Redistributing income under proportional representation: a correction (with M. Morelli), Journal of Political Economy, 127 (2019), 458-62.

Preferential votes and minority representation in open list proportional representation systems, Social Choice and Welfare, 50 (2018), 281303.

Size invariant measures of association: characterization and difficulties (with Yves Sprumont), Mathematical Social Sciences, 75 (2015), 115-122.


Working papers

Electoral Systems, Taxation and Immigration Policies (with M. Morelli)
When exposed to similar migration ows, countries with dierent institutional systems may respond with different levels of openness. We study in particular the different responses determined by different electoral systems. We find that Winner Take All countries would tend to be more open than countries with PR when all other policies are kept constant, but, crucially, if we consider the endogenous differences in redistribution levels across systems, then the openness ranking may switch.

Scoring Rules and Level-k Thinking in Games with Population Uncertainty (with F. Maniquet)
We study the functioning of scoring rules in presence of boundedly rational voters. We are interested in understanding their ability to i) deliver a clear winner of the election and ii) elect the Condorcet winner when it exists. We model bounded rationality in two ways: first, we assume that voters are unable to make correct predictions about how the others will vote, as in the level-k thinking theory; second, we assume voters are not perfectly able to compute the likelihood of pivotal events. In this framework, we find a trade-off between ensuring the election of the Condorcet winner, when it exists, and delivering a sure winner in presence of Condorcet cycles.

A New-Median Theory of Populism (with M. Morelli)
What explains the entry of populists with a positive chance of winning in a democratic competition? And why can’t incumbents preempt their entry by modifying their policy platforms? We propose a multidimensional preferences citizen candidate model to answer this question. We interpret one of the dimensions as immigration, globalization or robotization, topics typically emphasized by populist rhetoric with simple “closure” positions. We find that incumbents can prevent populist entry only if less than 1/3 of the population supports anti-immigration/globalization/robotization policies, or if the issue is not sufficiently salient. We also characterize the redistributive policies championed by populists when they enter the electoral competition.