Here, we expand on these studies by considering also retirement of the wife, thus distinguishing households in which the wife is a "housewife" from 'dual-earners'. We use a regression discontinuity approach to estimate the effect of each partner's retirement on household consumption. We use for the analysis data drawn from the French Consumer Budget Survey 2001 that collected two-week expenditure diaries. We find a significant and sizable drop in food and clothes expenditure upon retirement of the male partner. However, the drop in food expenditure is not robust to specification checks and it becomes statistically insignificant when dropping from the sample couples in which the wife is a housewife.
- dp7709