How robust are estimates of equilibrium real exchange rates

Sinopse

Increased attention is being paid to assessments of the actual values of countries' real exchange rates relative to their "equilibrium" values as suggested by "fundamental" determining factors. This paper assesses the robustness of alternative approaches and models commonly used to derive equilibrium real exchange rate estimates. Using China's currency to illustrate this analysis, the variance in estimates raises serious questions regarding how robust the results are. The basic conclusion from the tests used here is that, at least for China, small changes in model specifications, explanatory variable definitions, and time periods used in estimation can lead to very substantial differences in equilibrium real exchange rate estimates. Thus, such estimates should be treated with great caution.