Wirtschaft und Gesellschaft 42. Jahrgang (2016), Heft 4 Empirical studies using microdata often find no or a negative employment effect, whereas time-series models often find a positive relationship. Building on work by Kapteyn et al. (2004), a macroeconomic model is developed to estimate long-run cointegrating relationships. For a panel of 18 european countries, two enhancements to the literature are made. First, the average working hours of those full employed are used, limiting noise by part-time work. Second, cross-section dependencies are controlled for. The result of the ARDL model suggest a positive relationship between shorter working weeks and employment. 684