Le premier chapitre de cette thèse, intitulé « Optimal Vote Buying » (co-écrit avec Leon Musolff, Princeton University) s’intéresse à la corruption de comités. Il s’agit d’une contribution théorique dans laquelle nous montrons qu’il est généralement peu onéreux de corrompre une large majorité de membres du comité. En effet, lorsque le comité est largement corrompu, les membres considèrent que l’option soutenue par celui qui corrompt le comité sera probablement plébiscitée. Aucun vote ne peut alors faire basculer l’élection et les membres du comité sont prêts à accepter un pot-de-vin modeste en échange de leur vote. Dans le deuxième chapitre (co-écrit avec Emeric Henry, Sciences Po) nous étudions les interactions entre le vote et les normes sociales. Nous analysons un modèle dans lequel un groupe choisit sa propre règle. Dans une première étape, chaque membre vote pour ou contre une règle. Une fois le règlement établi, les membres du groupes choisissent une action dont les conséquences dépendent de la règle choisie. Nous étudions comment le vote et le comportement ultérieur interagissent. Enfin, le dernier chapitre (co-écrit avec Etienne Fize) traite de l’impact du service militaire obligatoire en France sur les comportements politiques. Nous trouvons d’une part que les individus ayant fait leur service militaire sont plus enclins à voter. Pour les élections de 2012, nous estimons que l’effet est de l’ordre de 4 points de pourcentage pour les élections présidentielles et atteint les 9 points pour le second tour des élections législatives. Nous étudions également les conséquences sur les préférences politiques. Nous trouvons que les anciens conscrits sont plus à droite.

In this paper, we show that sentencing norms vary widely even across geographically close units. By examining North Carolina’s unique judicial rotation system, we show that judges arriving in a new court gradually converge to local sentencing norms. We document factors that facilitate this convergence and show that sentencing norms are predicted by preferences of the local constituents. We build on these empirical results to analyze theoretically the delegation trade-off faced by a social planner: the judge can learn the local norm, but only at the cost of potential capture.

It is now well documented that individuals, on average, change their behavior when their actions are observed by others. Yet, there is no systematic way of measuring this dimension of preferences at the individual level. In this paper, we propose a novel experimental game to measure the individual sensitivity to image concerns. We show that few socio-economic characteristics can explain the level of image concern. One exception is that members of ethnic minorities seem to be more imaged concerned, in particular when observed by a member of other groups. Men (resp. women) are more image concerned when observed by women (resp. men). Finally, we show that more image concerned individuals tend to be more selfish and find evidence consistent with the fact that they try to avoid situations where their actions risk being visible.

Though a large literature on the determinants of turnout has ourished, there is scant evidence on the causal impact of turnout on policies implemented in practice. Using data on French municipalities and instrumental variables for turnout based on temperature and inuenza variations, we show that a one percent increase in turnout decreases on average the municipalities' yearly budget by 1.5 percent. This is mostly due to a decrease in spending on equipment or personnel. We show that this could be the result of a negative effect of turnout on the strength of the incumbent's majority combined with the fact that the incumbent promises higher budgets. We argue, in the context of a theoretical model, that these different facts could be natural consequences of the well documented incumbency advantage.

In situations where a biased sender provides verifiable information to a receiver, I study how strategic reporting affects the incentives to search for information. Research provides series of signals that can be used selectively in reporting. I show that the sender is strictly worse off when his research effort is not observed by the receiver: he has to conduct more research than in the observable case and in equilibrium, discloses all the information he obtained. However this extra research can be socially beneficial and mandatory disclosure of results can thus be welfare reducing. Finally I identify cases where the sender withholds evidence and for which mandatory disclosure rules become more attractive.

The sale of ideas through licensing facilitates the division of labor between the separate activities of research and development. This vertical specialization can improve the overall efficiency of the innovative process. However, these gains depend on the timing of the sale: the buyer of an innovative project should assume development at the stage at which he has an efficiency advantage. Using data from the pharmaceutical industry, we show that competition between potential buyers is related to the timing of licensing. Furthermore, the effect differs by the type of competitor. We then describe a class of models that yields predictions consistent with these empirical patterns. Our key insight is that increased competition may increase licensing delays and hence inefficiency.

Members of groups and organizations often have to decide on rules that regulate their contributions to common tasks. They typically differ in their propensity to contribute and often care about the image they project: in particular, they want to be perceived by other group members as being high contributors. In such environments we study, from both a positive and normative perspective, the interaction between the way members vote on rules and their subsequent contribution decisions. We show how endogenous norms can emerge. We study in particular the role played by the visibility of individual actions, votes or contributions. While making votes visible always increases welfare in our setting, making contributions public can be welfare decreasing as it makes some rules more likely to be rejected.

How effective is fact checking in countervailing “alternative facts,” i.e., misleading statements by politicians? In a randomized online experiment during the 2017 French presidential election campaign, we subjected subgroups of 2480 French voters to alternative facts by the extreme-right candidate, Marine Le Pen, and/or corresponding facts about the European refugee crisis from official sources. We find that: (i) alter- native facts are highly persuasive; (ii) fact checking improves factual knowledge of voters (iii) but it does not affect policy conclusions or support for the candidate; (iv) exposure to facts alone does not decrease support for the candidate, even though voters update their knowledge. We argue that the main channel is that fact checking increases the salience of the immigration issue.

We introduce a neighborhood structure in a waiting game, where the payoff of stopping increases each time a neighbor stops. We show that the dynamic evolution of the network depends on initial parameters and can take the form of either a shrinking network, where players at the edges stop first, or a fragmenting network where interior players stop first. We find that, in addition to the coordination inefficiency standard in waiting games, the neighborhood structure gives rise to two other inefficiencies, the first linked to the order of exit and the second to the final distribution of remaining nodes.