August 2012

IZA DP No. 6791: Spouses' Retirement and Hours Outcomes: Evidence from Twofold Regression Discontinuity with Differences-in-Differences

Earlier studies conclude that spouses' retirement strategies are not independent from each other and that policies affecting individuals in a couple are also likely to affect the economic behaviour of their partner. In this study, we exploit retirement age legislation in France as well as a retirement policy change to identify the effect of own and spousal retirement on spouses' hours. To this end, we use a Fuzzy Regression Discontinuity approach combined with Differences in Differences, for both spouses. The data for the analysis are drawn from French Labour Surveys pooled over thirteen years. The sample for the analysis includes over 85,000 dual-earner couples with spouses aged 50 to 70. We find evidence of large and significant jumps in the own retirement probability at the legal early retirement age for both men and women in a couple. We also conclude that the 1993 reform reduced significantly the probability of retirement at the early retirement age for married men while the effect was not significant for married women. Husbands' retirement probability increases significantly when the wife reaches early retirement age while her retirement probability is not responsive to his early retirement age. We conclude that hours fall significantly upon own and partner's retirement for both spouses. On average, her hours fall by 2.7 per cent when he retires while his hours fall by 5 per cent when she retires, implying an average reduction of one hour per week for women and two hours for men if their spouse retires.