According to the life-cycle theory of consumption and saving, foreseeable retirement events should not reduce consumption. Whereas some consumption expenditures may fall when they are self-produced (given higher leisure after retirement), this argument applies especially to housing consumption which can hardly be substituted by home production. We test this hypothesis using micro data for Germany (GSOEP) and find that income reductions when entering retirement have a negative effect on housing expenditures for tenants. For some econometric specifications, this effect is significantly stronger than the one of income changes at other times. While this result suggests that the strict consumption-smoothing hypothesis is violated for the subgroup of nonhome owners, the effect is quantitatively small, which explains the ambiguity of previous findings.