Price level wavelet co-movement across regions in Turkey

This paper attempts to examine whether there is price level convergence across regions in an emerging market, namely Turkey, by using first and second generation unit root tests and wavelet comovement analysis. The empirical investigations on price levels within a country would improve our understanding of price behavior and enable us to evaluate the findings of the previous studies conducted on both international and intra-national data. Moreover, the well-known slow price level convergence found in empirical studies using developed country data may be due to the market structure specific to economies at that stage of development which probably have different price dynamics. Motivated by this gap in the existing literature, we employ unit root tests on disaggregated price data for Turkey, which represents a major emerging market economy. We use monthly price data obtained from the Turkish Statistical Institute on a panel of 54 consumer products from 26 regions in Turkey for the period from January 2005 to December 2011. First, we produce half life estimates to examine the speed of the convergence of price levels. In conformity with the previous studies using data on emerging markets we find price level convergence of a slower rate compared to developed economies. We speculate that differences in market structures between developed and developing countries may be the underlying reason. Then, we investigate distance as a possible explanatory factor that could account for the rate of price level convergence. Our findings suggest that distance does not seem to be an important determinant of the regional price level convergence in Turkey, which implies the price effects of distance cannot be fully captured by simple intercity measures. Moreover, wavelet comovement analysis shows that price levels show differences before and after the Great Recession.