Questions about whether and how political leaders influence outcomes are fundamental to the study of politics. If politicians are responsive to constituent opinions, one would expect to observe similar outcomes across different leaders. In reality, though, public policy outcomes tend to vary systematically depending on who serves in elected office. This dissertation examines representation in American cities. New data reveal that American mayors, like politicians at higher levels of government, tend not to be highly representative of their constituents. For instance, they are more likely to be white males with prior political experience and white-collar careers. Business owners and executives are especially well represented. By providing compelling evidence that mayors can and do shape public policy, this research contributes to our understanding of the nature and importance of local politics. In addition, the study also begins to explore the causes and consequences of the overrepresentation of business owners and executives. The findings suggest that the nonpartisan election system used in a large share of local mayoral elections may create an interesting interaction with the preferences of voters that affects the electoral prospects of underrepresented groups. Moreover, this research implies that more equitable representation may result in policy changes that could benefit less affluent citizens.
