Asymmetry and (in-)stability of Okun’s coefficients in nine European countries

Abstract

We investigate Okun’s law in nine European countries from 1981 to 2021. We compare estimates by applying both the classic linear model and a robust M-estimation within a rolling regression framework to explore the (in-)stability of the coefficients over time. We do not observe substantial differences in the goodness-of-fit of the two approaches in terms of the in-sample mean absolute error. We confirm Okun’s law and its instability over time in the investigated countries, except for Norway, where the relationship between changes in the unemployment rate and GDP growth is relatively stable. We note some asymmetries in the trends across countries, reflecting differences in labour markets and productive systems. We also find some asymmetries and (temporary) shocks in the magnitude of the coefficient when including the most recent major economic slowdowns in the analysis, namely the Great Recession and the COVID-19 pandemic. For economists and policymakers, our findings provide insight and improve knowledge regarding the response patterns of unemployment to a country’s productivity.