Evaluating NAFTA – an impulse function approach

Abstract

Models developed under ex ante approach are mostly based on some utility function. The assumptions underlying these models are hard to meet. These models also have stringent data requirements. The models developed under ex post approach, on the other hand, are simple to use, avoid the need of using a utility function and require less information. However, these models only offer two points in time: one before a trade reform and the other after the reform. As such, they do not provide a region of acceptance or rejection, which enables us to test the models empirically. Our study overcomes those problems, develops an intervention model, estimates the impulse response function and estimates the short-term and the long-run impacts of NAFTA (the North American Free Trade Agreement) on the variable representing Canada's welfare. Our results show that both the short-term and the long-term impacts are negative. Therefore, we conclude that Canada's welfare has, in fact, deteriorated, or at least not improved due to NAFTA.