Evaluating the effects of working hours on employment and wages

Abstract

The paper investigates whether a decrease in standard working time (the stipulated weekly working time) might produce work-sharing, i.e. a redistribution of a given amount of work over a larger number of employees. To this end, we adopt a structural vector error correction model thought to be representative of the main effects that working hours have on output, employment and wages in Germany and the US. Impulse response analysis suggests that permanent negative shocks to working hours lead to: (i) a decrease in the real GDP, (ii) a significant fall in real wages, and (iii) a persistent decrease in employment. More interestingly, results obtained from error-variance decomposition show that employment movements are not driven by working hours and that weekly hour shocks account for a minimal portion of employment and output dynamics. The main policy result emerging from the empirical analysis suggests that the implementation of work-sharing programs has not clear beneficial effects on the labour market performances.