Working time reductions and monopsony power

This paper studies the consequences of working time reductions when labor markets may be monopsonistic. A toy model shows that the marginal utility of a small working time reduction is negative when workers control schedules but positive when firms set hours. However, the policy increases wage rates in perfect competition but decreases monopsonistic wage rates. I test these predictions empirically by evaluating the first-ever working time reduction in Belgium: the maximum 9h workday in 1910’s coal mines. I find that the policy had sizable negative effects on profits, employment and earnings. To assess welfare, I build a directed search model with matching frictions where firms have heterogeneous productivity and post wages and hours. Utilitarian welfare is expressed in terms of a sufficient statistic whose application to the 1910 reform suggests that the value of leisure was particularly large.