Much research shows that politicians represent public preferences in public policy. Although we know that there is representation, we do not understand the nature of the relationship in different policy areas. We do not know whether and to what extent representation varies across domains. Even where we find representation, we do not know what policy makers actually represent. This article explicitly addresses these issues, focusing on a set of nine spending domains in the United States. At the heart of the article is a simple conjecture: representation varies across domains, and the pattern is symmetrical to the pattern of public responsiveness to budgetary policy itself. Analysis of the relationships between opinion and policy over time in the different spending domains supports the conjecture. The patterns fit quite nicely with what we know about the influence of different issues on voting behavior in American national elections. Based on this analysis, then, it appears that politicians’ responsiveness to public preferences reflects the public importance of different policy domains.