Explaining variations in public employment

This article tests and evaluates existing and new models that seek to explain variations in postwar public employment in the countries belonging to the Organisation for Economic Co-operation and Development (OECD). The basic assumption is that the existing models are rather incomplete and should be extended by including institutional variables. The results on the basis of pooled time-series analysis (using Beck and Katz' "panel-corrected standard errors") seem to confirm this assumption. The empirical findings that are based on neo-institutional modeling do indeed outperform the older models. Yet, no model is able to account for changes in public employment in the 1990s.