The Asian Financial Crisis
by Morris Goldstein, Peterson Institute for International Economics
© Institute for International Economics. All rights reserved.
The turmoil that has rocked Asian foreign exchange and equity markets over the past eight months is the third major currency crisis of the 1990s. Its predecessors were the crisis in the European Monetary System in 1992-93, and the Mexican peso crisis of 1994-95.
Equity and currency markets in the most affected countries recorded huge drops—ranging from 20-75 percent—during the second half of 1997. Reflecting the effects of the crisis, growth rates for emerging Asia (excluding China) this year are now expected to be only marginally positive (1-2 percent), and Thailand, Indonesia, and South Korea are likely to suffer outright recessions.
The crisis will also have significant spillover effects outside the region. The IMF recently revised downward its 1998 projection for global growth from 4.5 to 3.8 percent. In the United States, most analysts estimate that the crisis will cause growth this year to be one half to three-quarters of a percentage point lower than would otherwise be the case.
This paper has a dual objective: first, to explain how the crisis rose and spread; and second, to outline the kinds of corrective policy measures that would help end the crisis and reduce the chances of a recurrence.
Origins of the Crisis
Like other financial crises of years past, the Asian crisis can be traced to a set of interrelated problems. In this case, three factors predominated: financial-sector weaknesses cum easy global liquidity conditions, problems in the external sector, and contagion running from Thailand to other economies.
Financial-sector weaknesses cum easy global liquidity conditions—More than anything else, it was financial-sector weaknesses that got the Asian countries into deep trouble. During the 1990s, each of the ASEAN 4 economies (Thailand, Indonesia, Malaysia, and the Philippines) experienced a credit boom, that is, the growth of bank and non-bank credit to the private sector exceeded by a wide margin the already rapid growth of the real economy. The credit boom was stoked in part by large net private capital inflows, and much of it was directed to real estate and equities. This overextension and concentration of credit left the ASEAN 4 economies vulnerable to a shift in credit conditions. When that shift came, induced by a need to control overheating and to defend fixed exchange rates, it brought with it, inter alia, falling property prices and a rising share of non-performing bank loans. Because the credit boom ended earlier in Thailand and Indonesia, the effects were first visible there.
Vulnerability was also heightened—particularly in Thailand and Indonesia (and later on in South Korea)—because banks and their corporate customers, in an effort to lower borrowing cost, undertook most of their foreign borrowing at short maturities and in foreign currency. These liquidity and currency mismatches eventually took their toll—both in motivating speculative attacks and in limiting the authorities' room for maneuver.
These financial-sector problems could not have progressed so far were it not for long-standing weaknesses in banking and financial-sector supervision. Loan classification and provisioning practices were too lax; there was too much "connected" and "policy-directed" lending; state-owned banks did not pay much attention to the creditworthiness of borrowers; bank capital was often inadequate relative to the riskiness of banks' operating environment; and there were strong expectations of government bailouts should banks get into difficulties. On top of this, the quality of public disclosure and transparency was poor.
Of course, it takes two (lenders as well as borrowers) to tango. It is well to recall that the 1990s were a time of bountiful global liquidity conditions, with over $420 billion in net private capital flows going to Asian developing countries alone; spreads declined, maturities lengthened, and loan covenants weakened. The presence of historically low interest rates in Tokyo also gave rise to a large "carry trade," where funds could be borrowed directly from Japanese institutions or intermediated via U.S. lenders. And much like Mexico before its crisis, the ASEAN 4 countries were viewed as among the most attractive emerging-market borrowers, with a history of rapid economic growth, high saving and investment rates, disciplined fiscal positions, and growing integration with the world economy.
External-sector problems—Most of the affected countries had run moderate to large current-account deficits during the 1990s (in 1996, Thailand registered a current-account deficit of about 8 percent of GDP). For quite a while, these external deficits were viewed as "benign," since they did not result from large public-sector imbalances and since foreign borrowing was being used mainly to increase investment (rather than consumption). But in 1996 and 1997, concerns mounted on several counts.
For one thing, attention shifted from the quantity to the quality of investment. Too much of the investment was directed at speculative activities, overextended industries, over-ambitious infrastructure projects, and inefficient government monopolies. Competitiveness in the ASEAN 4 economies also seemed to be waning, as indicated, by appreciations of real effective exchange rates (relative to a ten year trend), by sharp slowdowns in merchandise export receipts in 1996, and by a perceived shift in regional comparative advantage toward China. Moreover, the sustainability of Asian external deficits seemed threatened by overproduction in certain industries (e.g., autos, memory chips, petrochemicals, steel, wood products, etc.) and by intense export competition in the region.
Contagion—The third contributing factor to the crisis was the contagion of financial disturbances across countries. The Asian crisis is unusual in that it originated in a small country (Thailand) and spread to a wider set of economies, including some larger ones.
In analyzing the channels of contagion, it seems unlikely that bilateral trade or investment shares with Thailand could have been the driving force: these shares are just too small to generate such wide-ranging contagion. Instead, two other channels are more plausible. One is the "wake-up call" hypothesis. It says that Thailand acted as a wake-up call for international investors to reassess creditworthiness of Asian borrowers and when they did that reassessment, they found that a quite a few of these economies had weaknesses similar to those in Thailand (that is, weak financial sectors with poor prudential supervision, large external imbalances, appreciating real exchange rates, declining quality of investment, etc.). The other major contagion channel results from the dynamics of devaluation. As one country after another undergoes a depreciation, the countries who have not devalued experience a deterioration in competitiveness, which in turn makes their currencies more vulnerable to a speculative attack.
And as the number of countries affected by the crisis grew, other multilateral channels of trade and capital flow interdependence also generated spillovers. For example, the problems at Indonesia's banks and corporates have rebounded to the disadvantage of Singapore's banks, the crisis-induced weakening of primary commodity prices has hurt Chile's exports, and the difficulties experienced by Korean banks have had adverse knock-on effects as far away as Russia and Brazil (since Korean banks were heavy purchasers of both Russian GKOs and of Brazilian Brady bonds and liquidated much of their holdings during the crisis).
How To Fix It
Just as the Asian crisis did not arise from a single source, there is not a single silver bullet that will fix it. The main policy imperatives can be grouped under the following four headings.
Financial sector reform in the ASEAN 4 economies and in South Korea—Because of the key role that financial-sector fragility played in motivating the crisis, efforts to overcome that fragility must be the centerpiece of the recovery strategy. Insolvent banks and finance companies should be resolved (equity holders should lose their stake and management should be replaced), while those that are undercapitalized should be recapitalized to meet international standards. Foreign ownership limits should be liberalized. Each of the crisis countries also should commit itself to upgrade significantly its system of financial supervision and regulation. A tangible way to illustrate this commitment would be for them to sign on to the Basle Committee's new "Core Principles of Effective Banking Supervision," and to ask the IMF and the World Bank to monitor their compliance with these principles.
Recovery and reform in Japan—As the region's largest economy, Japan has to become part of the solution to the crisis—not a tinder box that hinders crisis management. This means, first of all, that Japanese domestic demand needs to achieve a satisfactory rate of growth. This would enable Japan to absorb its fair share of Asian exports, and it would help to end the downward slide of the yen. The fiscal stimulus package announced by Premier Hashimoto in late February is once again much too small; it should at least be tripled. Second, Japan must get its bad loan problem behind it. This would increase the room for maneuver in rescheduling the bad loans of banks in the crisis countries (as well as boost confidence in Japan itself). The good news is that the Japanese authorities have finally discarded the fantasy that they can resolve their bad loan problem without recourse to sizeable public funds (30 trillion yen). The bad news is that public funds may be used to prop up weak banks, and that serious systemic reforms may once again be delayed.
The role of the IMF—The IMF has recently been subject to sharp criticism on many fronts. With the exception of the "moral hazard" argument, most of this criticism is off the mark.
Some argue that the crisis countries didn't need strong IMF medicine because they were merely victims of a (negative) shift of sentiment on the part of international investors. This ignores the serious financial-sector weaknesses and external imbalance problems outlined above. The market may have overshot once the crisis got started, but the "innocent bystander" hypothesis simply doesn't wash.
Nor does the criticism that the Fund is no longer necessary stand up to scrutiny. Without access to conditional official financing, countries are likely to respond to external deficits with larger deflations and greater resort to competitive devaluations and to trade and exchange controls (as dramatically illustrated by the experience of the 1920s and 1930s). Even with a near $50 billion rescue package, Mexico underwent a 6 percent decline in real output in 1995—its deepest recession in five decades. The Thai, Indonesian, and South Korean economies will probably contract this year by several percent. What would growth and social stability be in these countries without any official financing? And would it be better if the policy conditionality in these rescue packages were administered by individual creditor countries (with all the political overtones that would imply) rather than by an international institution with a constitutional mandate to provide conditional financing? Surely not.
The charge that the IMF is being too intrusive by making detailed recommendations about financial-sector reform and corporate governance misses the point that wholesale reform of banks, finance companies, conglomerates, and government monopolies is absolutely crucial if the crisis countries are to regain confidence and market access to private financing. After all, the crisis occurred in good measure because these needed reforms had been too long delayed.
The Fund has also been criticized for recommending higher interest rates, bank closure, and tighter fiscal policy. But experience suggests that it is difficult to stabilize a rapidly declining currency without a temporary period of high interest rates (recall the case of the Mexican peso in early 1995), and that allowing insolvent banks/thrifts to remain open encourages "gambling for resurrection" that can add significantly to the ultimate public-sector tab of banking problems (recall the U.S. saving and loan crisis). If there was a mistake in Indonesia, it was that the authorities did not close enough insolvent banks.
In retrospect, a somewhat easier fiscal stance may well have been appropriate for the crisis countries but the Fund has shown that it is willing to be flexible on fiscal targets when growth turns out to be weaker than expected. In any case, the initial ineffectiveness of the rescue packages in stabilizing markets was not primarily due to the monetary and fiscal targets in the Fund programs but rather to other factors (namely, convincing market participants that structural reform was really going to happen, political instabilities in the crisis countries, and uncertainties about how the short-term debt overhang was going to be resolved).
The one criticism of the Fund (and of the main creditor countries) that stands on the firmest ground is the "moral hazard" argument, that is, the provision of insurance by the official sector that acts as a subsidy to risk taking and results in too many resources being channeled into the insured activities. Despite the conclusions drawn by the G-10 after the Mexican crisis (e.g., that governments should resist pressures to assume responsibility for the external liabilities of their private sectors), actions by the official sector in the Korean crisis helped insulate private creditors of Korean banks from bearing the full consequences of poor lending decisions. If the official safety net becomes wider and wider over time, we can expect private lenders to increasingly channel international capital flows to those borrowers who are deemed "too large to fail." The only thing that will deter excessive risk taking is the expectation of taking a significant loss, which requires that lenders sometimes do take a significant loss. Equity and bond holders have taken large losses but many large creditors of banks have been shielded more than they should have been. Without the deterrent of losses, market spreads on emerging-market paper will not price risk appropriately, and pleas to banks and corporates to make good internal risk management the first line of defense against crises will likely fall on deaf ears. In addition, bailing-out large uninsured creditors makes it difficult to sell the principle of equitable burden-sharing and could jeopardize public support for IMF funding more generally.
Improving the crisis prevention/crisis management architecture—As in the aftermath of the Mexican crisis, the official sector ought to use the heightened awareness of crisis vulnerability to improve the international architecture of crisis prevention and crisis management. Three items should receive high priority in any Halifax II discussions.
First, a way needs to be found to facilitate a more orderly rescheduling of private external debt. The fact that such procedures had not been well thought out prior to the current crisis has meant that official rescue packages have turned out to be much larger than anyone anticipated, with longer-term moral hazard effects having to play second fiddle to short-term (but admittedly important) financial stability concerns. What is needed is a set of agreed principles and procedures that can reduce the uncertainty surrounding the restructuring and rescheduling of private debt. The discussion should focus, inter alia, on how to limit the use of blanket government guarantees, on the application of the IMF's "lending into arrears" policy, on the representation of debtor and creditor groups in rescheduling negotiations, and on the options that will be most helpful in resolving the problem (e.g., equity for debt exchanges, the substitution of long-term for short-term instruments, etc.).
A second priority item concerns banking supervision and financial-sector infrastructure in emerging economies. Now that agreement has been reached on the Basle Core Principles of Effective Banking Supervision, a target date should be set for having countries sign on to it. In addition, countries should be encouraged to adopt domestic bankruptcy codes that respect the rights of foreign and domestic creditors and to enact deposit insurance arrangements that protect small retail depositors.
Item number three on my proposed agenda falls under the heading of data standards and public transparency and disclosure. More specifically, the IMF's Special Data Dissemination Standard should be amended to cover timely reporting of the maturity and currency composition of external debt, as well as net international reserves (inclusive of government commitments in the forward exchange market). On a parallel track, monitoring of the Basle Core Principles should pay particular attention to more timely reporting and to a stricter (standard) definition of non-performing bank loans.
To conclude, despite the breadth and depth of the Asian financial crisis and the limited success achieved thus far in combating it, the successful resolution of the crisis does not require either a new game plan or a new IMF. Much more has been right than wrong about the overall design of official rescue packages, and there is still time to improve the market discipline and burden-sharing aspects. A sustained upward turn will come when the crisis countries have made enough progress in implementing structural reforms to convince markets that things have really changed, and when there is enough rescheduling of private debt to reduce uncertainty and moral hazard.