NAFTA and Mexico's current economic crisis: short-run and long-run perspectives. (North American Free Trade Agreement)

INTRODUCTION

Four main propositions form the core of this article. First, Mexico's current economic crisis has important consequences for the eventual success or failure of the North American Free Trade Agreement (NAFTA) and its possible extension to other nations (NAFTA-PLUS). Second, the Mexican crisis is not a short-run problem as asserted by both U.S. and Mexican policy-makers. The crisis in Mexico reflects long-term structural problems including employment generation, low productivity per worker, a large external debt, a highly skewed income distribution and inadequate investment in infrastructure. The third proposition follows logically from the second: the most recent financial aid package, which is based on a short-term perspective, can at best provide only short-run stability to the Mexican economy. Fourth, it is possible to design appropriate policies within the framework of NAFTA to address Mexico's long-term structural problems.

The organization of the article is as follows. The next section contains a brief methodological note. Then I describe the origins of the current crisis from a short-term perspective and examine the short-term consequences of the crisis and the policy response to the crisis. The following section makes the case that the current crisis reflects long-term structural problems. The last section outlines an alternative policy designed to alleviate Mexico's long-term structural problems.

A METHODOLOGICAL NOTE

The methodology used in this article consists of logical deductions based on elementary economic principles. The main concepts are from macroeconomic theory and international trade theory including exchange rate theory. These ideas are supplemented with selected facts from the recent economic history of Mexico. The concept of economic time is essential to the analysis.

Economists make a fundamental distinction between short-run and long-run economic activity. In macroeconomics as well as microeconomics this distinction in timing is crucial since both the nature of the analysis and the policy implications that may be derived from the analysis will be affected. For example, macroeconomists use substantially different analytical tools to examine short-run fluctuations in economic activity (the business cycle) than they use to examine problems associated with long-run economic growth. Appropriate policy recommendations to cope with a recession (a short-run phenomenon) are very different from policy recommendations concerning long-run growth.

During the NAFTA debate in 1993, proponents of the agreement frequently cited results from Computable General Equilibrium (CGE) models concerning the likely effects of the agreement.(1) While the models differed in terms of specification and purpose, the authors of the models reached remarkably consistent macroeconomic conclusions. NAFTA, by lowering tariffs and increasing trade, would have positive effects on production, prices, and employment in the three nations. In relative terms, the CGE models predicted that the largest positive effects of NAFTA would be observed in Mexico.

While CGE models were the analytical device of choice for many economists who examined the potential effects of NAFTA, these models have been strongly criticized and are inappropriate tools for an examination of the current crisis in the context of NAFTA. The models have been criticized particularly for their reliance on neoclassical theory which has a built in bias towards "free-trade." It is difficult to imagine a neoclassical model indicating that there would be no gains from trade. Second, the models are essentially static (or comparative static) representations of a dynamic problem. Third, the estimation of the parameters of the models could not be accomplished with reference to data on a NAFTA-like free trade agreement - none had previously existed. Fourth, econometric models (CGE or not) are most likely to fail when they are most needed.(2) That is, when conditions change rapidly (as in the current crisis) the models are least likely to be useful. Given the previous criticism of these models and the large shocks to the Mexican economy being considered here, no formal model will be developed.

THE ORIGINS OF THE CRISIS: A SHORT-TERM PERSPECTIVE

From a short-term perspective, the beginning of the Mexican crisis can be dated with considerable precision. On December 19, 1994 Mexican Treasury Secretary Jaime Serra announced that there would be a one time 14 percent adjustment in the peso-dollar exchange rate. Because Mexico's new President, Ernesto Zedillo Ponce de Leon, (who took office on December 1) had previously announced that there would be no devaluation of the peso, Mr. Serra was careful to explain that this "adjustment" was not a devaluation.(3) The peso, which had traded for between 3.0 and 3.4 to the dollar for most of the year, reached 4.0 to the dollar by the end of the next business day. Before the end of the month, it was apparent that the new exchange rate could not be maintained. By late January 1995, the peso fell to six to the dollar. The peso-dollar exchange rate reached a peak of slightly more than 7.0 in late February. By late May, the peso was selling for slightly less than 6.0 to the dollar.

The large, sudden and apparently unanticipated devaluation of the peso signalled the start of Mexico's current crisis. The devaluation was a highly visible symptom of the crisis but was neither the cause of the crisis nor Mexico's most serious problem. An examination of the timing of the devaluation will shed light on this point.

Purchasing power parity (PPP) theory can be used to examine the contention that the peso had been over-valued relative to the dollar for three or four years.(4) As can be seen in Table 1, the peso-dollar exchange rate remained relatively stable between 1991 and 1994 even though the inflation rate in Mexico, as measured by the Consumer Price Index (CPI), averaged 15.1 percent annually while in the U.S. the CPI averaged only 4.6 percent per year. PPP theory suggests that, in the absence of intervention in foreign exchange markets, large inflation rate differentials will, at least eventually, lead to a currency depreciation in the nation with the higher inflation rate.

The peso-dollar market was not, however, free of intervention. The Salinas Administration had two main reasons for wanting to avoid a major devaluation. First, during 1992 and especially during 1993 a major devaluation would have been a serious threat to the passage of NAFTA by the …