What drives governments with similar revenues to publicly provide very different amounts of goods for which private substitutes are available? Key examples are education and health care. In this paper, I focus on pre-primary education in Brazil. I show that areas with higher median income and more income inequality allocate less of an exogenous revenue increase to education and more to public infrastructure, which does not have private substitutes. I exploit a 1998 education finance reform to ensure that my results are credibly causal and do not merely reflect omitted variables that affect both tastes and income. This result can be explained by two separate theories. The first, "collective choice," hypothesizes that richer people are more likely to consume the private version of a good, especially as their incomes diverge from those of others in their area. Because they do not use the publicly-provided good, they do not support government spending on it. The second, "political power," hypothesizes that the rich are able to exercise control over public policy that is roughly proportionate to their share of income, as opposed to their share of votes. Thus, the poor may have little ability to make government publicly supply the goods they most demand. I attempt to apportion the estimated effect between these two theories by examining whether Brazilian municipalities with Participatory Budgeting, a form of governance designed to ensure that the poor have a strong role in policy making, allocate exogenous revenue differently than other municipalities. My results suggest that both the collective choice and political power theories are operative.