Abstract: Democratic theorists have long argued that states can create more resilient democracies through education. Educational investments are thought to produce more economic equality and instill in citizens greater capacity and responsibility to participate in politics. Using a geographic regression discontinuity design and township level data from Antebellum New York State, we examine whether state funding for common schools led to positive economic and political outcomes. Our estimates support the view that a democratic culture emerged not only because of initial favorable endowments, but also because of subsequent government decisions to fund education. New York townships that received more school funding later had higher median earnings, lower earnings inequality, and higher levels of voter turnout. Our findings support the view that maintaining democracy requires active investments by the state, something that has important implications for other places and other times—including today.

Abstract: Many large survey courses rely on multiple professors or teaching assistants to judge student responses to open-ended questions. While the adoption of, and training on, a grading rubric can help alleviate some of the most extreme versions of bias, there remains the opportunity for students with similar levels of conceptual understanding to receive widely varying assessments. We detail how this can occur, and argue that it is an example of differential item functioning (or interpersonal incomparability), where graders interpret the same possible grading range differently. Using both actual assessment data from a large survey course in Comparative Politics and simulation methods, we show that the bias can be corrected for by a small number of “bridging” observations across graders. We conclude by offering best practices for fair assessment in large survey courses.

Abstract: Countries that use proportional representation (PR) tend to have higher levels of redistribution. However, persuasive research on electoral system choice has demonstrated that countries only adopt PR under particular circumstances, such as the presence of a strong left-wing opposition or a need for coordination between opposing economic actors. We therefore ask if the strong relationship between the use of PR and redistribution is due to PR electoral rules or to these background factors. Taking advantage of an electoral reform to early 20th-century Norwegian local elections, we find that municipalities that were mandated to use PR increased tax rates and resources spent on the poor but also redistributed less in the first place. We further show that the reform did not increase left-wing party seat shares but did increase political mobilization. This evidence is consistent with moderate parties increasing redistribution in order to preempt left-wing party gains.

Working Papers

Abstract: Could the type of political institution used at the local level impact the balance of power in national politics? If so, how? This paper argues that certain political institutions, like local partisan elections, empower local party branches, which in turn change mobilization patterns in national elections. Furthermore, the type of party that is most likely to benefit from this dynamic are parties that use local branches and members as a central component of their electoral strategy. To test this intuition, I take advantage of a quasi-experimental local governance arrangement in early 20th century Sweden that made local institutions dependent on a population threshold. I find that municipalities that used elections and representatives for local governance, as opposed to a more traditional community meeting, had higher vote shares for the Social Democrats in national elections. Moreover, this is a pattern that is especially pronounced in areas with active party branches, while it disappears in areas without local activists. The result demonstrates that local-level political institutions can significantly impact national-level politics with consequences for institutional reform outcomes and citizen welfare. 

Abstract: Do local-level political institutions matter for national-level electoral outcomes? We explore the effects of a 2011 electoral reform in Poland that replaced proportional representation with single-member districts in municipal council elections in some but not all municipalities. We contend that the reform benefited specific political parties, which affected local party committees and gave them a relative advantage in subsequent national elections. Difference-in-differences analysis and placebo tests demonstrate that increases in the share of independent municipal councilors have benefited the already well-established parties and solidified the national-level political landscape. These findings demonstrate that local institutions impact the linkages between voters and national parties, which ultimately affect democratic robustness.

Work in Progress

Abstract: In the wake of Shay’s Rebellion of 1786 and the Whiskey Rebellion of 1794 political elites and others in the United States expressed fears that instead of participating actively in the electoral process, disgruntled groups of people might instead attempt to air their grievances through violent unrest. One proposed response to this was to better educate the population, which helped lead to the creation of “common school” funds by a number of states. In a staggered difference in differences setting that allows for dynamic effects over time, we examine whether the creation of these common school funds might help explain an important political phenomenon observed by historians during this period: greatly increased engagement of citizens in the electoral process. The results we find are supportive of this interpretation. They also point to another important fact that those who propose new civic education reforms ought to consider. Given the inevitable time lag between beginning to learn and being able to vote, it takes several decades for an educational reform to affect a large fraction of the electorate.

Abstract: Conventional wisdom says that democracy and taxation are tightly linked. Foundational theories, such as the Meltzer-Richards model of redistribution, predict that expanding the pool of citizens that decide on taxation should increase the tax burden, especially on the rich. However, this dynamic has proven difficult to find empirically. In this paper, I fill this gap by leveraging the uneven extension of suffrage in historical Norwegian municipalities and a rich dataset on local taxation types and levels. I find that overall tax rates are unaffected by democracy, but that the types of taxes collected are impacted by regime type. The results have implications for our understanding about how regime type impacts economic policies.