Chapter 2
East Asia’s financial crisis: causes, evolution, and prospects

Back to Contents

The financial crisis that began in Thailand, and led to the 20 percent devaluation of the baht on July 2, quickly spread to other countries in Southeast Asia and eventually to the Republic of Korea, and had repercussions around the globe. Between June 1997 and the end of the year the median currency devaluation in 12 of the largest emerging markets was 39 percent,1 and in the 5 East Asian countries hardest hit by the crisis—Indonesia, Korea, Malaysia, the Philippines, and Thailand—it was 80 percent. The International Finance Corporation’s (IFC) emerging stock market index dropped 25 percent between June and December, and its Asian index fell 53 percent. Among the group of 12 countries the median rise in short-term interest rates (from July 1 to each country’s peak rate) was more than 600 basis points.

The crisis began in mid-1997 with intense pressures on the Thai baht. Efforts to prop up the currency—including intervention, increases in interest rates (to 18 percent in June 1997 compared with 12 percent in January), and restrictions on foreign speculators—ultimately failed because by mid-1997 domestic companies sought to protect themselves from foreign exchange risks to their balance sheets by repaying foreign debt and hedging their foreign exchange exposure. After spending $8.7 billion in reserves to defend the currency (and undertaking $23 billion in forward contracts maturing over the next 12 months), Thailand’s central bank let the exchange rate float on July 2. By the end of the year the baht had depreciated 93 percent and the stock market had fallen 34 percent (in dollar terms) relative to June 1997. The devaluation of the baht triggered a collapse of market confidence in Indonesia, Malaysia, and the Philippines, causing their exchange rates to plummet in July and August 1997. Further declines in currencies and equity markets followed. During the second half of 1997 equity prices in the region fell 50–75 percent (in dollar terms).

The crisis began to affect other economies in October, when speculative pressures intensified against the Hong Kong dollar, the Korean won, and the Taiwan (China) dollar, accompanied by sharp falls in stock markets in these economies. A severe plunge in Hong Kong’s equity market precipitated a global drop in equity prices (affecting Eastern Europe, Latin America, Japan, Europe, and the United States) and increased pressures on the currencies of developing countries. In November Korea, the world’s eleventh largest economy, became the object of the largest ever international rescue package. Most countries in Latin America and Eastern Europe were able to maintain exchange rate policies or stabilize rates after the initial devaluation. But the monetary (and in some cases fiscal) tightening required to restore confidence has undermined growth prospects in several emerging markets.

Back to top
Back to Contents

This chapter draws the following conclusions in analyzing the causes and implications of the still-unfolding crisis.

• The East Asian crisis differs from previous developing country crises—such as the Mexican peso devaluation of 1994 or the debt crisis of the 1980s—in that private sector financial decisions were the main source of difficulties. Public sector borrowing played only a small role. Furthermore, East Asia’s crisis occurred despite a benign international environment, with low international interest rates and solid global growth in output and exports.

• Despite impressive macroeconomic performance and prudent fiscal policies, East Asian economies have become increasingly vulnerable during the 1990s. The most pressing problems are in the financial sector, where distorted incentives, lax regulatory standards, poorly managed financial liberalization, and inadequate disclosure and supervision have encouraged excessive risk taking, particularly in terms of maturity and currency mismatches. Financial sector weaknesses led to a misallocation of investment and a buildup of nonperforming loans. Large capital inflows amplified the problems of the financial sector and fueled domestic demand which, coupled with the depreciation of the yen relative to the dollar (to which these countries were closely tied), caused real exchange rates to appreciate.

• The increase in vulnerabilities does not fully account for the spread and depth of the crisis, however. The severity of currency and stock market declines and the few warnings from market participants indicate that a self-fulfilling loss of market confidence played an important role. With the loss of confidence and substantial uncovered foreign exchange exposure, currency depreciation became self-perpetuating: the rise in the local currency value of liabilities impaired balance sheets, lowered stock prices, and increased demand for foreign exchange to cover open positions. Increased demand for foreign exchange led to further currency depreciation, and so on. Both healthy and insolvent firms suffered because of the lack of transparency (investors, unable to distinguish among firms, withdrew from all of them), the effect of the currency depreciation on dollar–denominated debt, the increases in interest rates to defend the currency, the contraction in credit resulting from the rapid drop in equity of highly leveraged financial institutions (due both to their own losses and to the insolvency of their borrowers), and increased uncertainty and the economic downturn.

• The fallout from the East Asian crisis triggered a steep decline in portfolio flows to developing countries toward the end of 1997, and little recovery is expected in the near term. Private flows to developing countries will likely decline moderately in 1998, and downside risks are significant.

The crisis and its causes

The East Asian crisis differs from previous financial crises in developing countries in several ways: private decisions were the main source of difficulties, public borrowing played a limited role, and inflation was low relative to most other developing countries (and had declined in the 18 months preceding the crisis). Unlike Latin American countries in the 1980s, most East Asian countries had been running a fiscal surplus prior to the crisis (table 2.1). Unlike Mexico before the 1994–95 peso crisis, public debt was not a relevant factor; the short-term maturities of private, not public, debt were the main source of financial vulnerability. And excessive consumption by the private sector was not an issue in East Asia, which has high savings rates.

Table 2.1 Fiscal balances and gross national savings in East Asian countries, 1993–96
(percentage of GDP)

Source: World Bank and IMF data.

Back to top
Back to Contents

The crisis also differs from previous ones in that it occurred against a benign international background, with high rates of growth in world trade, low international interest rates (both nominal and real), and declining spreads on international borrowing (see chapter 1), which implied low debt servicing costs and a favorable environment for additional borrowing and equity flows. Still, external factors did play a role, in the form of the sizable appreciation of the U.S. dollar during 1996–97. Southeast Asian currencies’ close links to the dollar contributed to exchange rate appreciation, which was an important factor in the lower export growth rates relative to the extraordinary growth in the first half of the 1990s. And although Korea’s real effective exchange rate was stable, a weaker market for computer chips in 1996 hurt its economy.

A number of factors made the East Asian economies especially vulnerable to a sudden crisis of confidence. As noted, these problems (including misallocation of investment and maturity and currency mismatches) were rooted in private sector financial decisions. But policies—especially mismanaged exchange rates, insufficient financial regulation, and implicit or explicit government guarantees—also played an important role in creating the incentives that led to the particular size and character of external financing and internal resource allocation. Of course, every loan involves a decision to lend as well as to borrow. Thus lenders, many of them international banks, share with borrowers responsibility for the consequences of bad loans. Similarly, responsibility for resolving these problems lies not just with bank supervisors in the borrowing countries but also with bank supervisors in the lending countries—particularly if the international community believes that there is sufficient systemic risk to the global economy to warrant intervention.

The buildup of short-term, unhedged debt left East Asian economies vulnerable to a sudden collapse of confidence. The loss of confidence led to capital outflows, and thus to depreciating currencies and falling asset prices, which further strained private balance sheets and so proved self-fulfilling. And once started, the scale of capital outflows may have had little to do with currency mismatches or even financial sector weaknesses, except to the extent that they contributed to perceptions of weaknesses in the economy. East Asia’s close integration with international financial markets makes it easy to export capital (although in a major confidence crisis even countries with relatively closed capital accounts can experience substantial capital flight). In fact, there is evidence that domestic investors were major participants in the outflow of capital—an experience similar to that in Mexico during 1994–95. The downward spiral has accelerated as financial problems have led to restricted credit, undermining the real sector and further contributing to financial fragility. If economic problems fuel social and political tensions, the crisis will deepen further.

Manifestations of vulnerability

Large amounts of private capital have flowed to East Asia in recent years, although the scale and phasing have differed significantly among countries. Between 1994 and 1996 net private capital inflows as a share of GDP increased considerably in several East Asian countries—for example, by 7 percentage points in Malaysia, 6 percentage points in Indonesia, and 5 percentage points in the Philippines (table 2.2). Only in Thailand did net private capital flows as a share of GDP remain stable, though averaging more than 15 percent over the three years. The surge in flows reflected these countries’ strong economic performance, including rapid growth, sustained improvements in macroeconomic balances (public sector balances, inflation), and structural changes that have fostered a market-led, outward orientation since the late 1980s. The cyclical downturn in international interest rates in the early 1990s provided the initial impetus for the surge in flows (particularly portfolio flows); continued increases reflected structural changes that have increased the responsiveness of capital to cross-border investment opportunities (see World Bank 1997c).

Table 2.2 Net private capital flows to East Asia, 1994–96
(percentage of GDP)

Source: World Bank data.

Back to top
Back to Contents

Absorbing large inflows of private capital and deploying them productively presented challenges to economic management. Despite the good macroeconomic track record of the East Asian countries, three sources of vulnerability emerged: large current account deficits and misallocated investment, bad loans and currency and maturity mismatches, and misaligned real exchange rates and lost competitiveness.

Large current account deficits and misallocated investment. Although East Asia has high savings rates (averaging more than 30 percent of GDP in Indonesia, Korea, Malaysia, and Thailand during 1993–96), foreign capital was required to finance its even higher investment rates (which rose to about 40 percent of GDP in Korea, Malaysia, and Thailand and 34 percent in Indonesia during 1993–96). Still, financial crises result more from a loss of confidence in borrowers’ ability to repay than simply from the existence of large current account deficits. Indeed, the statistical link between current account deficits and currency crises appears weak (box 2.1).

Although (except for Korea) East Asia’s debt relative to GDP was much higher in 1996 than Latin America’s, its debt relative to exports was generally much lower (figure 2.1). (Only Indonesia’s debt to exports ratio—220 percent—might have been cause for concern.) Thus the size of the external debt-exports ratio was not necessarily a binding constraint on further borrowing. The large share of short-term debt owed by the private sector, however, increased the region’s vulnerability to internal and external shocks that could undermine investor confidence. Even where short-term borrowing was used to purchase long-term external assets (so that investors did not have substantial currency exposure), the refusal of creditors to roll over credit lines led to a liquidity squeeze because of the maturity mismatch.

Back to top
Back to Contents

The sustainability of large current account deficits partly depends on how the funds are used. If East Asian borrowers had invested the funds in diversified, high-productivity investments, the deficits would have been sustainable. But instead much of the capital inflow (and the coincident surge of domestic investment) appears to have been directed into risky, low-productivity investments. For example, a large portion of domestic credit flowed into nontradables, especially real estate. Market sources report that real estate loans accounted for about 25 percent of outstanding bank loans in Malaysia and the Philippines2 and 20 percent in Thailand. The surge in the region’s real estate investments rapidly outpaced demand: by 1996 vacancy rates had reached nearly 15 percent in Bangkok and Jakarta. The poor quality of real estate investment in Thailand was reflected in the stock market, where the index for building and furnishing companies collapsed from a peak of just below 8,200 in late 1994 to 1,100 in late 1997. The concentration of loans in the highly cyclical property market led to asset price bubbles and made the banking system vulnerable to a downturn in domestic demand. And the concentration of bank portfolios in nontradables—along with currency mismatches—made them more vulnerable to a depreciation in the currency.

More generally, an increase in incremental capital-output ratios in East Asia may reflect the poor quality of a large portion of new investment during the 1990s (box 2.2). Although in recent years the share of investment in GDP rose across East Asia, GDP growth rates remained roughly constant and even fell in some countries. For example, the efficiency of Korean investment declined in the years preceding the crisis, with the share of investment in GDP rising from 30 percent in 1983–89 to 38 percent in 1995–96 while GDP growth fell from 10 percent to 8 percent. The increase in the capital intensity of output partly reflected a shift away from labor-intensive sectors (such as garments, footwear, and light assembly) toward what were considered strategic sectors for Korean industrialization (including automobiles and electronics). High capital spending led to overcapacity in many sectors, as the country’s chaebol (industrial conglomerates) pursued investments hoping to build market share both domestically and overseas. To some degree, however, the decline in GDP growth relative to investment could also reflect changes that lowered the return on the capital stock—such as shifts in international demand for computer chips—rather than a decline in the productivity of new investment.

Back to top
Back to Contents

The low productivity of investment and the growing vulnerability of the corporate sector can also be seen in the drop in Korean companies’ return on assets (to 1 percent in 1996) and the spate of bankruptcies in 1997 (prior to the spread of the crisis to Korea). Similarly, the return on equity in Indonesia, Malaysia, and Thailand declined between 1992 and 1996 to below money market interest rates, indicating that there was no compensation for the risk of investing in these economies.

Bad loans and currency and maturity mismatches. Low-productivity investments were reflected in banking systems’ high nonperforming loans, which according to market sources reached 19 percent of loans in Indonesia, 16 percent in Malaysia, 13 percent in the Philippines, and 17 percent in Thailand in the immediate aftermath of the devaluation.3 Conservative estimates suggest that by the end of 1997 the equivalent of 6.5 percent of GDP was required to restore the capital adequacy of Korea’s merchant, commercial, and specialized banks. Some East Asian banking systems were made more vulnerable by the fact that a lot of external borrowing (especially in the past few years) was short term and was used to fund local currency–denominated assets, particularly in real estate and other nontradables. Even when banks avoided currency mismatches, their clients often had significant foreign exchange exposure, so the banks remained vulnerable to a fall in the exchange rate.

The extent of currency and maturity mismatches varied considerably among East Asian economies. Foreign exchange exposure was particularly high in Thailand, where the banking system’s net foreign liabilities rose to 20 percent of domestic assets in 1996, compared with 3 percent in 1990. By mid-1997 short-term debt had risen to 67 percent of total debt in Korea and 46 percent in Thailand, compared with 19 percent in the Philippines and 25 percent or less in every Latin American country. And ratios of short-term debt to reserves were high in Indonesia, Korea, and Thailand, but lower in Malaysia and the Philippines (table 2.3).

Table 2.3 Ratios of short-term debt to total debt and to reserves in East Asia and Latin America, mid-1997
(percent)

Source: Bank for International Settlements data; IMF International Financial Statistics; and World Bank data.

Misaligned real exchange rates and lost competitiveness. During the 1990s real exchange rates appreciated by 18 percent in the Philippines, 13 percent in Malaysia, 9 percent in Thailand, and 5 percent in Indonesia (by June 1997 relative to the 1988–92 average; figure 2.2 and table 2.4). Korea maintained a stable real exchange rate despite a 27 percent drop in its terms of trade in the three years prior to September 1997.

Back to top
Back to Contents

Table 2.4 Real effective exchange rate indexes, December 1996–December 1997
(1988–92 average=100)

Source: J.P. Morgan data.

The loss in competitiveness in Southeast Asian economies was reflected in a substantial slowdown in exports. During 1990–95 Southeast Asia’s export revenues grew by an average of 18 percent a year (in dollar terms). In 1996 and early 1997, however, export growth slowed markedly—well below the growth rate of world trade as a whole—as volume growth fell and dollar prices declined precipitously (figure 2.3). In Thailand export revenues rose just 1 percent in the year preceding June 1997. The rapid growth in export revenues during the first half of the 1990s may have given companies and financial intermediaries reason to believe that high unhedged dollar debt could be adequately serviced. The export slowdown, however, limited their ability to do so.

Continue with East Asia’s financial crisis: causes, evolution, and prospects

Back to top
Back to Contents