Comments: A Mixed Record for IMF Advice
- Alexander Swoboda, and Peter Kenen
- Published Date:
- December 2000
Since Takatoshi Ito has mentioned that I invented the term “Washington consensus,” let me make it clear that I did not use it in the sense that he does, as a synonym for die-hard advocacy of liberalization. My usage was in reference to the lowest common denominator of policy advice being given by the Washington institutions to Latin American countries as of 1989.
The market fundamentalism of Ronald Reagan’s first term was already being superseded by the return of rational economic policymaking, and it was becoming clear which of the ideas of the Reagan years were going to survive and which were not—monetary discipline but not monetarism, tax reform but not tax slashing, liberalization of trade and foreign direct investment but not complete freedom of capital movements, deregulation and privatization but not the elimination of prudential supervision of financial institutions or of regulations designed to preserve the environment. I acknowledge that Ito is in good company in using the term in this corrupted way, but that does not prevent me from deploring the fate that has befallen the phrase.
Box 1 shows the 10 criticisms of IMF policies that I have detected in Ito’s paper. The first five are taken directly from his second table. The next three refer to East Asia and come from the text of his paper. The last two are also to be found in his paper but relate to IMF policy in other parts of the world. I will discuss each of these 10 criticisms and express my own judgment on the degree of validity of each.
Excessive Fiscal Austerity
The East Asian countries all had strong fiscal positions prior to the outbreak of the crisis, in contrast to what has been typical in crises elsewhere in the world. This meant that there was no need to tighten fiscal policy, which only served to deepen the recession the countries had to suffer. The IMF has conceded its initial error in insisting on such tightening, as we have heard from several speakers, including Michael Mussa, during this conference. Indeed, the IMF rather quickly adjusted its policy on this issue, as Ito describes. Thus, while my judgment is that the critics’ position on this issue is valid, I also feel they make too much of it.
Box 1.Criticisms of IMF Policies
Takatoshi Ito’s List
Excessive fiscal austerity
Excessive monetary austerity
Inadequate exchange rate support
Financial sector restructuring
Inadequate restructuring of foreign financial liabilities
Capital account liberalization
Complacency over tesobonos and GKOs
Support of pegs in Russia and Brazil
Excessive Monetary Austerity
One of the leading participants in this debate is World Bank Chief Economist Joe Stiglitz, and it is also the case that his position has been exaggerated by preceding speakers, so, without wanting to get into some kind of institutional brawl, I need to set the record straight. Stiglitz has not argued that there is a conflict between a need for tight money to prevent a collapse of the exchange rate and a need for loose money to support domestic activity that should always be resolved in favor of domestic concerns. Rather, he has argued that the textbooks err in taking it for granted that higher interest rates will always and inevitably help support the exchange rate. He justifies this by pointing out that a risk-neutral investor is not concerned with the interest coupon he is promised but, rather, in the product of the interest coupon and the probability that he will actually receive the coupon rate.1 Because the latter—the probability of actually receiving payment—must be expected to decline as the interest rate rises, the impact of a higher interest rate on the exchange rate is, in principle, ambiguous. This is an intellectually interesting point that has been overlooked by the textbooks, including mine.
Could the paradoxical case, where a higher interest rate weakens rather than strengthens the exchange rate, actually arise in practice? I find this would be quite implausible when real interest rates are relatively low. One cannot rule it out, however, if the authorities jack up the interest rate to some astronomical level, which the market then sees as threatening private sector debtors (and perhaps even the government) with imminent bankruptcy if maintained for any length of time. Such a change may well be interpreted as a panic move and undermine rather than restore confidence. In other words, what we may have here is a nonlinearity, rather like the Laffer curve. If that is so, then it is critically important for us to begin to get some handle on the circumstances in which the paradoxical case is likely to arise, to make sure that we do not inadvertently push countries onto the wrong side of the “Stiglitz curve,” where a higher interest rate is counterproductive, from both exchange rate and domestic standpoints.
Inadequate Exchange Rate Support
The case where I worry most that there was a failure to provide adequate support to the exchange rate is in Indonesia. The Indonesian authorities reacted to the Thai crisis by widening their crawling band, which showed no obvious symptom of being overvalued, so as to give greater scope for market forces to absorb any shock that contagion might transmit to them in due course. For some weeks the rate stayed at the strong edge of the band, but then contagion struck and the rate was allowed to go from one side of the band to the other, and to escape from the weak edge, in a matter of hours and without any attempt to mount a defense. After that there was a rush to cover foreign exchange debts, which then further weakened the rupiah, which then sparked fears of political collapse, which then ignited capital flight, which then caused the economy to collapse. It seems to me that, by far, the best chance of short-circuiting this disastrous chain of events would have been to mount a robust defense of the band. (I am not saying that this would have been sure to succeed, for the element of truth in the fad about hollowing out exchange rate options that has been so popular during this conference2 is that intermediate regimes are more vulnerable to speculative attacks than either currency boards or floating rates. I am arguing that the decision not to mount a defense sealed Indonesia’s fate.)
Financial Sector Restructuring
The art of financial sector restructuring is to introduce a package of measures that is accepted by the public as definitive. In general, one expects that some institutions will have to be closed or merged, and there is no need to ensure that the depositors in or lenders to those institutions will remain whole, but the public needs to be credibly persuaded that those that remain open the day after will remain open for the duration, and that there is no danger of their suffering losses if they leave their deposits where they are. This will involve either recapitalization of the banks that remain open, or the extension of explicit or implicit deposit insurance to deposits in them. In none of the crisis countries of East Asia was there a credibly definitive financial restructuring such as was called for. Instead, a number of finance companies or banks were closed, and the public seems to have been left with the impression that this was a first installment. I find it difficult to believe that the IMF staff were unaware that this was not a best-practice way of dealing with the problem, but, if they were so ignorant, it is a pity that there was not a more active communication between the World Bank and IMF during the critical months of late 1997. There certainly were people not far away for whom what I have set out above was not news.
Ito is critical of the IMF for having exploited the crisis to get countries to implement structural reforms that had little to do with securing a quick exit from crisis. Let me acknowledge that I can understand the temptation to use such an occasion to press for long-needed reforms. Ever since Bolivia decided to use the need to stabilize its hyperinflation in 1985 as an occasion to introduce a bold program of structural reform (an initiative that so impressed Jeffrey Sachs that he subsequently persuaded Poland to do a similar “big bang” reform in 1989), many reformers have regarded crises as providing an opportune occasion to press their cause. There is a world of difference, however, between the national government itself deciding that it wants to exploit such an opportunity and an international organization deciding that it will use its moment of maximum leverage to press for reforms. The difference lies in local ownership of the reform program in the first case and absence of such in the second case. In the absence of ownership, the temptation to press for reform should be resisted.
We have in recent years gotten so used to the notion that greater transparency is a good thing that I fear I will cause shock by suggesting that it can be too much of a good thing. I can recall life in the British Treasury in 1968, when we were fighting to prevent a second sterling devaluation that the market was baying for but that would have made no economic sense. At the time, the figures of U.K. reserves published at the end of each month benefited from a generous measure of creative accounting. Had we suddenly come clean in 1968, I doubt if there would have been any chance of staving off the dreaded second devaluation. Of course it is good that the creative accounting has long since been abandoned, but I believe it was a sensible decision to delay honesty until such time that it would not precipitate a crisis. In the same way, I do not think it was wise of the IMF to force Thailand to reveal in October 1997 that it had mortgaged all its reserves. Honest accounting, like capital adequacy, is something to be done after a crisis is safely past and then maintained in the future, not something to be rushed into in the midst of a crisis.
Inadequate Restructuring of Foreign Financial Liabilities
There is criticism that the debtor countries should have been assisted in rolling over their foreign liabilities (and perhaps in making debt-to-equity conversions) sooner than occurred; for example in Korea, rollover should have occurred in the first rather than the third week of December 1997. Ex post, it is easy to endorse this view, and we have indeed heard endorsement of it by members of the IMF staff, including Michael Mussa, during this conference. Ex post, I endorse it too, but honesty compels me to say that I am not sure I would have pushed it ex ante, for the reason laid out so starkly by Martin Wolf in the preceding session. He argued that one needs to make a definitive decision in a crisis between providing enough liquidity to ward off a crisis (the Bagehot approach) on the one hand, and restructuring liabilities on the other. An intermediate solution, first giving a financial package and then, if it doesn’t work, restructuring the debts, promises the worst of both worlds, both encouraging moral hazard before the crisis and then guaranteeing that the crisis will actually materialize. Here, I suspect, we have a case where the intermediate solution really makes no sense.
Capital Account Liberalization
I have had a long-standing disagreement with the IMF on capital account liberalization. On a first approximation, one can say that the East Asian crisis was caused by vastly premature liberalization of the capital account, something that the IMF had been pushing right up to, and even after the outbreak of, the crisis. Indeed, it seems that, at least in the case of Korea, the IMF embodied a requirement for further capital account liberalization in the program its client was expected to sign. I am happy to note that, from what we have heard during this conference, the IMF now seems to have backed off from this misguided cause. I would hope that in the future it will not merely tolerate measures to ward off excessive capital inflows of the type adopted by Chile and Colombia, but will actively encourage all emerging market members threatened by too much short-term money to adopt measures of this general type.
Complacency Over Tesobonos and GKOs
Ito charges the IMF with having complacently watched the buildup of tesobono liabilities in Mexico in 1994 and of GKOs in Russia in 1998. At least in the case of Mexico, this charge appears to be valid.
Support of Exchange Rate Pegs in Russia and Brazil
Ito’s final charge is that, in strong contrast to IMF pressure on Asian countries to float their currencies, it supported currencies pegged at overvalued exchange rates in both Russia and Brazil. It is clear in retrospect (even if some of us were not following events in Russia closely enough to have recognized it at the time) that the ruble was overvalued enough to have given Russia an acute case of Dutch disease. And it had been clear for some time that the Brazilian real was overvalued. In the latter case, however, I am not sure that the IMF had much choice but to support the Brazilian authorities in their chosen strategy in the fall of 1998, when the world financial crisis appeared genuinely threatening, and the Brazilian choice was to try and tough it out with a quasi-fixed exchange rate. That strategy bought a few months of extra time for the world financial system to calm down, and my judgment is that those few months proved invaluable.
The IMF’s record, in my assessment, is mixed. In a number of contexts I have argued that it does not merit the criticism that it has been receiving. It is a relief also to hear from Mohsin Khan that, in general, the evidence shows that countries do better, other things being equal, with IMF programs than without them. It is even more of a relief to have just heard from Kiettisak Meecharoen that Thailand is now on the road to recovery. But, at the end of the day, I have to say that I agree with Ito that this was not the IMF’s finest hour.
The argument is reinforced if investors are risk-averse rather than risk-neutral.
My guess is that this fad will prove short-lived as countries discover that floating rates preclude the sort of sustained export-led growth that was at the core of the East Asian miracle that preceded the crisis.