NAFTA and Economic Activity along the U.S.-Mexico Border
Peach, James T., Adkisson, Richard V., Journal of Economic Issues
Trade agreements between nations, such as the North American Free Trade Agreement (NAFTA), are important institutions with far-reaching domestic and international consequences.  The NAFTA negotiations during 1992 and 1993 generated an intense debate over the potential macroeconomic consequences of a free-trade agreement for the United States. That debate continues six years after the first provisions of NAFTA took effect on January 1, 1994. 
The purpose of this paper is to examine changes in economic activity in the U.S.-Mexico border region in the years surrounding NAFTA's implementation. We restrict the analysis to the U.S. side of the border. This region deserves special consideration for two reasons. First, because of its proximity to Mexico and its preNAFTA industrial structure, the impacts of NAFTA in the border region were likely to be more intense than in non-border regions of the nation. Second, the region attracted considerable attention during the NAFTA debate. For example, the NAFTA side agreement on environmental issues and the creation of the Border Environmental Cooperation Commission (BECC) were largely the result of concern about the environmental effects of increased NAFTA-related economic activity in the border region. In addition, the creation of the North American Development Bank (NADBANK) was a direct consequence of concern over inadequate border region infrastructure.
The border region has been defined in many different ways [Hansen 1981; Stoddard 1991; Sharp 1997]. Here, the border region is defined to include the 25 U.S. counties that are contiguous to Mexico's northern border. The main focus is on eight border region Metropolitan Statistical Areas (MSAs) that contained 91.4 percent of the region's population in 1998 (see Table 1). In addition, recent trends from the four border states (California, Arizona, New Mexico, and Texas) and non-MSA border region counties are also examined.
Several regional characteristics are worth noting. First, the region is far from homogeneous. In terms of economic, cultural, and political features, San Diego, California, and Brownsville, Texas are no more appropriately placed in the same region than Oklahoma City, Oklahoma, and Paris, France. Second, the region's population has been growing faster than the nation as a whole for more than 50 years. Even if migration to the border were to stop suddenly, the region's population would continue to grow more rapidly than the national average for several decades [Peach and Williams, forthcoming]. Third, border region per capita income (with the important exception of San Diego) is low relative to the national average. In 1997, for example, the six poorest MSAs in the nation were located in the border region, and 14 Texas border counties ranked among the nation's poorest 100 counties [U.S. Department of Commerce 1999]. Fourth, when compared to national data, regional unemployment rates are high and labor force pa rticipation rates are low (see Table 1).
Expected NAFTA Impacts: National and Regional
When NAFTA was implemented, the United States had already been involved in a free trade agreement with Canada which was signed in 1988 and implemented in 1989. As a result, NAFTA would bring little change to U.S.-Canada trading rules. Mexico had been unilaterally opening its economy in the years immediately prior to NAFTA, but the prospect of further opening U.S. borders for trade with an obviously lower-wage and less developed nation caused concern to many in the United States. [Weintraub 1993, 12]. A 1992 survey of attitudes toward NAFTA indicated that Americans generally expected a positive overall socioeconomic impact on the United States. The same survey revealed that Americans expected an overall loss of U.S. jobs [Nicholson et al. 1994]. These apparently conflicting views reflect the uncertainty surrounding NAFTA's expected outcomes.
While there was uncertainty regarding NAFTA's expected impacts, there was also consensus on certain points. …