구매력평가 가설은 성립하는가?

Does the Purchasing Power Parity Hypothesis Hold? -The Panel Data Case of Big Mac Hamburger Prices-

The appropriate aspect of McDonald s Big Mac Hamburger for international price comparison is that the Big Mac is identical goods across countries. This study presents an empirical analysis of nonstationary panel data constructed by Big Mac prices of 27 countries for 16 years. To illustrate whether the purchasing power parity hypothesis holds or not, recently developed powerful methods such as panel unit root test, panel cointegration test and panel estimation and inference techniques are employed. Main findings in dynamic OLS framework indicate followings. First, We cannot accept the strong-form purchasing power parity hypothesis directly, on account of the null of (1,-1) cointegrating vector between the nominal exchange rate and Big Mac price ratio being rejected. Second, Harrod-Balassa-Samuelson effect exists. A country s nominal exchange rate is systematically undervalued more with respect to less per capita GDP. Third, in case the Harrod-Balassa-Samuelson effect is included, we cannot reject null of (1,-1) cointegrating vector. In conclusion, the purchasing power parity hypothesis could conditionally hold in the panel data case of Big Mac Prices.