Sayonara Dollar Peg: Asia in Search of a New Exchange Rate Regime

February 1, 2000


The latest currency crisis in Asia has raised a number of important issues relating to exchange rate management. First, why has the dollar-peg system, which seemed to have served Asian countries so well in the past, become so vulnerable? Second, is the de-linkage of Asian currencies from the US dollar just an emergency measure that should be reversed once the crisis is over? Finally, if the dollar-peg system should be abandoned sooner rather than later, what are the alternatives? This paper seeks to answer these questions by applying the tools of economic analysis to the current circumstances of Asian countries.


Led by Thailand, many developing countries in Asia were forced to abandon their traditional dollar-peg system and to allow their exchange rates to float in the summer of 1997 amidst intense currency attacks by speculators. After a period of sharp depreciation accompanied by high volatility, exchange rates in these countries bottomed out against the US dollar in the autumn of 1998, thanks to the yen’s rebound and a return of confidence in the economic prospects of these countries. However, most Asian currencies have not restored their stability against the dollar seen before the crisis. With the dust now settling, it is the right time to explore what should be done about the exchange rate system in order to restore stable economic growth.

The latest crisis in Asia has revealed the vulnerability of the traditional dollar-peg system widely adopted by the Asian countries. First, fluctuations of the yen against the dollar have led to sharp swings in export performance as well as net capital inflow. Second, as synchronization between Asian and US growth rates has weakened, leaving interest rates to be determined in the United States has led to macroeconomic instability. Finally, countries committing to fix their exchange rates against the dollar are vulnerable to speculation.

Corresponding to these weaknesses of the dollar-peg system that come from diverse sources, the remedies proposed span the gamut from moving to more flexible systems to strengthening the dollar-peg system. If the widely fluctuating yen-dollar rate is the root of the problem, Asian countries should peg to a basket of currencies in which the yen carries a substantial weight. If the loss of monetary independence is the major source of instability, allowing the exchange rate to float is recommended. If unwarranted speculation is blamed, the dollar-peg system should be strengthened rather than abandoned for a more flexible regime.

Although it has become fashionable to eliminate intermediate regimes as viable options in favor of either free floating or firm fixing, neither of these polar regimes seem to be appropriate for Asia’s developing countries at this stage. A more promising strategy is to strengthen the institutional framework, both at the national and international levels, to reduce the vulnerability of the intermediate regimes to speculative attacks.

Pegging to a Single Currency versus a Basket of Currencies

With the yen fluctuating widely against the dollar, the traditional policy of pegging their exchange rates to the dollar has led to macroeconomic instability in Asia’s developing countries (Figure 1) Every time the yen appreciates against the dollar, the economic growth rate of Asia (the Asian NIEs, the ASEAN countries, and China as a group) picks up, as happened between 1986 and 1988 and again between 1991 and 1995. These periods were also accompanied by an expansion of the bubble economy, with asset prices rising sharply. The reverse is true when economic growth in these countries decelerated and the asset bubble burst on the back of a weaker yen in 1989-90 and again between mid-1995 and the autumn of 1998. The latest downturn in Asian economic growth that became apparent since 1996 can largely be explained by the sharp depreciation of the yen against the dollar. The weakening yen also led to a marked deterioration in Asia’s export performance and current account balances in 1996, paving the way for the currency crisis.

Given the very large impact on economic growth from fluctuations in the yen-dollar rate, stability of the yen-dollar rate is crucial for stability in the regional economy. Unfortunately, the yen-dollar rate is beyond the control of Asia’s developing countries, or even of the major industrial powers, including Japan and the United States. One way for an Asian country to insulate its economy from the adverse effects of a widely fluctuating yen-dollar rate is to peg to a basket of currencies in which the yen carries a substantial weight. The higher the weight assigned to the yen, the stronger the synchronization between the host country’s currency and the yen. By allowing the domestic currency to follow the yen up (down) when the yen strengthens (weakens) against the dollar, pegging to a currency basket should help stabilize exports. Should Asia’s developing countries adopt this policy, the ups and downs in their business cycles would be moderated in magnitude, and the strong correlation between Asian economic growth and the yen-dollar rate would become a phenomenon of the past.

The “optimal” weight assigned to the yen in this currency basket should differ from one country to another, depending largely on how much the host country competes with Japan. Under the traditional dollar-peg system, higher income countries that compete directly with Japan are more vulnerable to fluctuations in the yen-dollar rate than lower income countries that have trade structures complementary to that of Japan; their exports surge when the yen is strong and plummet when the yen is weak. To stabilize exports, they need to pay more attention to the stability of their currencies against the yen. Thus, South Korea should peg closer to the yen than Thailand, which in turn should assign a higher weight to the yen in its reference basket than Indonesia, for example.

While a system of pegging to a basket of currencies with an appropriately chosen weight for the yen should contribute to macroeconomic stability in Asia, it shares with other fixed exchange rate systems such demerits as lack of monetary autonomy and vulnerability to speculative attacks. Compared with a system of pegging to a single currency, a basket-peg system is harder to administer and the commitment to the peg is interpreted as weaker. At the micro level, the cost of transactions is likely to be higher because all bilateral rates fluctuate.

Fixed versus Floating Exchange Rates

As with all fixed exchange rate systems (the extreme case being a monetary union with a single currency), pegging to the dollar under free capital mobility implies the abandonment of an independent monetary policy. Since the host country has to match domestic interest rates to the US level to maintain the parity rate, it has little room to use monetary policy to restore stability when experiencing economic shocks. For countries whose business cycles synchronize with that of the United States, the costs of giving up monetary autonomy are relatively small, as they are likely to pursue a policy stance similar to that of the United States most of the time anyway. For Asian countries facing shocks that affect the two sides of the Pacific asymmetrically, however, leaving domestic interest rates to be determined in the United States may actually be destabilizing. Indeed, the synchronization between Asian and US economic growth has weakened markedly since the Plaza Accord in 1985. The correlation coefficient between Asian and US growth rates fell from 0.731 for the period 1971-84 to -0.193 for the period 1985-98 (Table 1).

The experience of Hong Kong illustrates how pegging to the dollar can be destabilizing in this new environment. The latest economic crisis in Asia, for example, has hurt Hong Kong more than the United States, so that following the monetary policy stance adopted by the United States has actually aggravated and prolonged the recession in Hong Kong. Indeed, Hong Kong has fallen into a liquidity trap where monetary policy has become totally ineffective to stimulate demand because interest rates cannot fall below the US level (currently at 8.75% in terms of the prime rate, which is far above the “natural” floor of zero percent for a liquidity trap). Making things worse, real interest rates in Hong Kong have risen to a double-digit level on the back of a falling inflation rate (Figure 2.)The adverse effect of rising real interest rates on domestic demand has acted to offset the positive effect of deflation (through stimulating exports) on economic growth. This situation contrasts sharply with the pre-crisis period when negative real interest rates (reflecting high inflation) fuelled the bubble economy. In this way, under Hong Kong’s dollar-peg system the counter-cyclical movement in real interest rates (that is, the tendency for real interest rates to decline during a boom and to rise during a recession) has made the economy inherently unstable.

One way for an Asian country to restore monetary autonomy is to adopt a floating exchange rate system. Although with high capital mobility, domestic interest rates are closely linked to the overseas level, the monetary authorities maintain control over the money supply, and monetary policy works mainly through its impact on the exchange rate. When imperfect substitution between assets denominated in domestic and foreign currencies is taken into consideration, there should also be some room for domestic interest rates to deviate from the overseas level. In the long run, a floating rate regime allows the host country to choose its own rate of inflation that may differ from its trading partners.

Floating exchange rates have their own demerits, however. Foreign exchange markets are driven by herd behavior rather than by rational expectations, and the post-Bretton Woods period has witnessed high volatility and misalignments (chronic deviations from equilibrium rates) of exchange rates among major currencies that have been allowed to float. Freely floating exchange rates are particularly harmful to developing countries with underdeveloped financial markets and highly open economies. Compared with the industrial countries, their exchange rates would be more volatile and the availability of financial instruments for hedging against exchange risks would be limited.

De-dollarization versus Re-dollarization

The increasing mobility of capital has increased the vulnerability of fixed exchange rate systems to speculative attacks. Hedge funds now have the leverage to finance very large positions in the foreign exchange market, making it more and more difficult for central banks to defend fixed exchange rates, as the recent experience of the Asian countries vividly illustrated.

Defending a pegged exchange rate system can be costly in terms of the need to hold a large amount of precautionary foreign exchange reserves and to maintain high interest rates when under speculative attack. Furthermore, an unsuccessful attempt to defend a peg may be accompanied by adverse consequences. Market participants usually interpret a sharp devaluation as an indicator of failed macroeconomic management, which prompts further capital outflow. By raising the debt burden of banks and companies with foreign debts in local currency terms, a devaluation may lead to serious credit crunch problems and falling domestic investment.

Measures to strengthen the dollar-peg-system, including dollarization and currency boards, have been proposed to avoid currency attacks in Asia’s developing countries.

In its purest form, dollarization amounts to replacing the national currency with the US dollar. By eliminating the national currency, there is no longer any need to defend it by raising interest rates or by intervention in the foreign exchange market. In the absence of foreign exchange risks, domestic interest rates should follow closely rates in the United States. This benefit, however, comes at the cost of forgoing once and for all the option(s) to change the exchange rate and/or to pursue an independent monetary policy, which may be very high given the lack of synchronization between Asian and US economic growth. At the same time, dollarization also implies a loss of seigniorage for the host country, as the monetary authorities need to draw down foreign exchange reserves (thus forgoing interest income) and/or borrow dollars in international markets (thus incurring interest costs) to redeem the local currency in circulation. The central bank also loses its function as the lender of last resort, because it can no longer provide liquidity (by printing money) to the banking system in case of a crisis.

Short of eliminating the local currency, some countries have sought to enhance credibility by adopting currency boards. This involves rigidly linking the value of domestic money to that of a foreign currency (the dollar in the case of Hong Kong) and tying the domestic monetary base firmly to the level of foreign exchange reserves through legislation. Unlike complete dollarization, currency boards allow the retention of seigniorage and the option of devaluation in case of emergency. The latter may, however, turn out to be a disadvantage in some circumstances. Since countries with currency boards have not eliminated all possibility of devaluation, they are not fully immune from speculative attacks. The automatic adjustment mechanism under a currency board hinges crucially on interest arbitrage. Under normal conditions, a deterioration in the balance of payments position that puts downward pressure on the exchange rate, for example, would shrink the money supply and put upward pressure on domestic interest rates, which in turn would attract capital inflows to support the exchange rate and to keep interest rates from rising. Should confidence in the authorities’ ability (and determination) to defend the official parity fade as a result of some external shocks, however, containing capital outflow would require maintaining domestic interest rates at an extremely high level, with adverse effects on economic growth and the soundness of the financial system. Furthermore, since under a currency board system only the monetary base is supposed to be backed up by foreign exchange reserves and the credit created by the banking sector is many times this amount, the authorities may possibly run out of reserves amidst intense speculation. In this case, they have no choice but to abandon the official parity, or even the currency board system itself. Hong Kong’s experience suggests that a currency board system is too rigid to allow pursuing an independent monetary policy, but not rigid enough to convince speculators that the risk of devaluation is zero.

Polar versus Intermediate Regimes

The selection of a country’s exchange rate regime is not a dichotomous choice between fixed and floating exchange rates but involves a spectrum of options including-in order of descending commitment to maintain fixed rates-monetary union, currency board, adjustable peg, crawling peg, basket peg, target zone or band, managed floating, and free floating. With various types of intermediate regimes becoming more vulnerable to speculative attacks in this age of global financial integration, some analysts have argued that policy authorities are left with no options but to choose between the two polar regimes: monetary union at the fixed end and free floating at the flexible end. As discussed above, however, joining a monetary union (as in the case of dollarization) involves a complete abandonment of monetary autonomy, while free floating is likely to be accompanied by volatility and misalignments of exchange rates. Indeed, the argument in favor of polar regimes over intermediate regimes has been based on the firm belief that intermediates regimes are not viable rather than that the polar regimes are more desirable. A preferred approach is to improve upon the intermediate regimes with explicit exchange rate targets to make them less vulnerable to currency speculation.

In an attempt to combine the merits of both the fixed and floating rate regimes while minimizing their demerits, John Williamson of the Institute for International Economics has proposed a crawling band system wherein a central bank undertakes a public obligation to maintain the exchange rate within a wide, publicly-announced band around a parity that is periodically adjusted in relatively small steps to keep the band in line with economic fundamentals. A band performs the function of guiding the market where the equilibrium rate lies, and thus makes expectations stabilizing. While the exchange rate is free to fluctuate within the band, the central bank is obliged to intervene at the edge of the band to prevent the rate from going outside. The parity rate should be defined in terms of a basket of currencies instead of a single currency, and it should be adjusted not only to offset inflation but also to reflect permanent changes in fundamentals. At the same time, the band should be wide enough to allow the authorities more room to pursue an independent monetary policy.

The proposed crawling band system, like other intermediate regimes, has been criticized as being vulnerable to currency attacks, but it can be strengthened by various measures, including capital controls and international frameworks to stabilize exchange rates among major industrial countries as well as among the Asian countries.

Capital Controls versus Liberalization

Exchange rate policy does not work in isolation and should be considered in conjunction with other macroeconomic policies. Economic theory suggests that it is difficult, if not impossible, for a country to achieve simultaneously fixed exchange rates, free capital mobility, and an independent monetary policy (known as “the impossible trinity”), but a country under a fixed exchange rate system can restore monetary autonomy by imposing capital controls (Figure 3).

Although capital inflow has played a major role in Asia’s economic development, with imperfect financial markets perfect capital mobility may lead to market failures, and some form of government intervention should provide a second-best solution. The literature on the sequencing of economic reform, which has identified prerequisites for a successful opening up of the capital account, provides the strongest theoretical base for a developing country to impose capital controls. The latest crisis in Asia has confirmed that a developing country should not liberalize capital account transactions at a too early stage when banks’ risk appraisal is inadequate and monetary control is difficult. Premature opening up would not only lead to a misallocation of social resources but also invite speculation. The recent experience of the ASEAN countries has shown that this cost can be very high. In contrast, China is fortunate that it has maintained tight controls on capital flows, which insulate the country from speculative attacks on one hand and enhance the room to stimulate domestic demand by pursuing an expansionary stance of fiscal and monetary policies on the other.

Capital controls should also be considered as emergency measures to prevent systemic risks when speculation threatens the soundness of the financial system and deprives the government of tools of macroeconomic management. Indeed, in an attempt to stabilize the exchange rate and to reclaim autonomy over its monetary and fiscal policies, Malaysia imposed drastic capital controls in September 1998. In retrospect, these goals have been broadly achieved, and Malaysia has also been successful in using the breathing space to restructure its financial and corporate sectors.

At the onset of the Asian crisis, there was little sympathy in the West towards the hard-hit countries, which were said to be plagued by “crony capitalism” and “moral hazard problems.” With the negative repercussions of the crisis being felt not only in other emerging markets but also Wall Street in the summer of 1998, however, major shortcomings intrinsic to the current international financial system-excessive exchange rate volatility, speculative capital flows, and inadequacy of the IMF as a lender of last resort-also came to be widely recognized. This has led to more balanced discussions concerning the causes and policy implications of the Asian crisis, and prompted joint efforts by major industrial powers to reform the “international financial architecture.”

Table 1. Correlation between Asian and U.S. Economic Growth Rates [Back]

  1971-84   1985-98 (1985-96)
Asia 0.731 -0.193 0.175
China 0.139 0.161 0.262
Korea 0.584 -0.298 -0.087
Taiwan 0.857 -0.013 0.090
Hong Kong 0.705 -0.102 0.097
Singapore 0.461 -0.156 -0.074
Indonesia 0.436 -0.321 -0.282
Thailand 0.545 -0.343 -0.068
Malaysia 0.537 -0.343 -0.281
Philippines -0.189 0.204 0.243
Japan 0.616 -0.163 0.066

Note: Asia = NIEs + ASEAN + China.
Source: Compiled by the author based on official statistics of countries concerned.

Note: The views expressed in the CNAPS Working Paper Series are solely those of the authors and should not be attributed to the staff, officers or trustees of the Brookings Institution. The views expressed do not represent the views or policies of any particular institution or government.
The authors welcome comments.