Persistence in law of one price deviations: Evidence from micro-data

Using an extensive micro-price panel, we find a positive cross-sectional relationship between LOP persistence and the distribution margin, which we measure using sectoral U.S. data, as suggested by the classical dichotomy. The median level of persistence (across goods) is low, and there is no evidence of a border effect: the half-life of a deviation is about 19 months across OECD cities and just 1 month lower across cities in the U.S. Aggregating our micro-data using a variety of weighting methods shows PPP persistence to be in the range of 1-2 years, over the 1990-2005 period. These results challenge three widely held views: (i) the classical dichotomy is irrelevant; (ii) high persistence is a robust feature of aggregate real exchange rates; and (iii) border crossings necessarily generate greater real exchange rate persistence.