Alternative Scenarios of U.S.-Mexico Integration: A Computable General Equilibrium Approach

Abstract

This paper analyzes the economic impact on the U.S. and Mexican economies of creating a free trade area (FTA).The empirical analysis is based on a three-country (U.S., Mexico, and rest of world), multisectoral, computable general equilibrium (CGE) model.The model solves for prices, wages, profits, and the real exchange rate that achieve equilibrium in the goods markets; factor markets, and balance of trade.It includes international migration of rural and urban unskilled labor and rural-urban migration within Mexico.The model also solves for equilibrium sectoral production, consumption, exports, and imports in each country.The model represents a methodological advance over earlier multi-country trade models in its functional specification of sectoral import substitution possibilities and its inclusion of migration.The empirical results indicate that: ( 1 ) The removal of import protection in the two countries, with no other changes, will have a tiny effect on the U.S. and a small effect on Mexico.(2) Assuming that the creation of an FTA is accompanied by additional policy changes so that Mexico succeeds in shifting to an open development strategy, trade between the two countries expands and both gain significantly.( 3) The size of the potential gains is sensitive to assumptions about the possibilities for productivity gains and collateral policy changes accompanying the formation of an FTA.(4) The creation of the FTA and associated Mexican growth will generate adjustments in production and employment structure in both .economies, although the effects are proportionately far larger for Mexico.( 5) The impact of an FTA, by itself, on migration is small.On the other hand, successful Mexican growth significantly reduces the pressure for migration to the U.S. (6) The creation of an FTA, by itself, has almost no effect on real wages in either country.If Mexico succeeds in its shift in development strategy, real wages rise in both countries for most labor categories, but relatively more in Mexico.The effect on wages for unskilled workers in the two countries is very sensitive to assumptions about migration behavior.'Other comparators might be relations between Spain, Portugal, or Turkey and the European Community; Korea, Taiwan, and other Asian developing countries and Japan; or Franc-zone African countries and France.In these cases, the links are clearly not as strong as between the U.S. and Mexico.Reynolds and Tello (1983), Hinojosa (1989) and Weintraub (1990) discuss these links.2Wharton Econometrics Forecasting Associates (1984).3Reynolds, Hinojosa, and Bustamante (1991);Carnoy, Daley, and Hinojosa (1990); and Hayes-Bautista (1989) all discuss the implication of social, economic, and demographic trends on labor markets in the U.S. and Mexico.'See Balassa (1989);Chenery, Robinson, and Syrquin (1986); and Corbo, Krueger, and Ossa (1985) for surveys of economic performance in countries which have pursued open development strategies.The discussion below of the nature of the transformation draws on Chenery, Robinson, and Syrquin.'Such models have been widely used to study the impact of proposed trade liberalization in the Uruguay round of GATT negotiations.See Goldin and Knudsen (1990) and OECD (1990) for a summary of this work.The first multi-country CGE model was developed by Whalley (1985) to study the impact of the Tokyo round of GATT trade negotiations.Single-country CGE models used to analyze the impact of the Uruguay Round are surveyed by Robinson (1990).'The imbalance in North America is much greater than that between Northern Europe and Spain, Portugal, and Greece.The U.S.-Mexico gap also is wider than that between Japan and most of its Asian neighbors, except China.'See Gilpin (1989) on the political economy of emerging global trading blocs and a critique by Krasner (1989).Shott (1990) discusses the impact of FTAs in the context of global U.S. trade policy.