Empirical test for purchasing power parity using a time-varying parameter model: Japan and Korea cases

Abstract

This study examines the validity of the Purchasing Power Parity (PPP) hypothesis for Japan and Korea by using a smooth time-varying cointegrating regression model. When we use the usual approaches, including unit root and cointegration tests, we fail to find the existence of the PPP for Japan and Korea. However, we find there is a time-varying cointegrating relationship between the logarithm of nominal exchange rates and the logarithm of the Producer Price Indices (PPI) ratio for Japan and Korea. This relationship does support the PPP theory. Moreover, we also find that the exact PPP theory holds for some periods in Japan and Korea.

Keywords:

• Purchasing Power Parity
• time-varying parameter
• cointegrating regression
• foreign exchange

JEL Classification:

• C22
• F31

Notes

¹ For more details on this method, see Park and Hahn (1999).

² Following the convention in existing literature, we consider PPI as price indices. Since PPP theory addresses the law of one price in the international market, it is mainly concerned with tradable goods, the price level of which can be better measured using the PPI.

³ Aslan et al. (2010) also use Zivot and Andrews (1992) procedure to test PPP for Turkey.