Massive capital flows have been at the heart of every major currency crisis in the 1990s. Whether Mexico in 1994, Thailand in 1997, Russia in 1998, or Brazil in 1999, the stories are depressingly similar. High domestic interest rates, perceived stability stemming from rigid exchange rates, and apparently rosy economic prospects all attracted foreign funds into these emerging markets, lifting stock prices and helping finance bloated current account deficits. When these funds eventually trickled to a halt or reversed direction, significant corrections in macroeconomic policies became necessary. But governments often watered down or delayed reform, which increased investor uncertainty and nervousness over risk. As a result, more and more capital poured out of the countries and foreign exchange reserves dropped to dangerously low levels. Eventually, the governments had no choice but to abandon their pegged exchange rates and float their currencies. In Brazil and Russia, runaway fiscal deficits made the situation even more explosive.

In the aftermath of these crises, a number of influential academics have argued that the wild capital movements wrought by globalization have gone too far. In the words of Paul Krugman, "sooner or later we will have to turn the clock at least part of the way back" to limit the free mobility of capital. Bolstered by the growing number of capital-controls advocates, proposals for a new international financial architecture have focused on two types of controls: restrictions on short-term capital inflows, similar to those implemented in Chile between 1991 and 1998; and controls on capital outflows, like those Malaysia imposed in 1998. Both schemes try to reduce the "irrational" volatility inherent in capital flows and foster longer-term forms of investment, such as direct foreign investment, including investment in equipment and machinery.

Despite their good intentions, these proposals share a common flaw: they ignore the discouraging empirical record of capital controls in developing countries. The blunt fact is that capital controls are not only ineffective in avoiding crises, but also breed corruption and inflate the costs of managing investment.


Chile, which experimented with short-term capital controls during 1978-82 and 1991-98, has become a favorite test case for proponents of such measures. In both episodes, foreigners wishing to move short-term funds into Chile were required to first deposit their money with Chile's central bank for a specified amount of time -- at no interest. By stemming inflows, the policy aimed to mitigate capital volatility, prevent the currency from rising too quickly (a common result of accelerated capital inflows), and increase the central bank's control over domestic monetary policy. From 1978 to 1982, the controls were particularly stringent; foreign capital was virtually forbidden from entering the country for less than five and a half years. In this way, it was thought, the country would not be vulnerable to short-term speculation.

Proponents of controls cite all the above facts but miss the bigger picture. Indeed, their brand of wishful thinking misreads Chile's history and oversells the effectiveness of this policy. What you do not hear from them is that the draconian restrictions on capital inflows could not prevent Chile from going through a traumatic economic crisis from 1981 to 1982, which caused a peso depreciation of almost 90 percent and a systemic banking collapse. The problem lay with the largely unregulated banking sector, which used international loans to speculate on real estate and lend generously to bank owners, ultimately creating an asset-price bubble. When it burst, loans could not be repaid. Hence, many banks went under and had to be rescued by the government at a very high cost to taxpayers. A massive 1986 banking reform finally put an end to that by establishing strict guidelines on bank exposure and instituting rigorous on-site inspections. A healthy, strong, and efficient banking system emerged as a result, which to this day has helped Chile withstand the most recent global turmoil.

This historical episode underscores a key factor in evaluating restrictions on capital mobility: without effective prudential banking regulations, restrictions on capital inflows alone are unlikely to reduce a country's vulnerability. Moreover, capital controls may foster a false sense of security, encouraging complacent and careless behavior by policymakers and investors alike. South Korea's recent experience is a case in point. Until late 1997, international market players and local policymakers believed that Seoul's restrictions on capital mobility would inoculate the country from a currency crisis. Indeed, even after giving South Korea's central bank and private banks their next-to-lowest rating in early 1997, Goldman Sachs still argued that the nation's "relatively closed capital account" necessitated the exclusion of such gloomy data from its overall assessment of South Korea's financial vulnerability. Hence, Goldman Sachs played down the extent of the Korean won crisis throughout most of 1997. Had it correctly recognized that capital restrictions cannot truly protect an economy from financial turbulence, it would have accurately anticipated the South Korean debacle just as it forecast the Thai meltdown.

Brazil is another example that capital-control advocates should reconsider. Restrictions on short-term capital inflows in 1997 and 1998 lulled Brazilian policymakers into complacency. They repeatedly argued that their controls would preclude a Mexican-style currency crisis. As it turned out, they were wrong. Once the collapse of the Brazilian real became imminent in the autumn of 1998, domestic and foreign investors alike rushed to flee the country -- just as in Mexico in 1994.


Not surprisingly, most supporters of capital controls have focused only on Chile's experience during the 1990s, when 30 percent of all capital inflows had to be deposited for one year, at no interest, with the central bank, and direct foreign investment was required to stay in the country for at least one year. This policy was implemented in June 1991, when the newly elected democratic government of President Patricio Aylwin became concerned over the exchange rate and inflationary effects of the rapidly growing capital inflows. Aylwin sought the support of exporters, who resisted a strengthening of the currency, while taking a firm anti-inflationary stance. Accordingly, he turned to capital controls.

Despite positive media coverage and popularity with some academics, there is no firm evidence that this policy actually achieved its goals. First, Chile's short-term foreign-denominated debt was almost 50 percent of all debt from 1996 to 1998. Second, capital controls failed to slow the strengthening of Chile's currency. Throughout the 1990s, Chile's real exchange rate rose more than 30 percent despite capital controls. Finally, the argument that restrictions on inflows help increase central bank control over domestic monetary policy is tenuous at best. Tightening restrictions increases domestic interest rates only slightly and temporarily.

Chile's capital controls have also carried another price -- by inflating the cost of capital. Large firms, with their easy access to international financial resources, can always find ways to circumvent the controls; smaller firms are not so fortunate. As a result, a prohibitively high cost of capital not only distorts the true cost of investment but discriminates against small- and medium-sized businesses. In fact, some analysts have calculated that investment costs for smaller firms exceeded 20 percent in dollar terms in 1996 and 1997; larger firms, on the other hand, could access the international market with dollar loans at a cost of only 7 or 8 percent per annum.


As their advocates would have it, temporary controls on capital outflows would allow stricken countries to lower interest rates and implement pro-growth policies without worrying about investors pulling out for fear of devaluation. Controlling capital outflows would also buy economies time to restructure their financial sector in an orderly fashion. Once the economy is back on its feet, so goes the argument, controls can be dismantled.

Again, the historical evidence flies in the face of this reasoning. According to two studies of 31 major currency crises in Latin America, countries that tightened controls after a major devaluation did not post better performances in economic growth, job creation, or inflation than those that did not. The Latin American debt crisis of the 1980s illustrates how ineffective these controls really were. Nations that imposed controls on capital outflows -- Argentina, Brazil, Mexico, and Peru -- muddled through but suffered rising inflation, worsening unemployment, and a long and painful decline in growth. Moreover, the stricter controls encouraged neither macroeconomic restructuring nor orderly reforms aimed at increasing efficiency and competitiveness. In fact, the opposite happened: politicians experimented with populist policies that ultimately deepened the crisis. Mexico nationalized the banking sector and confiscated dollar-denominated deposits. Argentina and Brazil launched new currencies while setting price controls and expanding public spending. In Peru, stricter controls on outflows allowed President Alan Garcia to whittle away the basis of a healthy and productive economy, squandering international reserves and pursuing a hyperinflationary policy. To make things even worse, in none of these countries did controls on capital outflows successfully stem capital flight.

Chile and Colombia, which did not tighten controls on capital outflows, provide an interesting contrast. These two states attempted to restructure their economies. Chile even implemented a modern bank supervisory system that greatly reduced domestic financial fragility. Accordingly, both countries emerged from the debt crisis significantly better off than the rest of the region. In fact, they were the only two large Latin American countries to experience positive growth in per capita gross domestic product and real wages during the "lost decade" of the 1980s.


The recent financial crises have dealt a severe blow to the International Monetary Fund's credibility. The IMF badly miscalculated the Mexican collapse of 1994, prescribed the wrong policies in East Asia in 1997, and offered vastly inadequate rescue packages to Russia and Brazil in 1998. This succession of embarrassments reflects the fact that the IMF'S structure does not allow it to operate effectively in the modern world economy, where investor confidence and the frank, uncensored, and prompt dissemination of information are crucial. Sadly, international politics is likely to stand in the way of true IMF reform. After much talk about a new architecture, we will probably end up with a slightly embellished IMF that will continue to miss crises, throw good money after bad, and ultimately try to rationalize why currency crises persist.

Economists have long recognized that the issue of international capital movements is highly complicated. In the absence of strong financial and banking supervision in both lending and borrowing countries, unregulated capital flows may indeed be misallocated, generating major disruptions in the receiving nations. Many academics have rightly argued that relaxing controls on capital movement should therefore follow, not precede, market-oriented macroeconomic reform and the establishment of a reliable supervisory system for domestic financial markets. Governments should lift controls on capital movements carefully and gradually -- but they should be lifted.

We must understand what capital controls can and cannot do. The historical record shows convincingly that, despite their new popularity, controls on capital outflows and inflows are ineffective. The best prescriptions to combat financial turmoil, now as then, are sound macroeconomic policies, sufficiently flexible exchange rates, and banking reforms that introduce effective prudential regulations and reduce moral hazard and corruption. Without a solid financial groundwork, emerging markets will remain as fragile as a house of cards, easily blown down by the first breezes of turbulence.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Sebastian Edwards is Henry Ford II Professor of International Economics at UCLA'S Anderson Graduate School of Management. He was Chief Economist for Latin America at the World Bank from 1993 to 1996.
  • More By Sebastian Edwards