June 1998, Volume 35, Number 2

The financial crisis that erupted in Asia in 1997 vividly demonstrated the risks associated with free capital flows. But capital mobility is a goal worth pursuing, given the potential benefits. With the right policies, countries can manage the risks while increasing their access to global financial markets.

But capital flows also expose countries to external disturbances and can have a destabilizing effect. The dangers of sudden outflows are well understood, but capital inflows also carry risks—they may create difficulties for monetary policy management and inflation control as well as for exchange rate stability and export competitiveness. This is particularly true in countries with vulnerable financial sectors and inappropriate macroeconomic policies.

The long-running debate about the desirability of unrestrained capital movements intensified in the wake of the financial crisis that rocked several Asian economies in 1997 and 1998 (see box). Do capital controls have a role in today's world economy? What other steps can be taken, at both the national and the international levels, to help countries minimize the potentially disruptive effects of capital flows on their economies?

Capital flows and the Asian crisis

The Asian countries hit hardest by the crisis of 1997 and 1998—Indonesia, Korea, and Thailand—had pursued diverse approaches to opening up their capital accounts. Although they achieved only partial liberalization, they all experienced significant growth in capital inflows (until the crisis, almost half of total capital inflows to developing countries went to Asia—nearly $100 billion in 1996). Indonesia liberalized outflows relatively early and inflows only gradually. In 1989, it eliminated controls on foreign borrowing by banks but reintroduced them two years later because of concerns about excessive borrowing. It continued, however, to liberalize inflows to corporations, allowing borrowing for trade finance, sales of securities to nonresidents, and foreign investment in the domestic stock market. Korea took a gradualist approach. It liberalized outflows first and did not begin liberalizing inflows into its securities markets until the mid-1990s. In 1992, nonresidents were given limited access to the Korean stock market, and the types of securities that resident firms could issue abroad were expanded. Foreign exchange banks were authorized to borrow abroad, but direct foreign borrowing by corporations was controlled. In contrast, Thailand sought to attract foreign inflows, offering tax incentives to foreign investors, setting up a special facility (the Bangkok International Banking Facility) to channel inflows through the banking system, and allowing foreign investment in Thai securities markets. Capital outflows were liberalized gradually, however. The impact of capital account liberalization ultimately depends on how efficiently capital flows are used. In all three countries, liberalization brought significant growth to capital inflows channeled through domestic banking systems that were inefficient and unsophisticated and that contributed to their short-term maturities. Excessive lending to interrelated entities distorted incentive structures. The rapid expansion of bank credit strained credit assessment capabilities, and funds flowed into unprofitable or speculative activities. Policy weaknesses led to the accumulation of unsustainable levels of foreign debt by domestic firms. The three countries had pegged their currencies to the U.S. dollar, and the high domestic interest rates needed to sustain the pegs both attracted short-term inflows and encouraged domestic firms to borrow in foreign currencies. Much of this foreign borrowing was unhedged on expectations that the pegged exchange rates would be maintained indefinitely. Government guarantees, implicit as well as explicit, encouraged lending by foreign institutions and undue risk taking by domestic firms. Risk aversion by foreign lenders also biased inflows toward the short term. All these factors added to the risks of abrupt capital flow reversals. On the whole, however, markets took an overly sanguine view of the three economies based on their historically solid performances, inter alia because there was insufficient accurate information on current economic conditions. When signs of weakness began to appear, markets overreacted, triggering massive capital outflows and causing the currencies to depreciate to levels far below those that would have been required to correct the initial overvaluations. In an effort to reduce inflows before the crisis and to stanch outflows during the crisis, Indonesia and Thailand resorted to capital controls. The controls not only failed to achieve the desired effect but also exacerbated the problem. They introduced new distortions and sent markets a negative signal during the crisis, discouraging new inflows at a critical juncture. The experiences of the Asian economies demonstrate the need to proceed with caution in opening the capital account, to anticipate and minimize the risks involved. Capital account liberalization needs to be undertaken as an integral part of economic reforms and coordinated with appropriate macroeconomic, exchange rate, and financial sector policies. The issue relates more to the sequence of reforms than to their speed. A number of countries would have benefited from relatively rapid liberalization. Increased competition in domestic financial markets would have resulted in stronger markets and institutions, a domestic interest rate structure that better reflected risks, and a weakening of vested interests, thus fostering the development of efficient financial systems, so notably lacking in many Asian economies. Most important, the liberalization of inflows through the banking system should have been supported by banking reforms, as well as by greater transparency and better information flows, to enable markets to make informed decisions and reduce the risk of subsequent reversals of market sentiment. It is also critical to recognize that a country's economic policies will be constrained by its choice of exchange rate arrangements. Therefore, attention has to be given to the maintenance of an appropriate, sustained, and consistent policy mix to prevent a country from attracting short-term inflows on such a scale that they cannot be absorbed.

The Bretton Woods system

The questions being raised in the debate on capital mobility are not new. Shortly before the end of World War II, policymakers from 44 countries met at Bretton Woods, New Hampshire, to discuss an institutional framework for the reconstruction of the world economy. Their goals included the establishment of a new, cooperative international monetary order. The key elements of this order were contained in the Articles of Agreement of the International Monetary Fund, which was established soon after the war ended to oversee the new international monetary system. This document includes provisions according to which member countries may exercise controls "as are necessary to regulate international capital movements" (Article VI, Section 3) and the IMF "may request a member to exercise" such controls (Article VI, Section 1).

At the time, the IMF's founders understandably gave priority to liberalizing trade and removing foreign exchange restrictions on current account flows rather than to liberalizing capital movements. International trade had been disrupted before and during World War II, and getting things back to normal was of the utmost urgency. It must also be remembered that capital movements played a less prominent role in the world economy fifty years ago than they do today.

But the issue of international capital movements and their implications for the management of national economies were debated at length. At the time of the Bretton Woods conference, it was widely believed that capital controls could preserve the independence of domestic policies—and, in fact, were necessary for this purpose. (The desire for a measure of autonomy in national economic policymaking is also behind the recent renewal of interest in capital controls.)

To ensure exchange rate stability in the postwar global economy, the Bretton Woods participants established what was known as the par value regime—a system of fixed exchange rates that could be adjusted after consultation with the IMF. In the early 1970s, however, the par value system was abandoned. It had proved to be incompatible with national economic objectives and with increasing capital mobility. The scale of capital flows and their importance for the world economy had grown considerably; this growth was due to numerous factors but owed much to the opening of trade and current accounts during the 1950s and 1960s.The growing interdependence of the world economy over the past few decades has been a key product of the more open trade relationships fostered by the Bretton Woods regime, which, in turn, helped bring about larger and closer financial and credit links between countries.

An essential legacy (and challenge) of the Bretton Woods order, therefore, has been the expansion of international capital flows and the greater integration of financial markets, particularly those in advanced countries. Another characteristic of the post–Bretton Woods order is that an important proportion of expanding capital flows has been private, rather than official, with commercial bank lending, securities trading, and direct investment flows accounting for a large share of the growth.

Over the past 25 years, many of the IMF's members have relaxed capital controls in the context of a general liberalization and deregulation of domestic financial markets. These developments have sparked debate on the appropriate sequencing of liberalization in two domains—the balance of payments, where the issue is whether the current or the capital account should be liberalized first; and the financial sector, where the issue is whether or not domestic financial liberalization should precede the opening of the capital account.

The role of government

The developments that have taken place in the international economy since the 1970s reflect, of course, many factors. Fundamental among them is a profound evolution in views about the relative roles of government and market forces in the economic process. At the time of Bretton Woods, and for a quarter of a century thereafter, it was generally accepted that governments and their policies were predominant forces in the economy. Governments were expected to take responsibility for basic economic objectives and economic performance.

The views held today as to the economic role of government and the limits of economic policy stand in stark contrast to those described above. First, the consensus is that the role of government is to allow and support, not to restrain or compete with, private initiative. Government's responsibilities for (and its contribution to) economic performance do not include direct management of the economy but, rather, involve maintaining a stable macroeconomic framework; supporting the economic infrastructure (human as well as physical capital); and developing an institutional infrastructure (which encompasses the establishment and safeguarding of appropriate legal, regulatory, and social frameworks; economic incentives; and a competitive, open, and liberal economy).

Second, the consensus that has developed regarding the limits of macroeconomic policy is based on both conceptual analysis and empirical evidence. On the conceptual front, critical for the consensus have been the theory of public choice, which found, based on the tools of market economics, that government officials would tend to maximize their own, as opposed to society's, utility; the rational expectations hypothesis, which established the need for stable government policy; and the related literature on credibility, which stressed the importance of predictable and steady policy courses to avert disruptions to the economy. Today, economic policy is viewed essentially as an instrument for providing a relatively stable framework in which market forces can operate. On the empirical front, most countries—advanced, developing, and transition alike—have moved in the direction of using economic policy as a means for encouraging market forces, rather than as an instrument for molding and competing with them.

The issues at stake

The issue of what is an appropriate policy response to international capital mobility must be examined against this empirical and conceptual background. On the empirical front, the question is: has economic integration gone so far that its costs exceed its benefits? And on the conceptual front, has the case for market forces been stretched to the point where it conflicts with a sustainable, sound balance between national and international interests?

In a nutshell, the dilemma is whether to press ahead with liberalization, handling conflicts in the international economy through the coordination of national economic policies, or to rely on controls. The response can be organized around two hypotheses: (1) market failures exist; and (2) international capital flows not only respond to the relative risks and returns in different national economies but also arbitrage across the spectrum of national economic policies in search of the soundest among them.

Market failures. These failures are often seen as a justification for government intervention. In the context of capital mobility, the issue of market imperfections arises when there is a divergence between capital movements and economic fundamentals—that is, when market discipline has failed to operate. This is the case, for example, when capital flows go to countries with unsustainable policies or, conversely, when financing in international capital markets is either unavailable to, or too costly for, countries with appropriate policy records and prospects.

Policy arbitrage. According to the hypothesis of policy arbitrage, differences in the quality of countries' economic policy management are an important determinant of the scale, terms, and direction of capital flows. Thus, capital will tend to move from countries with relatively weak policies to countries with sound policy records and good prospects—that is, capital markets respond to economic fundamentals. In this context, any conflicts that may arise between national and international interests are related not to the direction of capital flows but to their scale relative to the size of the recipient economies. If an economy is receiving a greater volume of inflows than it has the capacity to absorb, the inflows will pose problems for the management of economic policy—in particular, monetary and exchange rate policies.

It is in this context that arguments in favor of controls on certain types of capital inflows have typically been made. The rationale is that restrictions can complement other, more conventional actions (such as sterilization and fiscal adjustment) to keep inflows commensurate with an economy's absorptive capacity. Viewed from the standpoint of individual countries, such arguments have a measure of validity. But from an international perspective, it does not make sense for countries with sound policies to impose controls on capital inflows. A more rational approach would be to correct policies in those countries where they are inappropriate (and capital flows out). The proper instruments to achieve this are international policy surveillance and conditional lending.

International surveillance

The IMF has a critical role to play in ensuring the orderly liberalization of capital accounts, as well as an important contribution to make to the analysis of capital flows and capital markets. Through its surveillance activities (which consist of monitoring developments in member countries and the international economy and warning of impending problems) and the conditionality attached to its loans (which calls for borrowing countries to make certain agreed policy and structural changes), the IMF can provide capital markets with information on country policies, thus reducing instances of market failure and the need for market arbitrage.

The continued liberalization of capital flows is preferable to "re-regulation" for two fundamental reasons. One is the desirability of observing time consistency in the direction of policy. What would be the good of deregulating domestic financial sectors and liberalizing capital markets today only to revert to capital controls tomorrow? More to the point, policies that emphasize the importance of market forces and give governments the role of providing markets with an institutional framework that fosters appropriate economic incentives have served us well. The other reason is purely pragmatic: capital controls are likely to be ineffective, as they were in Indonesia and Thailand (see box).

Chile is often cited as a country that has successfully used capital controls to limit short-term inflows. While the assessment of the effectiveness of controls is difficult, they appear to have been effective mainly as a temporary measure: short-term inflows declined in the year the measures were introduced only to increase the following year. Moreover, Chile's experience with controls is different from East Asia's, partly because (1) Chile had already addressed serious banking sector weaknesses following an earlier banking crisis, and (2) Chile's more flexible exchange rate policy was more attuned to dealing with significant capital inflows.

In general, then, controls should not be included as a standard weapon in a country's policy arsenal. But they may play a limited and—if properly monitored—constructive role. For example, restrictions may be useful in providing the time needed for policy corrections to take effect, though it must be kept in mind that their presence can also permit delays in the introduction of adjustment measures. It is because controls are a double-edged sword and because of their externalities (controls imposed by one country typically affect other countries adversely) that international monitoring is necessary. Basic principles for such monitoring include evidence that controls are necessary to protect the balance of payments, that they are transitory and exceptional in character, and that policies are being put in place that will eventually eliminate the need for them—the principles the IMF has applied in monitoring restrictions on current account payments and transfers.

What is most important is to keep a steady course and not to backtrack on it arbitrarily. As a general rule, policy on capital movements should favor openness and liberalization. But liberalization is not licentiousness—hence the need for supervision: for a sound accounting, legal, and supervisory framework, as well as for effective machinery for enforcing prudential norms. Sound, clear, and effective prudential norms will be required at the national level. And, at the international level, consistency and harmony among those national norms will be needed to limit the scope for erosion of their effectiveness through "regulatory arbitrage."

We need to be clear on the course we wish to give to the international economy. Integration and interdependence are perceived as desirable objectives when the focus is placed on their potential benefits. But when attention is paid to the conflicts they may cause and the costs that result, they are seen as undesirable. Unfortunately, the benefits cannot be reaped without accepting the risks. The proponents of controls give undue weight to the costs of integration, such as those related to the constraints interdependence imposes on individual country actions. But other means can be found to resolve these conflicts. Important among them are international policy surveillance and conditional external financial support of individual country policies. To enable the IMF to be effective in this area, its membership is considering a proposal for an amendment to the Articles of Agreement that would make currencies convertible for all transactions. The choices we face are clear. We can devote resources to surveillance and conditionality to protect the "soundness" of integration. Or we can instead go the route of controls and re-regulation and seek to avoid the costs of integration by "disintegration."

Manuel Guiti�n is Director of the IMF's Monetary and Exchange Affairs Department.