Estimating Switching Costs from Aggregated Data

This paper proposes a method to identify and quantify switching costs using firm-level data. The method is based on a simple thought experiment that compares the behavior of already contracted consumers to the behavior of new consumers, as the latter are not subject to switching costs. In two panel regressions on new and quitting consumers, we look at the differential response to firm-specific price changes, and identify switching costs from the difference between the two. We apply our method to the Hungarian personal loan market. We cannot present the results before the publication of the corresponding GVH inquiry (see main text), expected in April 2008.