Who votes for mayor? Evidence from midsized American cities

ABSTRACT

Who, or what, governs the mosaic of the almost 39,000 general purpose local governments across the United States? While the determinants of voter turnout in both national and sub-national elections have been well studied, there is a dearth of empirical literature examining voter turnout at the municipal level. Utilizing an original dataset of 356 midsized US cities drawn from the Midwest, South, and Northeast regions, this paper ponders the drivers of mayoral turnout, and asks if electoral timing, competitiveness, or characteristics of the electorate best predict turnout. Sequenced hierarchical linear models and OLS regressions are employed to control for sub-national effects, and model results indicate that election timing greatly dictate who governs midsized American cities. While local competitiveness and characteristics of the electorate also matter, their explanatory value is greatly over-shadowed by timing.

KEYWORDS:

• Elected mayors
• local elections
• voter turnout
• local government
• local politics

Acknowledgments

I want to extend my sincere thanks and gratitude to two faculty members who assisted me while I was an undergraduate political science student at Illinois Wesleyan University: Professor Kathleen Montgomery and Professor (and Mayor of Bloomington, Illinois) Tari Renner. This paper would not be possibile without them.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1. One example is the placement of elections on particular days of the week. Per US law, federal elections must be set on, ‘the first Tuesday next after the first Monday in the month of November.’ Meaning the earliest possible date for an election would be November 2nd (if November 1st is a Monday), and elections can only transpire on week days. Local elections greatly mirror these rules and traditionally hold elections on either Tuesdays or other week days. Yet, empirical research does show that elections held on weekends garner higher voter turnout (Mattila 2003).

2. The VAP methodology for measuring voter turnout is in contrast to the eligible voter population method (VEP), which is argued for most notably by McDonald and Popkin (2001) and Holbrook and Heidbreder (2010). However, the real number of eligible voters in each municipality was not uniformly reported. Hence, the VAP methodology of turnout is used.

3. In terms of electoral timing and days of the week, over 95% of the elections transpired on weekdays.

Additional information

Notes on contributors

Austin M. Aldag

Austin M. Aldag is a Ph.D. student in the Department of City and Regional Planning at Cornell University. His research agenda includes, but is not limited to, local government, federalism, service delivery and intergovernmental relations.