Sociologists of consumption focus primarily on understanding the determinants and consequences of social variation in tastes. Patterns of variation in actual purchases, however, remain relatively unexplored. In this paper, the author proposes a framework for studying household expenditures. Drawing on the work of Mary Douglas, the author proposes a classification of goods and services into four spending bundles that are defined in relation to their contribution to social reproduction strategies. The author first tests the construct validity of the classification using household spending data from the Consumer Expenditure Survey. The author then examines the relationship between cultural capital and patterns of actual spending. In contrast to expectations derived from the sociology of consumption, the author finds that cultural capital is only weakly associated with how households actually spend their money. Overall, this suggests that sociologists need to incorporate issues of spending into the predominantly cultural focus of much consumption research.