November 7, 1997
« More Economic Letters
Lessons from Thailand
After more than a decade of maintaining the Thai baht’s near-peg to the U.S. dollar, Thai authorities abandoned the peg on July 2, 1997. By October 24, market forces led the baht to depreciate by 60% against the U.S. dollar. The depreciation triggered a wave of speculation against other Southeast Asian currencies; over the same period, the Indonesian rupiah, Malaysian ringgit, and Philippine peso depreciated by 47, 35, and 34%, respectively. There also has been some speculation against other East Asian currencies, but at this writing the adjustment in these currencies has been more limited.
This Letter discusses why the peg of the Thai baht was abandoned and what lessons we may draw from this recent episode of currency speculation. It focuses on Thailand, which appears to have been most severely affected, and whose experience illustrates some of the factors that may have contributed to the spread of currency speculation to other Southeast Asian economies.
In some ways, the attack on the Thai baht was surprising, as Thailand has been one of Southeast Asia’s outstanding performers. Thailand’s growth in the first half of the 1990s averaged over 8%, before slowing to 6.4% in 1996. Although the inflation rate rose from 3.4% in 1993 to 5.9% in 1996, it was still moderate by the standards of emerging markets. Thailand’s domestic savings were high — about 34% of GDP — and the government has run fiscal surpluses in nine of the past ten years. These indicators contrast sharply to those of Mexico prior to the peso collapse in December 1994. In Mexico, domestic savings were low, GDP growth was sluggish, and the economy was still experiencing the residual effects of the triple digit inflation of the late 1980s.
What Thailand did share with Mexico was a heavy reliance on short-term foreign funds, attracted both by the stable baht and by Thai interest rates that were much higher than comparable rates in the Eurodollar market (in 1996, the spread between Thai and Eurodollar rates was around 5 percentage points at a one-month maturity). In 1996, the net rate of foreign financing as a percentage of GDP in Thailand (as measured by the current account deficit) was 8%, compared to 7% in Mexico at the time of the 1994 peso crash. In the first half of the 1990s, foreign financing supported a broad-based economic boom that was particularly visible in the real estate market.
The downside to Thailand’s economic boom became apparent after 1995, when economic activity slowed, led by a sharp reduction in Thai export growth. Export growth fell in part because of a steep appreciation in the real trade-weighted baht (against an inflation-adjusted currency basket comprising the yen, the dollar, and the deutschemark) of 21% between April 1995 and December 1996. This appreciation reflected the close link of the baht to a sharply appreciating U.S. dollar.
The economic slowdown was accompanied by a sharp interruption in Thailand’s real estate boom. By the end of 1996, office vacancy rates in Bangkok exceeded 20%, and a similar glut was apparent in commercial and residential property markets. Along with excess capacity in manufacturing and services, weakness in the real estate market put pressure on the financial sector. A dramatic prelude to the financial difficulties to come was the failure of the Bangkok Bank of Commerce, which incurred over $3 billion in bad loans and was taken over by the government in May 1996.
After the second half of 1996, concerns about sluggish export growth and the weak economy, accentuated by evidence of weaknesses in the financial sector, led to sporadic pressures on the baht to depreciate. However, the government sought to resist these depreciation pressures for several reasons. First, the stable baht had encouraged many Thai firms to borrow in dollars without hedging their foreign currency exposure, so a depreciation of the baht would increase their debt burdens (by increasing the amount of baht needed to service a given amount of U.S. dollar debt). Second, a sharp depreciation of the baht could increase the cost of foreign borrowing to some degree, as investors would demand to be compensated for increased baht volatility. Finally, a weaker baht would tend to increase inflation and reduce the purchasing power of domestic residents.
Confronted with these potential costs, the Bank of Thailand (BOT, the Thai central bank) chose to resist depreciation pressures in a number of ways. The central bank purchased baht with dollars in the foreign exchange market. The BOT also raised interest rates, which increased the attractiveness of the baht and also made it more costly for speculators to borrow baht in order to sell it in foreign currency markets. Measures to restrict foreigners’ access to baht also were adopted after a particularly severe episode of speculation against the baht in mid-May 1997.
In the course of using these various tools, the BOT encountered significant constraints on its ability to resist depreciation pressures. First, paralleling Mexico’s experience in 1994, Thailand found its reserves could be depleted very quickly if sentiment shifted against the baht. Indeed, the BOT’s foreign exchange reserves fell from $37.2 billion in December 1996 to $30.9 billion in June 1997 (prior to the baht float). In addition, the BOT reported in August 1997 that in the course of the year it had incurred off-balance sheet obligations to deliver $23.4 billion dollars in the forward market over a 12-month period, so that net reserves were actually lower. Second, raising interest rates adversely affected an already weak financial sector by dampening economic activity and raising the cost of funds for existing borrowers, who might then find it difficult to service their loans. Finally, restricting foreigners’ access to baht temporarily curbed speculation against the currency but could not prevent pressures on the baht to depreciate from other sources. For example, if holders of dollars were less willing to supply the amounts needed to finance Thailand’s foreign borrowing, the baht would tend to depreciate anyway.
To illustrate the implications of efforts to defend the baht, the figure shows the baht exchange rate and the Thai one-month interbank rate since early 1996. Starting in the summer of 1996, sporadic episodes of speculation against the baht were resisted by intervention and occasionally by very large increases in interest rates. After mid-May 1997, interest rates remained several percentage points above their 1996 levels, and by the time the peg was abandoned on July 2, interest rates had risen to over 24%. The steep increase in interest rates imposed significant costs on the domestic economy and the financial sector, which ultimately contributed to the abandonment of the peg.
In order to facilitate economic recovery, Thailand has adopted an economic stabilization program designed to repair the financial system and achieve certain macroeconomic goals. Operations were suspended at 58 of Thailand’s 91 finance companies, which had received Bt 430 billion (about $18 billion at the exchange rate prevailing on July 1, 1997, or over 9% of 1996 GDP) from the BOT in an effort to keep them afloat. A deposit insurance system is being set up, and the 25% ceiling on foreign ownership of Thai financial institutions has been raised. Thailand’s stabilization plan also would reduce the rate of foreign borrowing while maintaining an adequate supply of foreign exchange reserves, to balance the government budget (which went into deficit this year) and to bring greater transparency to government finances and enhance market efficiency. This program is supported by international assistance of close to $17 billion involving the IMF, Japan, and a number of economies in the Western Pacific Basin. In contrast to its leading role in mobilizing over $50 billion to support Mexico after the peso collapse, the U.S. has played no direct role in the international assistance program for Thailand.
Ever since the Mexican peso crisis, which caught many analysts by surprise, there has been much interest in improving the flow of information in emerging markets. The IMF has encouraged countries to subscribe to an international data dissemination standard that has contributed to the more timely release of information by central banks in emerging markets. While transparency could be further improved, it appears that the information available, along with a better understanding of factors that may trigger speculation against a currency, made it possible for some observers to anticipate the potential vulnerability of the baht peg.
However, the recent episode has highlighted two difficulties in the management of information flows that were not so apparent during the attack on the Mexican peso. First, conventional accounting measures sometimes provide a misleading picture of economic or financial conditions. For example, conventional foreign exchange reserve measures do not take into account off-balance sheet transactions, which were very large in Thailand. (As noted earlier, BOT now has disclosed its position in forward markets.) Accounting difficulties also cloud the assessment of weaknesses in the financial sector, not just in Thailand, but also in other emerging markets in the region. For example, conventional measures may understate the exposure of the banking sector to the volatile real estate sector because they do not reflect the extent to which lending to other sectors is backed by collateral in the form of real property.
Second, the recent episode has illustrated the extent to which openness has exposed Asian economies to sudden shifts in market sentiment, which can disrupt economic activity by triggering large changes in exchange rates and other asset prices. This experience apparently has made some Asian policymakers very wary about how available data are interpreted and used by financial market participants. It also has raised the stakes in the debate on whether policymakers should respond by further increasing transparency and giving even more leeway to market forces (the approach adopted in more developed countries), or by curbing the activities of market participants.
The recent episode of currency speculation in Thailand illustrates the risks associated with pegging the exchange rate in small economies that are open to capital flows and that have less developed financial sectors. Pegging encouraged an extended period of capital inflows that supported an economic boom and a run-up in asset prices that were not sustainable. Years of rapid growth also disguised poor lending and investment decisions that ultimately led to the closure of a large number of financial institutions. Under these conditions, indicators such as rapid growth and high saving and investment rates, or budget surpluses, provided an incomplete picture of the ability of the economy to respond to shocks.
Policymakers had no easy choices once pressures on the exchange rate emerged, as defending the peg or allowing the currency to depreciate both involved significant costs. It is apparent in hindsight that this dilemma could have been avoided by allowing the currency to appreciate during the earlier period of capital inflows, when there was no threat of a sudden loss in value in the currency. An appreciation of the baht would have dampened capital inflows, thus reducing the dependence on foreign financing and attenuating both the boom and subsequent contraction. The greater flexibility in exchange rates also would have encouraged domestic residents to hedge their foreign currency exposure. However, policymakers in small open economies are often reluctant to allow the currency to appreciate in response to capital inflows precisely because such appreciation dampens economic growth and reduces the competitiveness of the trade sector.
Finally, this recent episode reveals that in spite of progress made in improving information from emerging markets, conventional accounting measures may not always provide an accurate picture. In addition, it highlights the much greater exposure of Asian economies to shifts in market sentiment, raising the stakes in the debate on whether transparency should be further enhanced and financial markets further liberalized, or whether some of the activities of market participants should be curbed.
More Economic Letters
Pacific Basin Notes are published occasionally by the Center for Pacific Basin Studies. Opinions expressed in FRBSF Economic Letter do not necessarily reflect the views of the management of the Federal Reserve Bank of San Francisco or of the Board of Governors of the Federal Reserve System. This publication is edited by Sam Zuckerman and Anita Todd. Permission to reprint must be obtained in writing.
Please send editorial comments and requests for reprint permission to