Trade, Growth, and Welfare Linkages in North America: An Empirical Analysis MUNISAMY GOPINATH, P. LYNN KENNEDY, and TERRY L. ROE ABSTRACT Changes in welfare are computed as the combined effects of technological progress (TFF), and terms of trade for the NAFTA signatories. Nonparametric methods are used to derive indices of output, TFP, terms of trade, and welfare from aggregate time-series data. Results indicate the importance of both TFF growth and terms of trade to welfare in all three countries. An analysis of TFP growth suggest that increased openness of the Mexican economy augments this rate effect through, for example, the increased imports of more modem intermediate factors while, for the more open economies, the NAFTA trade effects are small. Hence, other Latin American countries joining the NAFTA agreement may experience gains in TFP growth similar to Mexico. INTRODUCTION The reduction of trade barriers brought about by the North American Free Trade Agree- ment (NAFTA) is expected to have diverse effects on its signatories (Kehoe and Kehoe 1993; Cox 1993). These studies suggest that NAFTA will generate significant benefits for Mexico with relatively small aggregate effects on the United States and almost no impact on Canada, while others (Burfisher et al. 1992) find welfare gains from reform to be rela- tively small. However, these contributions are either of the applied general equilibrium- calibration methodology or partial equilibrium and, hence, they tend not to focus on the time series or sources of growth effects in drawing inferences. This paper contributes to the latter. The envelope properties of the Gross National Product (GNP) function are exploited and an approach of Diewert and Morrison (1986) is used to measure the impacts of techno- logical change and terms of trade on welfare. This general equilibrium approach permits the nonparametric use of time-series data to compute indexes of real output and input, tech- nological change, terms of trade, and welfare effects from aggregate price and quantity data for each of the three signatories. A distinction is made between level variables that affect the rate of growth in GNP and the rate effect-that is, total factor productivity (TFP). Then, drawing upon fairly recent contributions to growth theory and the empirical approach suggested by Backus et al. (1992), a parametric analysis of the growth in total factor productivity is undertaken to provide insights into whether increased trade among the three signatories is likely to be growth-augmenting. The case of NAFTA is particularly interesting because the agreement is expected to have significant effects on the level of GNP as resources are reallocated to more profitable Munisamy Gopinath and Terry Roe l Department of Applied Economics, University of Minnesota, St. Paul, MN 55108; and Lynn Kennedy l Department of Agricultural Economics and Agribusiness, Louisiana State University Agricultural Center, Baton Rouge, LA 70803. North American Journal of Economics & Finance 6(2): 189-201 Copyright 0 1995 by JAI Press Inc. ISSN 1062-9408 All rights of reproduction in any form resewed