsight, have been asked to restructure its debts, are not impossible to find.

The problem is that the objective conditions that face the Fund in each of these cases is the same--a loss of access to international capital markets. In the midst of a crisis, it is simply impossible to do the kind of detailed ex post evaluation that allows us to identify the nature of the crisis. So what is one to do?

Clearly, the answer depends on what the costs of excessive intervention are. One is that a country adds the cost of a "rescue," including perhaps a bank bailout, to an already unsustainable debt. Then the country totters for a few more years before it enters a fresh crisis. In the meantime, the citizenry pays the price with low growth and high taxes. The second is that moral hazard is encouraged--not just the traditional versions where investors get complacent (the evidence is very mixed on this) or governments overborrow with citizens suffering the pain--but also the possibility that the domestic interest groups may have too little incentive to compromise on inflated budgets, excessive wage demands, or inefficient monopolies if they know the country will not be allowed to fall off a cliff. The point is that far from reducing the overall pain borne by countries, excessive intervention may enhance it.

But why would a member country agree to a rescue if it leads to only short-term relief but longer term pain for the country? The adage "any port in a storm" is probably much of the explanation. But my colleagues at the IMF Research department, Olivier Jeanne and Jeromin Zettelmeyer (2001), have proposed another. Bailouts may help governments shift the burden of a crisis off the shoulders of the domestic business elite on to domestic taxpayers. To the extent that the latter do not have an adequate voice, they bear the brunt of excessive intervention, and everyone else who has a voice is willing to go along.

To summarize then, in capital account crises, the Fund should intervene primarily in situations where the problem is temporary illiquidity or where there is a significant possibility of rapid structural change. But if it is hard to tell, even in the midst of a crisis, whether the underlying factors are temporary or more structural, the facts that emerge are unlikely to help decisionmaking. In such a case, discretion can only be harmful as it exposes the Fund to internal and external pressures to intervene. Not only will the decision be biased as a result of discretion, it also will be noisy because it will not be based on underlying fundamentals. Bias and unpredictability are not the best attributes of sound decisionmaking.

Suggestions for Reform

I think it is very important to create steady pressure for domestic change. Here is where I agree entirely with Allan Meltzer that some form of linking the safety net to domestic policies would be useful. For instance, access to Fund lending could be tied to a country's policies and reforms in normal times, as suggested by Jeanne and Zettelmeyer (2001). If a country follows sound policies and undertakes needed reforms there should be a presumption that if it faces a crisis, it is likely to be a liquidity crisis, or a solvency problem (such as a permanent terms of trade shock) that is not of its own snaking. The Fund should intervene in the former, and will be providing insurance in the latter case (with the country then snaking needed adjustments on its own accord), not an entirely bad use of Fund resources.

Offering a contingent stream of lending, based on past policy choices, would also provide incentives for continual improvement in policies. Moreover, if this approach was also applied to rich countries so they would get a grade on their policy report card that determined their access to credit, their incentives would change. Rich countries might not need to draw upon IMF credit, but the report card would be a summary judgment of the countries' policies. If the finance minister saw a poor evaluation on his country's report card, regardless of that country's need for IMF resources, that grade would still be important--for it can clearly be communicated to the press and to domestic policy circles. It is not just that this process would make naming and shaming of flagrant abusers of the international system more effective. Equally important, it follows the principle of universality: all countries are subject to this access limit and evaluation. Moreover, it makes the IMF a little more responsible for its evaluation if countries can stand up and say, "You gave me a terrible rating and I want you to justify why."

Clearly, the IMF would have an incentive to do a great job on the surveillance because, at the end of it, if the Fund did not come up with a justifiable rating, either the markets or the country would be upset. Of course, some would argue that the process would become completely uninformative, but if it is to be the basis for access to Fund resources, then the Fund would have every incentive to maintain its credibility.

Another reform to ensure that conditional solvency can be separated from insolvency, at least after the initial intervention, is to have some kind of a firm end-date in mind at the beginning of that process. It would be all too easy for the Fund to make the initial loan, see reforms half done, and then say, "Okay, we're going to lend some more because it is not done. If we leave now, the markets will take flight." On the other hand, if the Fund says well in advance, "We are going to leave by this date, and no more lending beyond that," it specifies an end-date, which will put pressure on the country to reform in a timely fashion because the country will eventually and predictably lose the safety net. This approach is similar to the cleanup clauses banks have in their lines of credit whereby clients have to clean up a loan by a certain date.

The final aspect that 'all this implies is that you also have to fix the governance process. If the Fund has a system of lending that is so intrusive, in the sense of coming into countries that are not in the program, but also evaluating their policies and putting a report card out there for all to see, there has to be trust among the members that this is not being done at the behest of one or the other large shareholder. This means that the governance has to be a little more independent of the political structures in the large member countries, while recognizing that the Fund cannot be fully independent--it is relying on taxpayers in these important countries, so they have a right to representation. The way to solve this, I think, is to get a little more oversight, but at longer intervals. That is, the large shareholders should be able to elect a board, but the board should be independent after that. Central banks stay more independent of their domestic government if day-to-day activities are not driven by the shareholders. If shareholders have the right to appoint only at the end of a predetermined period, it ensures that governors or directors have independence from the governments but do not have complete license. I think that such a governance system would lend greater credibility to the Fund.

Finally, I think the IMF has to recognize that economic situations of countries around the world have changed. The "old" representation that was in place in the past does not reflect today's realities. Therefore, the governance structure has to change. Also, changing the governance structure would be a way for the Fund to draw in new money, if need be, for carrying out its major functions. A number of countries now have both the need for insurance as well as the resources to create a common pool. Rather than the Fund continuing to rely on its old shareholders who have little need for Fund resources, as well as significant budgetary problems at home that cause them to question any augmentation of the Fund, the IMF could find ways to draw in resources from the willing. This might mean finding innovative ways to achieve necessary separations (and new links) between voting power, contribution of resources, and even control over some resources.

In sum, there is much to debate about the role of the IMF going forward, and I have offered some thoughts to stir the pot. I personally do not believe that the need for an organization like the IMF will diminish. I do not think it is a problem that the Fund has taken on new functions--which to a large extent have been thrust upon it by the membership--as some old ones have become less important. This has not been mission creep as much as mission push and the evolution of a learning, adaptive organization. It would also be a tremendous waste of the talented human capital that is now embedded in that organization if it were allowed to atrophy. Instead, we should debate reform of the Fund vigorously so that it can continue to play a vital role in the international arena. Discussions like the one the Cato Institute has organized are tremendously important in this process.

Reference

Jeanne, O., and Zettelmeyer, J. (2001) "International Bailouts, Moral Hazard and Conditionality." Economic Policy 16 (33): 409-32.

Raghuram Rajan is Economic Counselor and Director of Research at the International Monetary Fund. The views expressed in this article, which is based on remarks delivered at the Cato Institute's 22nd Annual Monetary Conference on October 14, 2004, are his own and do not necessarily reflect those of the IMF, its management, or its staff.