Estimating Models of Earnings Dynamics with the European Community Household Panel Data

Abstract

The aim of this paper is to use panel data on male workers to separate the permanent from the transitory component of earnings inequality for a number of European countries. Several authors have noted the need for long panel data sets when conducting such exercises. Unfortunately the data best suited to international comparisons across European countries (the European Community Household Panel) is limited to a maximum of eight years per individual. Therefore, before commencing with a detailed analysis of the ECHP we conduct a series of Monte Carlo evaluations of our proposed estimator for ECHP type data structures. Our analysis suggests that for moderate degrees of persistence in transitory shocks, eight years is sufficient to get precise estimates of the components of inequality. However, for stronger persistence, eight years is clearly problematic and longer panels are required. With this in mind, we estimate the permanent and transitory components of inequality across European countries. We find that the nature of inequality varies substantially across countries, with differences in permanent inequality accounting for the majority of the observed cross-country differences in aggregate inequality. We also briefly assess the effects of unemployment and education on the components of inequality.