Research Interests: Political Economy, Health Economics, Applied Econometrics




Electoral Uncertainty and the Stability of Coalition Governments

Abstract: In multiparty parliamentary democracies government coalitions frequently reshuffle the allocation of cabinet posts, and cabinets terminate before the end of the legislature. I interpret these events as equilibrium outcomes of a strategic interaction among political parties. Parties' incentives to terminate the government and seek early elections depend on future electoral prospects, and electoral opinion polls convey information regarding possible shifts in the electoral support. I develop a dynamic strategic model of government formation and termination, and structurally estimate the model using newly collected data on eleven Western European parliamentary democracies over the period 1970-2002. Using the estimated model I conduct counterfactual experiments aimed at evaluating the effects of poll informativeness and institutional features on the survival probabilities of coalition governments.


Good Old Spendthrift. The Fiscal Effects of Political Tenure

(with Andrea Mattozzi) – coming soon!

Abstract: Using a newly collected data set covering sixty-two countries over the period 1945-2006, we construct a number of measures of political tenure and we document a robust negative (positive) effect of political tenure on the size of government surplus (expenditure). We also provide a theoretical model that accounts for the established facts.


Negative Advertising and Political Competition

(with Amit Gandhi and Carly Urban)

Abstract: Why is negative advertising such a prominent feature of competition in the US political market? The typical election in the United States is a two-candidate race. This duopolistic competition provides stronger incentives for "going negative" relative to non-duopoly contests: when the number of competitors is greater than two, engaging in negative ads creates positive externalities for opponents that are not the object of the attack. In contrast, positive ads benefit only the advertiser. To investigate the empirical relevance of the "fewness" of competitors in explaining the volume of negative advertising, we collected information about all candidates running for a US non-presidential primary contest in 2004 and 2008. The nature of primaries provides us with a cross section of independent races and large variation in the number of entrants. We merge these data with specific information on the political advertisements aired during the campaigns from the Wisconsin Advertising Project. The findings are striking: duopolies are twice as likely to air a negative ad when compared to non-duopolies, and doubling the number of competitors drives the rate of negative advertising in an election close to zero. The estimates are robust to various specification checks and

the inclusion of potential confounding factors at the race, candidate and ad levels.

Negative Advertising and Political Competition – on line supplement



Caught in the Bulimic Trap? Persistence and State Dependence of Bulimia Among Young Women

(with Michelle Sovinsky and John Ham)                                                                         

Abstract: Eating disorders are an important and growing health concern, and bulimia nervosa (BN) accounts for the largest fraction of eating disorders (ED). Health consequences of BN are substantial and especially serious given the increasingly compulsive nature of the disorder. However, remarkably little is known about the mechanisms underlying the persistent nature of BN. Using a unique panel data set on young women and instrumental variable techniques, we document that unobserved heterogeneity plays a role in the persistence of BN, but strikingly up to two thirds is due to true state dependence. Our results, together with support from the medical literature, provide evidence that bulimia should be considered an addiction. Our findings have important implications for public policy since they suggest that the timing of the policy is crucial: preventive educational programs should be coupled with more intense (rehabilitation) treatment at the early stages of bingeing and purging behaviors. Our results are robust to different model specifications and identifying assumptions.


Selected Press Reactions: Chicago-Sun Times; Guardian; Science Update (radio feature for Science); Teen Vogue; USC news release; PBS Tavis Smiley

Supplement Material


Race, Social Class and Bulimia Nervosa

(with Michelle Sovinsky and John Ham)                                                                         

Abstract: In this paper we explore a serious eating disorder, bulimia nervosa (BN), which afflicts a surprising number of girls in the US. We challenge the long-held belief that BN primarily affects affluent White teenagers, using a unique data set on adolescent females evaluated regarding their tendencies towards bulimic behaviors independent of any diagnoses or treatment they have received. Our results reveal that African Americans are more likely to exhibit bulimic behavior than Whites; as are girls from low income families compared to middle and high income families. The fact that our results stand in stark contrast to the popular conceptions of who is most likely to struggle with bulimia may arise from differences in diagnosis across racial and income classes. Our findings have important implications for public policy since they provide direction to policy makers regarding which adolescent females are most at risk for BN. Our results are robust to different model specifications and identifying assumptions.


Selected Press Reactions: Chicago-Sun Times; Guardian; Science Update (radio feature for Science); Teen Vogue; USC news release; PBS Tavis Smiley



Education, HIV Status and Risky Sexual Behavior: How Much Does the Stage of the HIV Epidemic Matter?

(with Raül Santaeulŕlia-Llopis)

Abstract: We study the relationship between education and individual HIV status using nationally representative data (Demographic and Health Surveys, DHS) for 18 countries in Sub-Saharan Africa (SSA). Because the diffusion of knowledge on HIV prevention- hence, actual change in sexual behavior - may differ across education groups, we explicitly explore the possibility of a dynamic relationship between education and the probability of being infected with HIV over aggregate stages of the HIV epidemic. Our contribution is twofold. First, we construct an innovative algorithm that positions, for any set of countries, the country-specific evolution of the HIV epidemic in a unified framework - a normalized epidemiological space- to define stages of the HIV epidemic in a comparable manner across SSA countries. Second, using this framework, we exploit epidemiological stage variation across DHS countries and find that the relationship between education and individual HIV status is dynamic and significantly evolves with the course of the epidemic. Specifically, we show that the education gradient of HIV displays a large U-shaped (positive-zero- positive) pattern over the aggregate stages of the HIV epidemic.


Companion site