This paper aims to study the effect of the 35 hours week in France on productivity and employment.  France has  implemented  3  successive laws  to  promote  the 35  hours  week through financial incentives in order to improve work sharing.  None of the previous studies agree on the possible impact of such a policy. In this paper, we conduct a Granger Causality test in order to analyse the consequences of the decrease of working hours on productivity and then  employment.  The  results  show  no  causality from  the  number  of  working  hours  to employment. Our main explanation is that the law has never been fully applied and the period was not long enough to observe an important effect of the unemployment rate.  Besides, the level of wage compensation might not have been well calibrated. As a result of our study, we suggest that introducing such a policy, such as reducing working hours to increase employment, in countries where working hours are already short (40 or less) must be well examined before implementation and different characteristics of labour market must be taken into account.