Different approaches of modelling reaction lags: how do Chilean manufacturing exports react to movements of the real exchange rate?

Abstract

This study examines the relationship between export supply and the real exchange rate using annual Chilean data for the period 1960-1996. The hypotheses to be tested are first, that the real exchange rate does matter for the supply of exports - contrary to studies relying on quarterly data - and second, that the impact of a real depreciation only ceases to be positive and significant after about two-three years. Four different distributed lag models were considered as potentially adequate and useful to depict the impact of the real exchange rate over time. Even though all four models assumed different underlying lag structures, they all point to the importance of maintaining a competitive real exchange rate over time. The transfer function model is particularly well suited in shaping any lag structure in that it is not presumptive in form.