This message contains the text of Sidney Weintraub's January Issues in International Political Economy. In this issue, Dr. Weintraub compares exchange rate management in Mexico, Argentina, and Brazil. To go directly to the web version, click on:

http://www.csis.org/simonchair/issues200201.htm

For further information, please contact Veronica Prado at (202) 775-3123 or vprado@csis.org. She will be replacing me as Sidney Weintraub's research assistant effective this Friday, January 18th.

Best wishes to all,
Rebecca Tunstall


ISSUES IN INTERNATIONAL POLITICAL ECONOMY
January 2002, Number 25

REJECTING THE LESSONS OF EXPERIENCE
Sidney Weintraub

The turmoil that Argentina is now going through is so costly for the majority of the Argentinean population that it demands some answers to the question, "Did it have to happen?" The same question arose when the Mexican financial structure imploded in 1994. My single word answer in both cases is "No." In each case, the disaster followed a similar-although not identical-trajectory; and in each case the political and technical authorities made a bet that they could manage discredited practices. And, in each case, they failed, thereby visiting great suffering on their populations.

The precipitating event of the Argentine debacle was an overvalued and, ultimately, untenable exchange rate. The convertibility scheme adopted in 1991 was brilliant in its simplicity. By assuring the public that enough dollars were to be held in official reserves to cover the monetary base and thereby guarantee that pesos could be exchanged for dollars, on demand, even a skeptical public found the approach credible. Runaway inflation was stopped in its tracks. Overall economic growth rates soared, at least for a while. The technique was a variant of the Mexican experiment that began in 1987: using the exchange rate as the anchor to defeat inflation.

What "did in" the Argentine experiment was the appreciation of the U.S. dollar. This led, inevitably under one-to-one convertibility, to the appreciation of the Argentine peso, and despite the preaching of ideologues that it is immoral to alter an exchange rate, the price of a country's money-its exchange rate-does affect its exports and imports. On top of this, Brazil devalued its currency, the real, in 1999, and what Argentina faced was the appreciation of its currency while that of the country that took one-third of its exports depreciated. Argentina nevertheless retained the overvalued exchange rate and, in the process, further destroyed the country's competitive position. Inflation was kept in check, but the country has suffered economic downturn ever since.

This description omits other aspects of Argentina's economic, political, and social situation. These include relatively modest fiscal deficits (the public sector borrowing requirement averaged about 2 percent of GDP between 1991 and 1998, at the national and provincial levels combined); growing social inequality and poverty; too many feckless politicians who looked at government service as a source of enrichment; and reckless borrowing, public and private, domestic and foreign. Argentina met its debt obligations by more borrowing, but this was a large Ponzi scheme because the inability to earn enough foreign exchange from exports meant that the rigid one-to-one structure had to collapse, sooner or later.

Mexico, in 1987, adopted a form of price controls, under what was called a pacto de solidaridad, to lower its inflation. The pacto, or agreement, called for the private sector to limit price increases, labor to limit wage demands, and the government to control its expenditures. When the pacto started, Mexico devalued its currency sharply to leave room for appreciation of the peso in order to prevent devaluation from raising inflation. The pactos were kept in place, with annual modifications, from 1987 through 1994. The anchor for Mexico's anti-inflation policy was an increasingly overvalued peso. Inflation declined gradually, year after year, rather than abruptly as it did in Argentina after convertibility, but economic growth in Mexico in these years was modest.

As in Argentina, the Mexican political authorities (the president) and the technicians (the finance minister, with the cooperation of the head of the central bank) were unwilling to change the exchange-rate structure, a form of crawling peg. Instead, they took the risk that they could delay devaluation until some time in the future. There was general agreement that the Mexican peso was overvalued, and the way the authorities dealt with this in foreign borrowing was to issue short-term notes that had a dollar guarantee-the so-called tesobonos. This house of cards collapsed in December 1994, about three weeks after the previous president left office. The Mexican authorities gambled, and the population lost.

As in Argentina, there were other elements in the Mexican collapse. There were assassinations of political leaders during 1994; the U.S. Federal Reserve raised the federal funds rate six times during the year, seriously affecting Mexico; the government budget was kept in balance, but development banks were allowed to provide uncollected credit that amounted to some 3 percent of GDP; political corruption continued much as before; and foreign borrowing, on short terms and at dollar equivalents, rose sharply even as official reserves declined from capital flight and efforts to support the exchange rate. In the end, however, what did Mexico in was an overvalued exchange rate that was kept in place far too long.

In the case of Brazil, the scenario unfolded differently. The real plan was put in place in mid-1994 in an effort to reduce inflation without burdening economic growth. As in Mexico, in Brazil the stable exchange rate worked and inflation was reduced. Brazil weathered the tequila effect of December 1994, when the Mexican peso depreciated, without any sharp change in the real plan. However, Brazil was hit sharply by the aftereffects of the Russian debt default in 1998 because investors either had to cover positions or became nervous about keeping capital in Brazil given its precarious situation. In January 1999, Brazil devalued the real and took other corrective measures. Brazil did not have the turmoil that either Mexico or Argentina faced when they were forced to devalue under the most adverse circumstances. Brazil acted more quickly, before a deepening crisis made action exceedingly costly.

The currencies of Brazil and Mexico are both floating, and thus far, the mechanism is working. Argentina has announced that it will allow its currency to float for domestic transactions but that it will be controlled for foreign trade. That is, Argentina will have a dual exchange- rate system, justified by the assertion that because the one-to-one structure was in effect for so long, the country needed six months for moving to a single, floating currency. Argentina's new president, Eduardo Duhalde, has also stated that the country's opening to imports has been a disaster and that trade policy will become more nationalistic (protectionist).

At this point, two comments. Dual exchange rates have a history of failure, especially by abetting corruption. Argentina has tried protectionism before, especially under Juan Per?n, and it did little other than to increasingly impoverish the country. In the interest of Argentina's future, one can only hope that the dual exchange-rate system does not last very long and that the economic nationalism will not be too severe.

The deeper issue is to ask why government leaders pursue policies that have a history of failures. This is a task for psychologists, not economists, but I will give it a try.

		?	It is politically unpopular to devalue a currency; therefore, why not push off the unpleasantness for as long as possible? The suffering for doing this will come later, perhaps on someone else's watch.
?	Memories are short-those of the politicians and technicians who repeat past mistakes, and of the general public. In addition, memories often gild the past, forgetting that Latin America's poverty and inequality were created during periods of past protectionism.
?	Intelligent leaders become wedded to policies they put in place and are often emotionally unable to change what they have been doing. Domingo Cavallo was fabulously successful as economics minister in Argentina in the early part of the 1990s and dismally incorrect in his actions when he returned in the De la R?a government in the same position. In Mexico, in 1994, both President Salinas and Finance Minister Pedro Aspe were so adamant about not devaluing the peso that they excluded technicians from policy deliberations when they did not agree with this position.

It is understandable that many analysts and business leaders prefer fixed exchange rates. When things are going well, this provides a degree of certainty. However, the time to judge the effectiveness of an exchange-rate mechanism is when the system is under pressure. This explains the attractiveness of floating exchange rates for countries prone to economic shocks. It is clear from what is happening in Argentina and what took place in Mexico in 1994 that fixed rates-even a relatively hard fix of the type that Argentina had-can spell disaster when a country is under economic pressure. Perhaps this lesson will stick-at least for a time.

Issues in International Political Economy is produced by the William E. Simon Chair in Political Economy at the Center for Strategic and International Studies (CSIS), a private, tax-exempt institution focusing on international public policy issues. Its research is nonpartisan and nonproprietary.

CSIS does not take specific policy positions. Accordingly, all views, positions, and conclusions expressed in this publication should be understood to be solely those of the author.

? 2002 by the Center for Strategic and International Studies.