Teresa Esteban Casanelles

Teresa Esteban Casanelles

Research

Working Papers

Is an Image Worth a Thousand Votes? Evidence from a Natural Experiment in Political Advertising

This paper examines the causal effects of electoral campaigns on voting behavior in multi-party democracies, where voters choose among parties differing along multiple dimensions. Using novel geolocated data on street-level ads, I find that an increase in ads significantly increases a party’s vote share. The impact is heterogeneous across party and voter types, being larger for new parties and candidates, and more persuasive for ideologically aligned voters. The analysis supports ad salience and informativeness as the main mechanisms underlying the effect. Finally, results also show spillover effects across parties, with ads benefiting other ideologically similar parties.

Best Applied Paper by a Young Researcher, Econometric Society – European Meeting 2023

Media Coverage: Nada es Gratis

Better the Devil You Don’t Know? Economic Shocks, New Party Emergence, and Changes in Voting Behavior

This paper analyses the effects of an economic shock on the emergence of new parties and other changes in voting parties by using regional variation in the exposure to the shock. I find that a worsening of economic conditions as measured by unemployment rate leads to an increase in electoral competition and volatility. In particular, the deeper the effects of the recession in a area, the larger the number of new parties emerge and become more successful and there is an increase in the changes in vote shares. On the other hand, the vote share of parties previously in government decreases and a decrease in vote share concentration.

Incentives and Strategic Behaviour

with Duarte Gonçalves, accepted at Experimental Economics

How do incentive levels affect strategic behaviour? We address this with an experiment that separately identifies own- and opponent-incentive effects in two dominance-solvable games that differ in strategic complexity. Higher own incentives favour more strategically sophisticated actions and increase best responding to stated beliefs. Beliefs shift in a parallel direction and participants expect more sophisticated opponent actions. Furthermore, although higher own incentives increase belief accuracy, opponents with higher incentives are harder to predict. Higher incentives also increase response times, and longer response times are associated with better performance and more sophisticated actions and beliefs. Taken together, the evidence suggests that incentives affect strategic behaviour through two main channels: by reducing payoff-dependent mistakes and by increasing the effort devoted to reasoning, with the returns to that effort shaped by strategic complexity of the environment.

Work in progress