Faculty Name: Department Name

"The Spillover Effects of Conditional Cash Transfers on Primary Care and Non-beneficiary Health: Evidence from Brazil"

Abstract
Conditional cash transfer programs that incentivize health service-seeking behaviors have become a paradigm of poverty-relief in low- and middle-income countries, including those with health system capacity constraints. Targeting complexities mean many poor families may be excluded from program benefits, while still being dependent on increasingly-strained health systems. I investigate the impacts of the largest and longest-lived conditional cash transfer program in the world, Brazil's Bolsa Família, on the health of non-beneficiaries living in poverty. I use pre-Program survey data to model the probability of death for infants born into poor households. Model predictions generate a counterfactual with which to compare infant death risk for poor beneficiary and non-beneficiary households observed after the start of the Program. Infants from beneficiary families had a lower risk of death than predicted by pre-Program trends, while those from impoverished non-beneficiary families developed a higher-than-expected risk, over the same period. This evidence is consistent with the hypothesis of negative health spillovers. I further examine the Program's consequences for primary care dynamics as one likely mechanism for program spillovers. Using fixed-effects regression, I find that program expansions at the municipal level depressed local spending on primary care. Mediated regression suggests a fiscal substitution effect where primary care budgets compete for resources with new administrative expenses from monitoring compliance with health conditionalities. This research demonstrates why conditional program designs should include a supply strategy to accommodate service volume dynamics and administrative overhead from compliance monitoring. It also suggests program evaluations should consider broader system impact and non-beneficiary health.

"Social Identity, Threat-Perception and Prosocial Politics in Latin America"

Abstract
Support for democracy is low on average across Latin American countries, despite significant national and subnational variation. A robust literature links personal security to democratic values, though the mechanisms undergirding this relationship remain poorly understood. Some research has identified perceptions of insecurity as a better predictor of democratic values than self-reports of experience with violence itself. There seems, too, to be a cognitive component to prosocial behavior in the context of physical insecurity. Cooperative behavior in the presence of threat may be contingent on one’s assessment of the source of the threat and perceptions of personal versus shared victimhood. The literature that links insecurity to a distrust in political interdependence cannot yet account for these contingencies. I seek to address the complex interaction between social cleavages, experience with violence, and threat perception in explaining attitudes toward democracy using hierarchical modeling of Latinobarometro survey data. I analyze whether the type of narrative that informs an individual's threat-perception following a violent episode mediates the relationship between violence and support for democracy. I further examine the way these narratives vary depending on social identity and the type of violence experienced. The work synthesizes psychological research on cognitive processing of trauma with political science research on the ways in which voters' predispositions affect their electoral behavior.

"Waiting Room Politics: Using Electoral Signaling to Measure Health Good Clientelism in Brazil"

Abstract
In lower-capacity states with unmet demands or long wait times for public health services, opportunities for politicians to provide health services in a targeted fashion for electoral gain abound. In Brazil, clientelistic use of health procedures, medical devices and medical transport are recognized as a problem in popular discourse, and are even occasionally prosecuted. The phenomenon, however, remains difficult to demonstrate systematically, making both the scale of the problem and the context that facilitates it elusive. I propose a novel measurement strategy to capture a given municipality’s propensity toward health clientelism. Electoral ballots in Brazil identify candidates by self-declared nicknames which I argue often serve as an observable mechanism through which candidates signal a comparative advantage in the targeted distribution of health goods to voters. When a large percentage of successful city council candidates in a municipality identified themselves on the ballot by medical pseudonyms like “Dr Tonho,” “João da Ambulância” (João the Ambulance Driver), or “Luana da Farmácia” (Luana of the Pharmacy), I treat this as an indicator of the prevalence of health clientelism in a municipality. To validate the metric, I examine whether it predicts the per capita frequency of elective medium- and high-complexity hospital procedures, publicly-dispensed medical devices, and non-medical (administrative, maintenance, security) public-sector health personnel in a municipality during that city council’s four-year term. I then evaluate whether this dynamic is mediated by political factors like incumbency, partisanship and mayoral coalition membership, as well as contextual attributes like municipal capacity (completeness of death certificate documentation) and poverty rate.

"Income and Public Service Demands: Comparative Voter Efficacy in Brazil"

Abstract
Many models of public service distribution in democracies predict the poor will have high voting leverage over distributive policy due to a numerical advantage in universal suffrage political competition. Empirical studies do not bear this prediction out, especially in highly unequal democracies, where canonical models predict the poor will have the greatest leverage. This paper proposes an argument that explains the apparently low weight placed by politicians on the preferences of the poor with respect to public service policy in unequal democracies. I show that even when accountability mechanisms function properly in democracy, the poor may find themselves at an electoral disadvantage. This occurs when the poor's (likely higher) public service demands are divided more symmetrically across competing services. When the better-off pile the weight of their votes on fewer services, their votes are more responsive to a unit shift in spending, even if their total service demands are lower. This leads the spending priorities of vote-maximizing, tactically-spending politicians to more closely reflect the preferences of the more concentrated demands of the better-off than of those with higher total state dependence for services. I illustrate the argument and its implications using a study of local public health service allocation in Brazil in the context of a shock to the public primary care service dependency-level of a subset of poor voters induced by a federal transfer program. I contrast voter demands for services with vote responsiveness to service spending using original survey data I collected in Brazil in the two weeks prior the 2012 municipal elections. This research updates our understanding of accountability in unequal democracies, suggesting that the poor do not necessarily fail to hold democratic politicians accountable as many theories would suggest; rather, democratic politicians may have the incentive to prioritize the preference-ranking of the less state-dependent over those more dependent on the public services in question if the less-dependent also have less diffuse service preferences.

"Learning from Social Data when Researchers and Social Actors Share Prior Beliefs"

Abstract
Bayesian analyses are often critiqued on the basis of dubious exchangeability claims regarding the data. Not only must observed data be exchangeable, but prior "data" must be as well, and the observed data must also be exchangeable with the prior data---an assumption not typically justified by the practitioner. Yet social scientists often utilize social data---observed human behaviors that rely on human judgment---to make inferences. Social priors shared by the researcher are, therefore, non-exchangeable with social data. One common defensive argument offered by Bayesian practitioners is that as long as there is some component of new information in the observed data, repeated observation-updating cycles will still eventually produce a highly informative posterior distribution. In frequentist statistics we have power analyses---a way of estimating how much data we need to get desirable properties from our estimator. Here I develop a model that parameterizes the degree of non-exchangeability between the observed data and the prior data and offers a standard way to calculate how many observations are needed to achieve a parameterized definition of an "informative" Bayes' estimate.

Marika Csapo

Peer-Reviewed Publication

Other Writing

Working Papers

Book Project