Three essays on convergence and persistence of prices in the European Union.

摘要

This dissertation consists of three essays on the persistence and convergence of prices in the European Union (EU). The first essay studies price differentials within the Eurozone by comparing prices for specific goods between cities, and estimate their persistence. The second essay tests the effect of the single market and currency on price convergence in the EU. The third essay tests the effect of trade on price convergence.;In my first essay, using the Economist Intelligence Unit data I find that large price differentials exist across Europe. I estimate the persistence of deviations from the Law of One Price (LOP) by employing linear and nonlinear panel unit root tests. Based on the linear test, the estimated half-lives are 13 months for traded goods and 15 months for non-traded goods. Splitting the data series for individual goods into stationary and non-stationary components reduces the half-lives to 3 months for traded and 5 months for non-traded goods. The results of the nonlinear test suggest that the speed of reversion to parity is constant and independent of the size of the deviations, which implies prices based on good-specific data adjust in a linear fashion.;International treaties in the EU were expected to promote integration of consumer markets. In my second essay, I analyze the convergence of prices. Using high-income non-euro European countries as a control group, I test the effect of the euro and Single European Act (SEA) on price convergence in the EU. Using a difference-in-difference estimation strategy I find that the SEA, not the common currency, reduced price differentials and promoted price convergence. My third essay tests the effect of trade on price differentials. The fixed effects estimate shows strong correlation between price differentials and bilateral trade, a 1% increase in trade reduces price differences by 9%. This finding is driven by the Eastern and Central EU. Since trade is endogenous an instrument is required to estimate a causal relationship. Instrumental-variables strategy shows the effect of trade on price differentials for the Eastern and Central EU is even larger, at 13%, yet insignificant for the remaining EU members.