Decomposition into Tradables and Nontradables and the Purchasing Power Parity (PPP) Hypothesis of the Real Won-dollar Exchange Rate

Abstract

The purpose of this paper is to test the purchasing power parity (PPP) hypothesis using the won-dollar real exchange rate and analyze the effect of the decomposition into tradables and non-tradables on the change of the won-dollar real exchange rate. This paper decomposes the CPI-based real exchange rate into two parts according to Engel (1999); one is the relative price of traded goods between the countries, the other is a component that is a weighted difference of the relative price of nontraded-to traded-goods prices in each country. We construct this by comparing the component subsection weights in CPI. The empirical analysis of this paper consists of two parts as follows. First, we conducted a traditional time series analyses of the real exchange rate, tradable and non-tradable parts respectively, thereby testing the PPP hypothesis and other important hypotheses. Secondly, this paper conducted a Mean Squared Error (MSE) analysis to evaluate the relative contribution of tradable and non-tradable parts to the change of real exchange rate. From the time series analysis, it is not guaranteed that the PPP hyThe purpose of this paper is to test the purchasing power parity (PPP) hypothesis using the won-dollar real exchange rate and analyze the effect of the decomposition into tradables and non-tradables on the change of the won-dollar real exchange rate. This paper decomposes the CPI-based real exchange rate into two parts according to Engel (1999); one is the relative price of traded goods between the countries, the other is a component that is a weighted difference of the relative price of nontraded-to traded-goods prices in each country. We construct this by comparing the component subsection weights in CPI. The empirical analysis of this paper consists of two parts as follows. First, we conducted a traditional time series analyses of the real exchange rate, tradable and non-tradable parts respectively, thereby testing the PPP hypothesis and other important hypotheses. Secondly, this paper conducted a Mean Squared Error (MSE) analysis to evaluate the relative contribution of tradable and non-tradable parts to the change of real exchange rate. From the time series analysis, it is not guaranteed that the PPP hypothesis hold in the long run. The Balassa-Samuelson hypothesis is not either, since the sample size is too small to avoid the 'power problem.' In addition, the result of the Mean Squared Error analyses show that tradable goods are more important in explaining the won-dollar real exchange rate dynamics than that of the non-tradable goods. All in all, the results of this empirical analysis are in contrast with the explanation that if the long-run PPP hypothesis does not hold, it is mainly caused by the transaction cost and the non-tradable goods.