This paper uses microdata from the 1992 Statistics Canada Family Expenditure Survey to provide evidence that male and female incomes do not always exert identical influences on household expenditures. The novelty of the paper lies in its demonstration that, while incomes may be pooled for some categories of consumption (e.g. housing), the income pooling hypothesis must be rejected for others. We also go beyond simply rejecting the pooling hypothesis to ask how male versus female income is used. Our results stress the on-going importance of traditional gender roles. For example, we find that expenditures on child care increase only with women’s incomes_higher male income is not associated with higher expenditure on child care even when both spouses are full-time, full-year paid workers.