by John Williamson, The World Bank
Paper presented in Dhaka, Bangladesh
© Peterson Institute for International Economics
This paper was written while Mr. Williamson was the Chief Economist for the South Asia Region at the World Bank.
For almost two decades East Asia was held up as a role model for developing countries, by economists in general and by those from the World Bank in particular. Since last July the region has been engulfed by a severe economic crisis, and innumerable sages have emerged from the woodwork to proclaim that the miracle was really a mirage, or that the region's problems are the penalty it has to pay for deviating from free market orthodoxy, or that the crisis is a deliberate act of sabotage of the region's progress by the West. The purpose of the present paper is to offer an interpretation of the crisis that will allow us to assess the validity of such claims, and then proceed to ask how other developing countries, particularly those of South Asia, should respond to the crisis.
A Crisis that Surprised
No one foresaw a crisis of the character and severity that has engulfed East Asia. True, some of us worried that Thailand was taking a risk in committing itself to the defence of a fixed exchange rate (defined in terms of a single currency, which resulted in large cycles in the effective exchange rate as the dollar went on its roller-coaster) given that a massive current account deficit had developed, but we never imagined that the ensuing crisis might result in a devaluation of almost 50 percent, or a fall in the stock market of over 40 percent, or that output would decline by 8 per cent. Still less did it occur to us that a Thai crisis would result in rapid and widespread contagion. And if we had asked ourselves who would be likely to suffer from any contagion that might occur, we surely would not have selected as the likely victims the other East Asian countries, who boasted just about the strongest "fundamentals" in the world, at least as we have been accustomed to thinking of the fundamentals (in terms of macroeconomic variables like the fiscal balance, inflation, savings, investment, and growth).
Common Features of Victims
For several months I think most of us were quite unable to detect a pattern in which countries fell victim to the crisis. We were relieved that no South Asian country succumbed in a serious way; we were also surprised because all the countries of South Asia are weaker than those of East Asia in terms of what we have been accustomed to thinking of as the fundamentals. But gradually a pattern emerged. With one partial exception, all the victims had three things in common:
It is not too difficult to understand why a country with those three features should suffer contagion from the Thai crisis (whose initiation, I have argued, can be understood as a conventional exchange rate crisis). The Thai crisis raised questions about the sustainability of asset prices in any country whose economic situation appeared to resemble that of Thailand. Similarly, the Thai demonstration that exchange rates could depreciate severely caused a rush for dollars on the part of those with uncovered positions in foreign exchange, especially if these were short-term, which rapidly precipitated the very depreciation that agents were trying to insure themselves against. This depreciation weakened the balance sheet position of most companies with foreign exchange exposure (the exception being firms heavily engaged in exporting), which reinforced the asset price declines, and thus further undermined the solvency of the financial sector. These crises have been dubbed "new-style crises"2, in contrast to old-style or conventional balance of payments/exchange rate crises.
Countries (such as those of South Asia) that exhibited only the last of the three features listed above, namely a weak banking system, were not vulnerable to a new-style crisis. But it would be a mistake to draw too much comfort from this; they remain vulnerable to old-style crises, and indeed their vulnerability to such a crisis may well have been increased by events in East Asia, notably the devaluations, which have reduced the competitiveness of South Asian exports.
Explaining East Asian Vulnerbility
To say that the countries engulfed by the crisis were those with high asset prices, that had heavy exposure to the international financial markets, and that had weak banking systems immediately poses the question: how did countries get into this situation? It seems to me that there were at least three policy mistakes that contributed to creating the vulnerability in most or all of the victim countries.
The first, and the one highlighted in the most elegant piece of theorizing to have been spawned by the crisis so far (Krugman 1998), is the provision of explicit or implicit guarantees of the liabilities of financial intermediaries without the creation of a system of financial regulation and supervision adequate to restrain the moral hazard which such guarantees create. If the creditors of financial intermediaries believe they have no need to check the solvency of those to whom they are entrusting their assets, while the owners of the financial institutions believe that their losses will be socialized while they will reap the full benefits of the profits the institutions may make, they will have an incentive to undertake investments that are excessively risky from a social point of view. In the absence of a functioning system of regulation and supervision to restrain that moral hazard, the banks can be expected to acquire too many risky assets. That temptation may be expected to be particularly strong in the aftermath of financial liberalization, when bankers have been told it is their social duty to lend to those whom they judge it advantageous to lend to, and when they have not yet had the experience of a collapse to warn them against imprudent lending.
Guarantees without adequate regulation and supervision might go a long way toward explaining the weakness of the banking system, and they might go some way toward explaining how the prices of risky assets began to boom, but they certainly cannot explain how countries acquired a large exposure to the international financial market. To explain that, one needs also to recognize the second policy error, which was premature liberalization of the capital account of the balance of payments. It was this which allowed foreign as well as domestic wealth-holders to respond to the attractive (because presumed to be largely risk-free) returns offered by financial intermediaries, and thus provide the financial resources that in many cases powered asset price booms—booms that for a long time seemed to validate the excessively risky lending that was being undertaken. And in the process, of course, the countries acquired their large exposure to the international markets. I take the East Asian meltdown as decisive evidence showing the wisdom of the Tarrapore Committee (1997) in having argued that Indian capital account convertibility needs to be preceded by the establishment of a liberalized, well-supervised, domestic financial system.3 And I concede that a part of the blame for the East Asian crisis can legitimately be laid at the door of those in the West (including in the international organizations) who pushed for rapid, unconditional, and complete liberalization of capital account transactions.
While one can offer a complete explanation of the East Asian crisis in terms of those two policy errors, there is surely a third factor that contributed to the weakening of the banking systems in most or all of the victim countries. This is "crony capitalism", the favouring of particular enterprises controlled by the friends or relatives of the politically powerful. One of the ways in which the cronies received favors was in terms of access to credit, which weakened the banks when the cronies either were unable to repay or chose not to repay because they knew that their political patrons would defend them against the sanctions that are normally imposed on defaulters.
Interpreting the Crisis
If this is indeed how the East Asian meltdown developed, what does that imply for the sorts of claims mentioned at the beginning of this paper? Was the miracle a mirage? Was the disaster a consequence of deviating from the free market? Did the West sabotage the miracle?
Surely the miracle was no mirage. Per capita incomes rose faster in Korea and Thailand between 1970 and 1995 than in any other countries that started with a comparable level of income (see Table1). According to one estimate, poverty in Indonesia fell from some 58 percent in 1971 to just 8 percent in the early 1990s, surely the only time in history that a country has lifted half of its population out of poverty in a couple of decades, and dramatically superior to South Asia's reduction of perhaps 20 percentage points (from a similar figure to that of Indonesia in 1971 to perhaps 35 percent in the early 1990s). Suppose that poverty doubles in Indonesia as a consequence of the crisis, which is surely an extreme assumption. It would still be less than half that in South Asia, even though poverty rates in the two regions were similar a quarter century ago. The miracle involved building productive assets and teaching people how to make good use of them, and those investments and skills are not going to evaporate because of a financial crisis, no matter how severe. The miracle was real, and the people of those countries will again benefit from the assets that were built up when the crisis passes, as it will.
Can one explain the collapse by governments forcing their populace to save or invest or export too much, or otherwise distorting the market? One can argue that there was some over-saving, especially in Singapore—but Singapore is hardly at the eye of the storm. On the contrary, its over-saving has enabled it to build a financial position so strong that it was able to weather the storm with rather little damage. Advocates of the over-investment thesis seem unable to agree whether the problem was too much investment in nontradables (golf courses and shopping centres) that would not yield foreign exchange to service the debts that had been taken on, or too much investment in tradables so that they were satiating the world market. The latter version is quite implausible: most of the countries were importing more than they were exporting, and thus supporting rather than detracting from world demand. The former version is much more persuasive—but it amounts to saying the governments had let their former dedication to macroeconomic discipline, notably a competitive exchange rate, slip. And yes, crony capitalism had introduced distortions.
Did the West sabotage the miracle because it was jealous of the progress being made by a bunch of upstarts? The saboteurs were supposedly the speculators, who are presumed to have been responsible for causing the devaluations that initiated the crisis. No one can reasonably deny that the speculators were the proximate cause of the Thai devaluation (though it is not at all clear that they were responsible for the subsequent devaluations). The question is whether one believes that speculators act in response to a perception that prices (in this case exchange rates) are out of line with the fundamentals, thus presenting them with an opportunity to profit when the authorities are forced to face the facts, or whether they have the ability to make money by forcing a price away from its equilibrium level. This was the contention that Milton Friedman (1953) first challenged with his attempt to demonstrate that destabilizing speculation must be unprofitable (you make money by buying low and selling high, not vice versa). While I am among those who do not find this argument definitive in the context of a floating exchange rate (Williamson 1993), basically because of the possibility that stop-loss traders provide a source that can be milched almost systematically by traders who follow chartist rules (Krugman and Miller 1992), the equivalent argument in the context of a managed exchange rate system relies on governments being forced by the speculators to change their macro policies in a way that will validate the speculation (Obstfeld 1995). So far attempts to find instances where speculators have induced validating changes in government policies have failed. (I see not a shadow of evidence to support any wider charge that the West in general consciously set out to disrupt the economic progress being made in East Asia; on the contrary, most Western economists fell over themselves in trying to prove that the economic progress that they welcomed and admired was a consequence of the East Asian countries adopting their particular policy preferences.)
Is there any alternative interpretation at a similar level of generality that is suggested by the analysis in the preceding sections? I propose that we think of these countries as having been seduced by their own success into overconfidence and thence carelessness. They had lived so long without a macro crisis that they had forgotten the penalties of letting the exchange rate become uncompetitive, as well as the necessity of reacting with overwhelming force and determination when introducing a policy package designed to resolve a crisis. (Contrast Brazil's reaction to a speculative run last November with the initial dilly dallying in Thailand, Indonesia, and Korea.) They had enjoyed such a long boom that they mistook a speculative bubble in the asset markets for a natural market reward for success. They had profited by liberalizing their economies, and did not realise the lesson learned in South America in the early 1980s, that a necessary counterpart to financial liberalization has to be the institution of adequate regulation and supervision. The West told them that liberalizing international capital flows was a part of growing up, and they took the advice without questioning. Their poor had benefited spectacularly despite the favours bestowed on the cronies, so it did not seem particularly dangerous to allow the cronies ever bigger favours.
I shall frame my lessons with a particular view to how the South Asian countries should respond, although most of the things that I say apply also to other developing, or even in some cases developed, countries.
Avoidance of an Old-style Crisis. Since the South Asian countries exhibit only one of the three factors that precipitated the new-style crises in East Asia, I do not see much reason to fear similar crises in South Asia. However, the danger of an old-style crisis, one in which a balance of payments deficit precipitates a run on the currency, has been increased by the troubles of East Asia. The major factor here is the increased competitive pressure from East Asian countries following their massive devaluations, reinforced by the fall in demand for South Asian exports caused by the recessions in East Asia but somewhat attenuated by the weakening in the prices of primary commodities (notably oil) which are imported by both East and South Asia. The appropriate policy response is a tightening of macro policy, notably a more austere fiscal policy (including a reduction in quasi-fiscal deficits) and, if not necessarily an immediate depreciation, then certainly a greater willingness to allow the exchange rate to respond promptly to signs of a deterioration in the external accounts.
Continue to Pursue a South Asian Miracle. Just about every country in South Asia has in recent years set itself the goal of a major acceleration in its growth rate, inspired by the example of East Asia and its demonstration that rapid economic progress is possible if the policy stance is appropriate. Perhaps the most tragic consequence of the East Asian crisis would be if it dissuaded other countries from learning from the truly impressive achievements of East Asia over the past 30 years or so. There is no reason to believe that the policies that most economists hold responsible for nurturing the miracle of high and sustained growth with a reasonably egalitarian distribution of its fruits—macroeconomic stability, until 1997; high savings rates (aided by an early demographic transition); investment in education, first at the primary level; providing the private sector with a reasonably level playing field; export orientation; neutrality on industry versus agriculture; or even industrial policy (if you believe that was a factor)—were those that induced the crisis. On the contrary, it was the recent departures from those policies—the risks with macro stability involved in borrowing too much abroad and stabilizing the exchange rate rather than making sure it remains competitive, the failure to follow through from the successes in primary education by equivalent attention to secondary education, the violations of neutrality involved in growing preferences to the cronies—plus the sudden embrace of capital account convertibility, that set the stage for the crisis. With the exception of the growth of crony capitalism (and the neglect of the environment, though that is another story), pre-1990 East Asia remains a role model to be emulated.
Preempt New-style Crises. While achieving a couple of decades of East Asian rates of growth should be the primary objective of economic policy in South Asia, it is obviously desirable to avoid that growth being interrupted at some point in the future by the emergence of a new-style crisis. This demands avoiding those mistakes bred of over-confidence that have laid East Asia low in recent months. First, while it is probably undesirable to avoid all guarantees of bank liabilities, it is quite essential that these be accompanied by good regulations (requiring transparent accounting, capital requirements at least up to the international norm, prohibitions on connected lending, a legal system that exacts sanctions if loans are not serviced conscientiously, etc.) which are enforced by a competent and motivated body of supervisors. Second, it suggests that countries should eschew dogmatic commitments to capital account convertibility, and retain some ability to require financial actors to maintain a reasonably balanced currency position, and to make it expensive for domestic agents to borrow abroad short-term (as Chile and Colombia do). Third, it argues the need to abandon crony capitalism once and for all.
When this paper was written, in February 1998, I had started to hope that the countries of East Asia were beginning to emerge from the crisis. That hope proved premature, especially with regard to Indonesia, but as the paper is revised (in October 1998) the hope looks more secure, unless at least the crisis goes into a second round as it ripples further around the world. Even so, the costs of the crisis have been so massive that it is of obvious importance for other countries to learn how to avoid similar disasters in the future. South Asia certainly needs to learn that—but it needs even more to learn how to grow in the way that East Asia did for the preceding quarter century.
Friedman, M. (1953) Essays in Positive Economics. (Chicago: University of Chicago Press).
Krugman, P. (1998) ‘What Happened to Asia?', MIT (http://web.mit.edu/krugman/www/DISINTER.html).
Krugman, P, and Miller, M. (1992) ‘Why Have a Target Zone?', Carnegie- Rochester Conference Series on Public Policy 38 (June), pp. 279-314.
Obstfeld, M. (1995) ‘International Currency Experience: New Lessons and Lessons Relearned', Brookings Papers on Economic Activity 1, pp.119-211.
Tarrapore Committee (1997) Report of the Committee on Capital Account Convertibility (Mumbai: Reserve Bank of India).
Williamson, J. (1993) ‘Exchange Rate Management', Economic Journal, vol. 103, no. 1, pp. 188-97.
1. The partial exception is Hong Kong, which had an exceptionally strong banking system. However, I call the exception only partial (a) because agents might have worried as to whether the banks would remain solvent after the collapse of the property price bubble, given the extent of the banks' exposure to property lending, and (b) because Hong Kong has resisted rather than succumbed to the speculative pressures that were the proximate cause of the devastation elsewhere in the region.
3. If there is any conflict between the preconditions laid down by the Tarrapore Committee and the very ambitious timetable (3 years) for moving to capital account convertibility that they mentioned, it is surely the preconditions that should be decisive.