The Impact of Nafta on Mexico's Trade and Growth: An Econometric Exploration

Introduction It is known that Mexican trade has expanded significantly since 1983 to date. What has been the contribution of the North American Trade Agreement (NAFTA) to the expansion of Mexican trade? Has the expansion of Mexican trade contributed to the growth of the Mexican Economy? This paper investigates the historical relationship between trade and growth in Mexico, from 1970 to 1998. More specifically, it measures the importance of NAFTA on the opening of the Mexican economy. The above period includes two different stages of the Mexican economy. The first one, from 1970 to 1982, is a baseline for this research and represents the last phase of the industrial-import substation economy, characterized by widespread government interventions in the economy and by the OPEC bonanza. The second stage begins after the debt crisis of 1982 and runs up to the present. It is characterized by economic reforms such as privatization and liberalization of international trade. These economic reforms were later followed by democratization processes such as peasant revolts in Chiapas, and by urban political actions to establish a multiple party system. There exists a large empirical and theoretical literature on the impact of trade on growth during the 1970s and 1980s. Its findings show a positive correlation between free trade and growth (Edwards 1992; Barro and Sala-I-Martin 1995; and Sachs and Warner 1995). There exists also a literature that explains the channels through which free trade leads to foster growth (Grossman and Helpman 1989; Romer 1990; Vamvakidis 1998; and Manuel R. Agosin 1998.) There is also an emerging literature that represents in a more

1

explicit way the effect of institutional change on economic development (Campos and Nugent 1998). For the case of Mexico, Lustig's work (1998) is a major source for explaining the impact of structural adjustments on Mexican growth. It also advances explanations for the lower pace in most recent years. This paper follows the above research tradition for the case of Mexico. It studies the effectiveness of trade liberalization, and in particular of NAFTA, on the expansion of international trade, and the impact of trade on economic growth during the liberalization stage of the Mexican economy. This study builds on the sources-of-growth model, used by Barro and Sala-I-Martin (1995) with cross-section data, but here applied to one country over time. In explaining the expansion of trade, it builds on the propositions of the new institutional economics that treat institutions as explanatory variables. I. Institutional changes and opening of the Mexican economy The establishment of the NAFTA agreement between Mexico, Canada and the United States has been a major mechanism to open the Mexican economy to both trade and foreign investment. NAFTA, however was not the first nor the only institutional change aimed at expanding trade and inducing growth in Mexico. Two Mexican administrations, those of de la Madrid and C. Salinas de Cortari, implemented substantial unilateral trade liberalization policies before NAFTA. These paper studies, among others, what has been the relative importance of these two liberalization efforts. The expansion of trade also depends on macroeconomic policies such as the monetary system and exchange rate regimes. Have exchange policies been effective and important enough on the expansion of Mexican trade during the second stage? If so, have they operated as substitutes or complements for institutional changes? Using regression analysis, this study investigates the importance of trade liberalization on the rapid growth of Mexican trade since 1986 (Table 1). Trade, the dependent variable, is

ўА Carlos A. Benito is Professor of Economics in the Economic Department, at Sonoma State University.

2

measured by the proportion of exports plus imports relative to GDP. Trade liberalization began with Mexico's unilateral reduction and homogenization of import tariffs, implemented by the administrations of President de la Madrid and President Salinas de Cortari. These changes were followed by mutually agreed reductions of trade and investment barriers through the NAFTA framework. Mexico unilateral changes and NAFTA can be considered complementary explanatory factors. In this model, unilateral changes and NAFTA are treated as exogenous factors. A dummy variable with value one since 1986 to 1998 represent unilateral trade liberalization and a dummy variable with value one since 1994 to 1999 represents NAFTA contribution to liberalization. By these means, the impact of these two changes is compared against the base line of the previous stage (1970-82). Finally, the effect of exchange rate policies is controlled with the variable devaluation of the Mexican peso1. Two alternative measures of devaluation are used, the rate of change of the nominal exchange rate, and a dummy variable with value 1 for years when the rate of change was larger than 50%. Another alternative, not explored here, is to use the real exchange rate. benito@sonoma.edu, http://www.sonoma.edu/people/benito 1 In other studies about growth and trade, authors have used the real exchange rate as explanatory variable. Real exchange rates, however, are not policy variables directly controlled by policy makers. They are indirectly determined. Say, in an Australian type of model the real exchange rate is defined as RER = E. pt/Pn Where RER is the real exchange rate, E is the nominal exchange rate, pt is the world price of tradable goods, and Pn is the domestic price of non-tradable goods. In the Australian model, changes in the nominal exchange rate affects the real exchange rate. However, the price of non-tradables is also an important variable determined by the supply of and demand for non-tradable goods as affected by public finance and aggregate demand policies.

3

Table 1 Mexico Trade Liberalization: Summary Statistics Independent Variables Mean Standard deviation Variation coefficient % Openness: (X+M)/GNP) % 28.6 13.4 47 De la Madrid & Salinas' Reforms .55 .51 92 NAFTA implementation .17 .38 223 Nominal Devaluation Rate, % 32 45 141 Observations 29, between 1970 and 1998 Trade growth 1970-99: 4.8% 1986-99: 6.7% 1994-99: 9.8% Table 2 depicts of OLS regression coefficients. Variables representing institutional changes have a positive sign and are statically significant at probability values larger than 95%. Unilateral trade reforms and the NAFTA agreement explain almost 80% of the variations in trade or opening of the economy (estimation 1). They have a positive relationship with trade, and they are statistically significant with 95% of probability. The coefficients for nominal devaluation (estimations 2 and 3) have positive sign, as expected, but they are not statically