Individual Submission Summary
Share...

Direct link:

The Electoral Power of the Teacher

Saturday, November 7, 1:45 to 3:15pm, Property: Boston Marriott Copley Place, Floor: 3rd Floor, Room: Dartmouth

Abstract

Government employees participate in American elections at rates far higher than can be explained by their demographic factors alone. Yet both the sources of higher turnout among public employees and its implications for democracy and public policy have long been contested. We bring new evidence to bear on both debates by combining nearly two decades of administrative data on school employees in Ohio with validated voter turnout records.

Leveraging variation in the timing of teachers’ job transitions, we first estimate the causal effect of government employment on voting. We find that teachers become significantly more likely to vote in low-turnout school board elections after obtaining jobs in traditional public schools—but, crucially, not when their careers begin in charter schools, which are publicly funded but not governed by elected school boards. Becoming a public school teacher also increases participation in even-year November elections, when local school-related races do not appear on the ballot, indicating that self-interest alone cannot fully explain these turnout changes. We further find that these mobilization effects extend beyond public employees themselves, increasing turnout among other voters living in the same household who do not work in public schools.

Next, we use data from a second state, Iowa, and take advantage of a change in state law that consolidated the timing of school board and city elections. As in Ohio, we show that this change dramatically increased turnout among teachers in what were formally city-only elections. The increase was particularly pronounced among teachers who worked, lived, and voted in the same jurisdictions, and we again find substantial increases among other voters living in the same household as educators.

Even after accounting for these spillovers, voters with a direct employment stake in school board elections constitute a surprisingly small share of the electorate even in very low-turnout elections, suggesting that election timing alone cannot explain the overwhelming success of teacher-union–endorsed candidates.

Authors