Testing Deviations From Purchasing Power Parity (PPP)

The purpose of this paper is to study analytically how the presence of transportation costs in a model of deviations from PPP affects the testing procedure of the PPP hypothesis The analysis shows that in the presence of transportation costs traditional regression analysis will tend to reject the PPP hypothesis even if goods markets are well arbitraged, because the values of the regression coefficients are affected systematically by considerations that are independent of the degree to which markets are arbitraged. Thus, the content of the ppp approach cannot be tested satisfactorily without considering the systematic effects of transportation costs and other costs of goods arbitrage.