Does the interaction between shocks and institutions solve the OECD unemployment puzzle ? A theoretical and empirical appraisal

This paper adds to the already vast economic literature on the respective role of shocks and institutions in unemployment dynamics. Conclusions of existing studies widely diverge. Such divergences are mainly due to a weak theoretical framework underlying the models which have been estimated. We have tried to overcome such weaknesses by relying on a structural model based on a Phillips curve in order to obtain a reduced form unemployment equation. Once estimated, this reduced form accounts for the importance of macroeconomic shocks in explaining changes in unemployment. The introduction of institutional variables and the estimation of three potential effects on unemployment (level, persistence and sensitivity to shocks) lead to results, which are consistent with theoretical predictions. Nevertheless, the role of institutions in explaining changes in unemployment is limited. We use a panel data approach by pooling country data in order to disentangle fixed country effects.