PUBLICATIONS AND ACCEPTED PAPERS

Strategic Allocation of Irrevocable and Durable Benefits ​(forthcoming, American Journal of Political Science)
​Article

​Do Reelection Incentives Improve Policy Implementation? Accountability vs. Political Targeting ​(forthcoming, Quarterly Journal of Political Science)
Article

Cash transfers, clientelism, and political enfranchisement: Evidence from Brazil (Journal of Public Economics, 2019)
Article​

WORKING PAPERS

Can Descriptive Representation Help The Right Win Votes From The Poor? Evidence From Brazil
​(with Zuheir Desai)
(Revision requested, American Journal of Political Science)

The electoral success of the Right in poor nations is typically attributed to non-policy appeals such as clientelism. Candidate profiles are usually ignored, because if voters value class-based descriptive representation, it should be the Left that uses it. In this article we develop and test a novel theory of policy choice and candidate selection that defies this conventional wisdom: it is the Right that capitalizes on descriptive representation in high poverty areas. The Right is only competitive in poor regions when it matches the Left’s pro-poor policies. To credibly shift its position, it nominates candidates that are descriptively closer to the poor. Using a regression discontinuity design in Brazilian municipal elections, we show that Right-wing mayors spend less on pro-poor sectors than Left-wing mayors only in low-poverty municipalities. In high-poverty municipalities, not only does the Right match the Left’s policies, it also does so while nominating less-educated candidates. 

​Download

A Variance Decomposition Approach for Bounding Neighborhood Effects
​(with Scott Abramson)
​(Under Review)

Political scientists have pointed to neighborhood contexts as an important factor explaining individual voter behavior. This paper presents a variance-decomposition method for placing bounds on the impact of neighborhood effects on political attitudes. The method should be used to complement rather than to replace other empirical approaches used by the discipline in the study of neighborhood effects, as it provides less restrictive assumptions than most observational studies, and a broader perspective on the aggregate influence of neighborhoods than the typical experimental design. We also present two applications using data from the universe of voters in North Carolina. We estimate that neighborhood effects explain at most 1.9% of the total variation in individual turnout decisions, and 4% of the observed variation in party registration, after accounting for the race of households. In both applications, household-specific traits are a much more relevant factor determining those individual outcomes.

Upon Request

Sleeping with the Enemy: Effective Representation under Dynamic Electoral Competition
​(with Sergio Montero and Gabriel López-Moctezuma)
(Revision requested, American Journal of Political Science)

Electoral coalitions between ideologically incompatible parties - among other unconventional electoral strategies - may appear to threaten effective representation, signaling a breakdown of programmatic politics. However, this perspective overlooks parties' and voters' dynamic considerations. We propose and estimate a model of dynamic electoral competition in which a short-term ideology compromise, via an electoral coalition, offers opposition parties (and voters) the opportunity to remove an entrenched incumbent party from office, thus leveling the playing field in the future. This tradeoff provides a previously unrecognized rationale for electoral coalition formation. We take our model to data from Mexican municipal elections between 1995-2016 and show that coalitions between parties on opposite ends of the ideology spectrum have in fact served as an instrument of democratic consolidation.

Coming Soon

Larger Legislatures and the Cost of Political Brokerage: Evidence from Brazil
​​(Under Review)

This article shows that larger legislatures reduce the electoral power of incumbent parties in the executive. The electoral effects of legislature size have been largely overlooked by a literature that emphasizes its impact on policy outcomes. I estimate the effects of municipal council size on the results of mayoral, gubernatorial and presidential elections in Brazil. The regression discontinuity design exploits variation from a law that set non-linear council size caps after 2012. In a nutshell, every additional seat triggers a 10% vote loss for the candidates backed by the mayor's party. I also show evidence that these losses are a consequence of a breakdown in the political brokerage relationships that often characterize developing democracies: in Brazil, mayors exchange patronage for the councilors’ electoral support. Larger councils raise this transaction cost for the executive, more so when council and mayor have unaligned electoral incentives at the state/national levels.
​
​Coming soon

WORK IN PROGRESS

Pricing Partisanship: Evidence From North Carolina's Housing Markets (with Scott Abramson)

Building Weak Parties with Ambitious Politicians (with Sergio Montero and Gabriel López-Moctezuma)

Bureaucratic Performance and Party Recruitment

Political Targeting in Rural Insurance

​Do Programmatic Policies Induce Political Engagement?