In this paper I analyze the performance of the Michigan Model of Production and Trade in simulating the impact of trade liberalization under the North American Free Trade Agreement, reviewing the results of Brown, Deardorff, and Stern (1992). Because the NAFTA entered into force only part way through the phase-in of the U.S.-Canada FTA accord, I consider their joint impact on the pattern of relative trade flows. The methodology draws on the Fox (2000) analysis of the U.S.-Canada FTA. A substantial innovation in this paper is the reimplementation of the model using MPSGE/GAMS. Preliminary results suggest that the model performs best when simulating the impact on the already-substantial trade flows between U.S.-Canada and U.S.-Mexico. The expansion of certain sectors that had little pre-NAFTA trade highlights the difficulty of using a CES specification.