Files
Abstract
In a series of papers, Anderson and Neary have developed a trade restrictiveness index (TRI) that is firmly grounded in economic theory and that can be implemented empirically. In order to calculate TRIs in practice, one needs a model. This paper examines the robustness of TRI calculations to alternative model structures and it shows that the TRI is not very sensitive to changes in elasticities of substitution between factors of production, but it can be quite sensitive to alternative model structures (e.g. specific or mobile factor models). Also, the paper points out that in assessing trade restrictiveness over time, researchers need to be aware that changes in economic structure (e.g. factor accumulation and technological change) will alter the calculated values of the TRIs for unchanged trade policy. Therefore, researchers need to adjust the calculated values for TRIs for changes in economic structure if one wants a measure of changes in trade policy only.