Presented at the 2011 meeting of the American Political Science Association and the 2010 meeting of the Society for Political Methodology.
The effect of conditioning on an additional covariate on confounding bias depends, in part, on covariates that are unobserved. We characterize the conditions under which the interaction between a covariate that is available for conditioning and one that is not can affect bias. When the confounding effects of two covariates, one of which is observed, are countervailing (in opposite directions), conditioning on the observed covariate can increase bias. We demonstrate this possibility analytically, and then show that these conditions are not rare in actual data. We also consider whether balance tests or sensitivity analysis can be used to justify the inclusion of an additional covariate. Our results indicate that neither provide protection against overadjustment. pdf.
Presented at the 2013 meeting of the American Political Science Association and the 2014 meeting of the Midwest Political Science Association.
I study civilians' cooperation with an armed group in an irregular war. In the model, civilians differ in their valuation of siding with the armed group and make cooperation decisions without knowing others' motivations or cooperation choices. I find that a superior military force is not sufficient to bring high cooperation and that full cooperation can only be attained if military power is complemented by expectations of punishment for helping the enemy. The model challenges the idea that random violence aimed at punishing enemy cooperators is used when selectivity is difficult to implement, and it shows that indiscriminate reprisals induce lower levels of cooperation, even when enemy cooperators are less likely to be punished with selective methods. Finally, I find that communities that have a highly centralized process of decision making are expected to give their support to only one group of combatants and to be exposed to less violence. pdf.
Presented at the 2013 meeting of the Midwest Political Science Association and the 2013 meeting of the Southern Political Science Association.
Vote buying is widespread in developing democracies despite the secret ballot. What explains its resilience? I argue that brokers condition future payments on published electoral results to enforce these transactions and that this is effective at inducing voters' compliance only when the results of small voting groups are available. Using monitors' and citizens' reports of electoral manipulation and survey data from Colombia, I find a robust negative correlation between the size of the average polling station and various measures of vote buying. Evidence from a variety of identification strategies suggests that this relationship can be attributed to aggregate monitoring sustaining these transactions and not to the brokers' increased ability to identify compliers or other characteristics of places where polling stations are small. pdf. supplemental material.
Presented at the 2012 meeting of the Midwest Political Science Association.
How do politicians buy votes in secret ballot elections? I present a model of vote buying in which a broker sustains bribed voters' compliance by conditioning future bribes on whether her candidate's votes reach an optimally-set threshold. Unlike previous explanations of compliance, the threshold mechanism does not require brokers to observe individual voters' political preferences or even vote totals of the bribed voters. I show that when there is uncertainty about voters' preferences, compliance can be sustained as long as electoral results of small groups are available. If preferences are observed however, vote buying is not deterred by higher aggregation of electoral results. I also find that vote buying is facilitated when voters care about the welfare of other voters. Using survey data from Nigeria, I provide evidence consistent with the model's results. pdf.
Presented at the 2016 meeting of the American Political Science Association and the 2016 meeting of the European Political Science Association.
How do parties respond to the electoral manipulation attempts of their competitors? To answer this question we study the allocation of party representatives or poll watchers to polling stations in an electoral environment in which fraud, clientelism, and other manipulation tactics are common. Using polling station data from the Mexican Chamber of Deputies elections, we find a robust positive correlation between the presence of party representatives and that party's vote share. We present further evidence that suggests that such correlation can be attributed to party representatives directly influencing the electoral results. We also formulate a game theoretic model of party representative allocation and structurally estimate its parameters. The results show that parties send their representatives where they expect their opponents to send their own as well as to more competitive districts and precincts within the district. pdf. supplemental material.
Presented at the 2017 meeting of the Midwest Political Science Association, the 2017 meeting of the European Political Science Association, and the 2017 meeting of the American Political Science Association.
How does electoral manipulation affect elected officials' incentives to satisfy their constituents? The literature has highlighted the role of elections as a way to keep politicians accountable to the voters. Yet, we know little about how electoral processes whose integrity has been compromised affect public officials' actions while in office. This paper presents a formal model of electoral accountability in which electoral manipulation can occur during the campaign. Three findings are derived from the model: i) rent extraction increases with levels of electoral manipulation, ii) the value of holding public office is positively related to rent extraction for high values of office, and iii) electoral manipulation increases with the value of office. Using a fuzzy regression discontinuity design that exploits rules determining the sizes of polling stations in Colombia, we find a positive effect of vote buying during mayors' elections on the likelihood of the election winner being sanctioned for violating laws governing public officials. Consistent with our theory, the data show that higher values of office are not linked to less sanctions, but are associated with more reports of vote buying.
Presented at the 2012 meeting of the Midwest Political Science Association.
We propose an estimator of national electoral thresholds of representation and of electoral disproportionality from observed seats/votes data. We apply the procedure to 119 electoral systems used in 417 elections to the lower house across 36 European countries since WWII. We find that over half of these systems exhibit a statistically significant positive national threshold of representation. Furthermore, the two modal electoral system configurations in our data involve positive thresholds with proportional allocation for parties exceeding thresholds (37.82% of all systems) and disproportional seat allocations with (statistically) negligible thresholds of representation (30.25% of all systems). We also develop procedures to evaluate model fit and to test for statistically significant changes in electoral institutions. Finally, we propose a new coding scheme to map empirically complex electoral institutional provisions into a predicted level of electoral disproportionality and electoral thresholds; the resulting model accounts for over 50% and almost 90% of the variation in estimated thresholds and disproportionality, respectively.
Presented at the 2015 meeting of the American Political Science Association.
This paper documents a robust positive correlation between community group participation and occurrences of vote buying attempts in Latin America. Instrumental variable estimates and results of panel data models that account for time-invariant unobserved voter characteristics indicate that a more vibrant associational life facilitates this widespread form of electoral manipulation. Contrary to the expectations derived from the traditional literature on social capital, the findings show that institutions of civic participation can potentially be exploited to increase the efficiency of electoral strategies of manipulation, reducing accountability and affecting the quality of democratic processes. pdf.
Post-instrument covariates are often included in IV analyses to address a violation of the exclusion restriction. We demonstrate that even in linear constant-effects models with large samples: 1) such conditioning turns a(n) (natural) experimental study into an observational study, 2) invariance between IV estimates (with and without post-instrument covariates) does not imply that the exclusion restriction holds, 3) OLS with an omitted variable will often have less bias, 4) measurement error in the post-instrument covariate does not necessarily attenuate the effect, and 5) the bias of OLS and IV are related. Therefore, if used, IV with a post-instrument covariate should always be paired with OLS, and results should be discussed in concert. We illustrate these points with a re-analysis of Acemoglu, Johnson, and Robinson (2001), showing that for the paper's claims to be valid, at least 35% of the variance in the causal variable must be due to measurement error. pdf.