Peer-Reviewed Articles

2015. Who Wants an Independent Court? Political Competition and Supreme Court Autonomy in the Argentine Provinces, 1984-2008.  The Journal of Politics, 77 (1) (with Marcelo Leiras and Agustina Giraudy).

Political competition should protect judicial autonomy. A host of studies produce evidence that is consistent with this expectation. The influence of political competition operates through two distinct mechanisms: fragmentation and turnover. Most empirical studies treat them as mutually reinforcing. We explain why each of these effects should be most clear when the other one is inactive: when power is concentrated only the expectation of turnover may protect judicial autonomy; when turnover seems unlikely only fragmentation should prevent interferences on the judiciary. We test these hypotheses using an original data set comprising all justices that served in the 24 provincial supreme courts in Argentina between 1984 and 2008. Results of a survival model with competing risks support our argument. The effect of fragmentation is discernible when turnover seems unlikely. The expectation of turnover restrains incumbents particularly when power is concentrated.

Other Publications

2015. Graphical Presentation of Regression Discontinuity Results (with Natália S. Bueno). The Political Methodologist, blog and print newsletter.

Working Papers

"When the Church Votes Left: How Progressive Religion Hurts Gender Equality"

How do progressive religious leaders shape the objectives and electoral fortunes of the left? While existing research focuses on the effect of religious organizations’ morally conservative policy preferences, the effect of religious ideas about economic redistribution is often overlooked. I argue that church leaders who advance doctrinal interpretations that favor progressive economic policies will mobilize their adherents in support of the left, as long as doing so does not advance policies that contradict the church’s moral agenda. When these leaders advance conservative moral policies, gaining their support requires the left party to moderate its position on this policy dimension. I test this argument using original archival data from the Catholic Church and drawing on a natural experiment in Brazil after Pope John Paul II’s appointment in 1978. Leveraging plausibly as-if random variation in bishop vacancies, as well as Pope John Paul II’s systematic appointment of conservative bishops to posts where progressives previously held court, I study the effect of progressive bishops on the electoral success of the left-wing Workers’ Party. I find that the left’s electoral prospects suffered significantly in places where progressive bishops were replaced between 1978 and Brazil’s first democratic elections. The party’s stronger performance in progressive dioceses can be partly explained by its access to religious networks, which allowed it to build organizational structures that delivered an electoral advantage. However, the electoral benefits of religious support came at the cost of marginalizing the political demands of the party’s feminist faction, thereby decreasing women’s access to party tickets and the implementation of policies that favor gender equality.

This paper is part of a broader book project.  A description of the project and a chapter outline can be found here.

"​Religious Minorities and the Costs of Secularization: Theory and Evidence from Brazil"

​Despite conventional wisdom that increases in religious competition lead to secularization of church-state relations, religious minorities face tradeoffs when deciding whether to advocate for secularism. While dismantling state favoritism can offer ascendant minorities the chance to compete with majority religions, it can also benefit competing minority denominations, diluting the benefits of religious liberties for confessional groups seeking to grow their base of adherents. I test this argument through a study of religious instruction in local public schools—a policy area often cited as crucial for majority religious denominations—in the context of the massive recent increase in the political power of Evangelical Protestants in Latin America. I leverage a regression discontinuity design in Brazil that compares municipalities where a political coalition supported by the Universal Church of the Kingdom of God (UCKG)—a minority Brazilian Pentecostal denomination—barely won or lost the municipal elections. I first show that a UCKG victory reduces by 10 percentage points the proportion of public schools in which Catholic religion classes are taught. This result suggests that religious minorities often prefer to restrict denomination-specific policies when they have the political clout to do so. Yet, I find robust evidence that, although the incumbency of the UCKG coalition reduces Catholic adherence, it is other Pentecostal churches—not the UCKG—that attract new members. More broadly, my argument suggests reasons why religious favoritism may sometimes endure across countries with different religious traditions and levels of religious competition.

"Is Paying Taxes Habit Forming? Theory and Evidence from Uruguay" with Thad Dunning, Felipe Monestier, Rafael Piñeiro and Fernando Rosenblatt. (Under Review​)

Citizens can develop persistent, self-reinforcing habits of interaction with tax bureaucracies and other organs of the state. Yet, policy interventions can also foster or disrupt habits, often with unanticipated consequences. We study a policy in Montevideo, Uruguay that randomly assigns tax holidays—that is, year-long interruptions of payments—to punctual taxpayers. The program is designed both to reward and induce tax compliance, a critical facet of state capacity. We find that far from fostering compliance, interrupting the habit of paying taxes results in a substantial and prolonged reduction in payments after the holiday’s conclusion. Consistent with our theory, failure to pay has self-sustaining effects. Yet, for taxpayers with a strong reserve of compliance habit, negative effects eventually decay. We use field and survey experiments to isolate the habit mechanism from alternative explanations. Our findings have implications for understanding virtuous or vicious cycles in civic participation.

"Design-Based Analysis of Regression-Discontinuity Designs: Evidence from Experimental Benchmarks" (with Natália S. Bueno and Thad Dunning). 

"Paying Not to Vote: Pocketbook Considerations and the Shape of the Electorate" (with Germán Feierherd and Gerson Julcarimna).

Work in Progress

The Effects of Homeownership on Political Participation  (with Craig McIntosh, Felipe Monestier, Rafael Piñeiro and Fernando Rosenblatt).
Registered Pre-Analysis Plan