Globalization and Its Discontents Revisited

by Joseph E. Stiglitz

  If enough firms fail to repay their loans, banks may even collapse. A collapse of even a single large bank can have disastrous consequences. Financial institutions determine creditworthiness. This information is highly specific, cannot easily be transmitted, and is embedded in the records and institutional memory of the bank (or other financial institution). When a bank goes out of business, much of the creditworthiness information it has on its borrowers is destroyed, and that information is expensive to recreate. Even in more advanced countries, a typical small or medium-sized enterprise may obtain credit from at most two or three banks. When a bank goes out of business in good times, many of its customers will have difficulty finding an alternative supplier of credit overnight. In developing countries, where sources of finance are even more limited, if the bank that a business relies upon fails, finding a new source of funds—especially during an economic downturn—may be nearly impossible.

  Fears of this vicious circle have induced governments throughout the world to strengthen their financial systems through prudent regulation. Repeatedly, free marketeers have bridled against these regulations. When their voices have been heeded the consequences have been disastrous, whether in Chile in 1982–83, in which Chilean gross domestic product fell by 13.7 percent and one in five workers was unemployed, or the United States in the Reagan era, where, as we noted earlier, deregulation led to the savings-and-loan debacle, costing American taxpayers more than $200 billion.

  A recognition of the importance of maintaining credit flows has similarly guided policy makers in trying to deal with the problems of financial restructuring. Fears about the adverse effects of this “destruction of informational capital” partially explain why the United States, during the S&L debacle, closed down very few banks outright. Most of the weak banks were taken over by or merged into other banks, and customers hardly knew of the switch. In this way, the information capital was preserved. Even so, the S&L crisis was an important contributing factor to the 1991 recession.

  Inducing a Bank Run

  Although financial system weaknesses were far more pervasive in East Asia than in the United States, and the IMF’s rhetoric continually focused on these weaknesses as underlying the East Asia crisis, the IMF failed to understand how financial markets work and their impact on the rest of the economy. Its crude macromodels never embraced a broad picture of financial markets at the aggregate level, but were even more deficient at the microlevel—that is, at the level of the firm. The Fund did not adequately take into account the corporate and financial distress to which its so-called stabilization policies, including the high interest rates, contributed so strongly.

  As they approached the problem of restructuring, IMF teams in East Asia focused on shutting down weak banks; it was as if they had a Darwinian model of competition in mind, so the weak banks must not survive. There was some basis for their position. Elsewhere, allowing weak banks to continue to operate without tight supervision resulted in their making highly risky loans. They gambled by making high-risk, high-return loans—if they were lucky, the loans would be repaid, and the higher interest rates would bring them back to solvency. If they were unlucky, they would go out of business—with the government picking up the pieces—but that is what would happen to them in any case if they did not embark on the risky loan strategy. But too often, such risky loans indeed turn out to be bad loans, and when the day of reckoning comes, the government faces an even bigger bailout than if the bank had been shut down earlier. This was one of the lessons that had emerged so clearly from the U.S. savings-and-loan debacle: the refusal of the Reagan administration to deal with the problem for years meant that when the crisis could no longer be ignored, the cost to the taxpayer was far larger. But the IMF overlooked another critical lesson: the importance of keeping credit flowing.

  Its strategy for financial restructuring involved triage—separating out the really sick banks, which should be closed immediately, from the healthy banks. A third group were those that were sick but reparable. Banks are required to have a certain ratio of capital to their outstanding loans and other assets; this ratio is termed the capital adequacy ratio. Not surprisingly, when many loans are nonperforming, many banks fail to meet their capital adequacy ratio. The IMF insisted that banks either shut down or quickly meet this capital adequacy ratio. But this insistence on banks quickly meeting capital adequacy standards exacerbated the downturn. The Fund made the kind of mistake that we warn students about in the first course in economics, called “the fallacy of composition.” When only one bank has a problem, then insisting on its meeting its capital adequacy standards makes sense. But when many, or most, banks are in trouble, that policy can be disastrous. There are two ways of increasing the ratio of capital to loans: increasing capital or reducing loans. In the midst of a downturn, especially of the magnitude of that in East Asia, it is hard to raise new capital. The alternative is to reduce outstanding loans. But as each bank calls in its loans, more and more firms are put into distress. Without adequate working capital, they are forced to cut back on their production, cutting into the demand for products from other firms. The downward spiral is exacerbated. And with more firms in distress, the capital adequacy ratio of banks can even be worsened. The attempt to improve the financial position of the banks backfired.

  With a large number of banks shut down, and with those managing to survive facing an increasingly large number of loans in distress, and unwilling to take on new customers, more and businesses found themselves without access to credit. Without credit, the one glimmer of hope for a recovery would be squashed. The depreciation of the currency meant that exports should have boomed, as the goods from the region became cheaper, by 30 percent or more. But while export volumes increased, they did not increase nearly as much as expected, and for a simple reason: to expand exports, firms needed to have working capital to produce more. As banks shut down and cut back on their lending, firms could not even get the working capital required to maintain production, let alone to expand.

  Nowhere was the IMF’s lack of understanding of financial markets so evident as in its policies toward closing banks in Indonesia. There, some sixteen private banks were closed down, and notice was given that other banks might be subsequently shut down as well; but depositors, except for those with very small accounts, would be left to fend for themselves. Not surprisingly, this engendered a run on the remaining private banks, and deposits were quickly shifted to state banks, which were thought to have an implicit government guarantee. The effects on the Indonesia banking system, and economy, were disastrous, compounding the mistakes in fiscal and monetary policy already discussed, and almost sealing that country’s fate: a depression had become inevitable.

  In contrast, South Korea ignored outside advice, and recapitalized its two largest banks rather than closing them down. This is part of why Korea recovered relatively quickly.

  Corporate Restructuring

  While attention focused on financial restructuring, it was clear that the problems in the financial sector could not be resolved unless the problems in the corporate sector were effectively addressed. With 75 percent of the firms in Indonesia in distress, and half of the loans in Thailand nonperforming, the corporate sector was entering a stage of paralysis. Firms that are facing bankruptcy are in a state of limbo: it is not clear who really owns them, the current owners or the creditors. Issues of ownership are not fully resolved until the firm emerges from bankruptcy. But without clear owners, there is always a temptation for current management and the old owners to strip assets, and such asset stripping did occur. In the United States and other countries, when companies go into bankruptcy, trustees are appointed by the courts to prevent this. But in Asia there were neither the legal frameworks nor the personnel to implement trusteeships. It was thus imperative that bankruptcies and corporate distress be resolved quickly, before stripping could occur. Unfortunately, IMF’s misguided economics, having contributed to the mess through the high interest rates which forced so ma
ny firms into distress, conspired with ideology and special interests to dampen the pace of restructuring.

  The IMF’s strategy for corporate restructuring—restructuring the firms that were effectively in bankruptcy—was no more successful than its strategy for restructuring banks. It confused financial restructuring—entailing straightening out who really owns the firm, the discharge of debt or its conversion to equity—with real restructuring, the nuts-and-bolts decisions: what the firm should produce, how it should produce its output, and how it should be organized. In the presence of the massive economic downturn, there were real macrobenefits from rapid financial restructuring. Individual participants in the bargaining surrounding bankruptcy workouts would fail to take into account these systemic benefits. It might pay them to drag their feet—and bankruptcy negotiations are often protracted, taking more than a year or two. When only a few firms in an economy are bankrupt, this delay has little social cost; when many firms are in distress, the social cost can be enormous, as the macroeconomic downturn is prolonged. It is thus imperative that the government do whatever it can to facilitate a quick resolution.

  I took the view that the government should play an active role in pushing this financial restructuring, ensuring that there were real owners. My view was that once ownership issues were resolved, the new owners should set about the task of deciding the issues of real restructuring. The IMF took the opposite view, saying that the government should not take an active role in financial restructuring, but push for real restructuring, selling assets, for instance, to reduce South Korea’s seeming excess capacity in chips and bringing in outside (typically foreign) management. I saw no reason to believe that international bureaucrats, trained in macromanagement, had any special insight into corporate restructuring in general, or the chip industry in particular. While restructuring is, in any case, a slow process, the governments of Korea and Malaysia took an active role, and succeeded within a remarkably short period of time, two years, in completing the financial restructuring of a remarkably large fraction of the firms in distress. By contrast, restructuring in Thailand, which followed the IMF strategy, languished.

  THE MOST GRIEVOUS MISTAKES: RISKING SOCIAL AND POLITICAL TURMOIL

  The social and political consequences of mishandling the Asian crisis may never be measured fully. When the IMF’s managing director Michel Camdessus, and G-22 finance ministers and central bank governors (the finance ministers and central bank governors from the major industrial countries, plus the major Asian economies, including Australia) met in Kuala Lumpur, Malaysia, in early December 1997, I warned of the danger of social and political unrest, especially in countries where there has been a history of ethnic division (as in Indonesia, where there had been massive ethnic rioting some thirty years earlier), if the excessively contractionary monetary and fiscal policies that were being imposed continued. Camdessus calmly responded that they needed to follow Mexico’s example; they had to take the painful measures if they were to recover. Unfortunately, my forecasts turned out to be all too right. Just over five months after I warned of the impending disaster, riots broke out. While the IMF had provided some $23 billion to be used to support the exchange rate and bail out creditors, the far, far smaller sums required to help the poor were not forthcoming. In American parlance, there were billions and billions for corporate welfare, but not the more modest millions for welfare for ordinary citizens. Food and fuel subsidies for the poor in Indonesia were drastically cut back, and riots exploded the next day. As had happened thirty years earlier, the Indonesian businessmen and their families became the victims.

  It was not just that IMF policy might be regarded by softheaded liberals as inhumane. Even if one cared little for those who faced starvation, or the children whose growth would be stunted by malnutrition, it was simply bad economics. Riots do not restore business confidence. They drive capital out of a country; they do not attract capital into a country. And riots are predictable—like any social phenomenon, not with certainty, but with a high probability. It was clear Indonesia was ripe for such social upheaval. The IMF itself should have known this; around the world, the IMF has inspired riots when its policies cut off food subsidies.

  After the riots in Indonesia, the IMF reversed its position; food subsidies were restored. But again, the IMF showed that it had not learned the basic lesson of “irreversibility.” Just as a firm that was bankrupted by the high interest rates does not become “un-bankrupted” when the interest rates were lowered, a society that is rendered asunder by riots induced by cutting out food subsides just as it is plunging into depression is not brought together when the food subsidies are restored. Indeed, in some quarters, the bitterness is all the greater: if the food subsidies could have been afforded, why were they taken away in the first place?

  I had the opportunity to talk to Malaysia’s prime minister after the riots in Indonesia. His country had also experienced ethnic riots in the past. Malaysia had done a lot to prevent their recurrence, including putting in a program to promote employment for ethnic Malays. Mahathir knew that all the gains in building a multiracial society could be lost, had he let the IMF dictate its policies to him and his country and then riots had broken out. For him, preventing a severe recession was not just a matter of economics, it was a matter of the survival of the nation.

  RECOVERY: VINDICATION OF THE IMF POLICIES?

  By the beginning of the new millennium, the crisis was over. Many Asian countries were growing again, their recovery slightly stalled by the global slowdown that began in 2000. The countries that managed to avoid a recession in 1998, Taiwan and Singapore, fell into one in 2001; Korea is doing far better. With a worldwide downturn affecting the United States and Germany as well, no one talked about weak institutions and poor governments as the cause of recessions; now, they seemed to have remembered that such fluctuations have always been part of market economies.

  But although some at the IMF believe their interventions were successful, it’s widely agreed that serious mistakes were made. Indeed, the nature of the recovery shows this. Almost every economic downturn comes to an end. But the Asian crisis was more severe than it should have been, recovery took longer than it needed to, and prospects for future growth are not what they should be.

  On Wall Street, a crisis is over as soon as financial variables begin to turn around. So long as exchange rates are weakening or stock prices falling, it is not clear where the bottom lies. But once the bottom has been reached, the losses are at least capped and the worst is known. However, to truly measure recovery, stabilization of exchange rates or interest rates is not enough. People do not live off exchange rates or interest rates. Workers care about jobs and wages. Although the unemployment rate and real wages may have bottomed out, that is not enough for the worker who remains unemployed or who has seen his income fall by a quarter. There is no true recovery until workers return to their jobs and wages are restored to pre-crisis levels. Even as the new millennium began, incomes in the countries of East Asia affected by the crisis are still 20 percent below what they would have been had their growth continued at the pace of the previous decade. In Indonesia, output in 2001 was still 5.3 percent lower than in 1997, and even Thailand, the IMF’s best pupil, had not attained its pre-crisis level, let alone made up for the lost growth. This is not the first instance of the IMF declaring victory prematurely: Mexico’s crisis in 1995 was declared over as soon as the banks and international lenders started to get repaid; but five years after the crisis, workers were just getting back to where they were beforehand. The very fact that the IMF focuses on financial variables, not on measure of real wages, unemployment, GDP, or broader measures of welfare, is itself telling.

  The question of how best to manage a recovery is difficult, and the answer clearly depends on the cause of the problem. For many downturns, the best prescription is the standard Keynesian one: expansionary fiscal and monetary policy. The problems in East Asia were more complicated, because part of the problem was weaknesses
in finance—weak banks and firms with excess leverage. But a deepening recession makes these problems worse. Pain is not a virtue in its own right; pain by itself does not help the economy; and the pain caused by IMF policies, deepening recession, made recovery more difficult. Sometimes, as in Latin America, in Argentina, Brazil, and a host of other countries during the 1970s, crises are caused by profligate governments spending beyond their means, and in those cases, the government will need to cut back expenditures or increase taxes—decisions which are painful, at least in the political sense. But because East Asia had neither loose monetary policies nor profligate public sectors—inflation was low and stable, and budgets prior to the crisis were in surplus—those were not the right measures for dealing with East Asia’s crisis.

  The problem with the IMF’s mistakes is that they are likely to be long-lasting. The IMF often talked as if what the economy needed was a good purgative. Take the pain; the deeper the pain, the stronger the subsequent growth. In the IMF theory, then, a country concerned about its long-run prospects—say twenty years from now—should swallow hard and accept a deep downturn. People today would suffer, but their children at least would be better off. Unfortunately, the evidence does not support the IMF’s theory. An economy which has a deep recession may grow faster as it recovers, but it never makes up for the lost time. The deeper today’s recession, the lower the likely income even twenty years from now. It is not, as the IMF claims, that they are likely to be better off. The effects of a recession are long-lasting. There is an important implication: The deeper the recession today, not only is output lower today, but the lower output is likely to be for years to come. In a way, this is good news, since it means that the best medicine for today’s health of the economy and the best medicine for tomorrow’s coincide. It implies that economic policy should be directed at minimizing the depth and duration of any economic downturn. Unfortunately, this was neither the intention nor the impact of the IMF prescriptions.