Working Papers:
The Electoral Effects of Voting Technology: Evidence from Bulgaria
(with Teodora Tsankova), April 2025
Related policy brief: Whose Vote Counts
abstract
Can voting technology improve the integrity of elections in developing democracies? We study this question in the context of Bulgaria’s transition from paper ballots to voting via direct-recording electronic machines -- a measure introduced with the goals of improving the accuracy of elections, enhancing ballot secrecy and reducing opportunities for human interference with the voting process. Our empirical strategy leverages a sharp discontinuity in the rule for the allocation of voting machines across polling stations, and variation in the implementation of machine voting over nine consecutive general elections. We document two main results. First, machine voting significantly increases the share of valid votes, effectively increasing the likelihood that votes -- especially those cast by less educated, elderly or ethnic minority voters -- are counted toward the electoral outcome. Second, machine voting causes a large and significant reduction in turnout, particularly in poor and rural areas. Decomposing this decline, we find that it is driven by a reduction in votes for parties that were locally dominant at baseline, while we find no change in votes for other parties. We conduct representative surveys to further investigate mechanisms related to the reduction of bought or fictitious votes, as well as alternative mechanisms related to voters’ aversion to new technologies.In Their Shoes: Empathy through Information
(with Marianne Andries, Leonardo Bursztyn, Thomas Chaney and Alex Imas), February 2025
Revise & Resubmit – Quarterly Journal of Economics
abstract
We explore the mechanics of empathy. We show that information about an outgroup can potentially activate and magnify empathy when presented in conjunction with an experience simulating their struggles. This response increases the willingness to help the struggling group, but it is only activated when the information comes before the experience and not after. We provide evidence for this effect in an immersive virtual reality experiment where participants (“witnesses”) simulate the struggle of unauthorized migrants (“protagonists”). These results are then replicated in a series of controlled lab experiments. We show that this effect operates through an increase in interpersonal similarity, or relatability. If information shifts perceptions of relatability, which changes people’s experience when witnessing the protagonist’s struggles, then it magnifies their empathetic response and drives them to engage in more prosocial behavior. Together, our evidence suggests that the ability to put oneself in the shoes of another person or group can be enhanced by activating empathy through simple information provision.Experience, Narratives, and Climate Change Beliefs
(with Ruben Durante, Elliot Motte and Eleonora Patacchini), November 2024
Revise & Resubmit – Journal of the European Economic Association
abstract
Linking the location and timing of US-based natural disasters to large-scale electoral survey data, we study how the experience of a natural disaster affects climate change beliefs and how experience interacts with ideology. Contrary to the predictions of standard learning models, we find evidence for divergence in beliefs: exposure to the same disaster event increases stated climate change and environmental concerns among liberals but decreases them among conservatives, widening the ideological gap by 11-17%. We further provide evidence of conflicting ideological media discourse on climate change in the aftermath of disasters by applying GPT as a novel text annotation approach. Our findings are consistent with natural disasters making the debate around climate change and partisan cleavages on this issue more salient and further polarizing initial beliefs. We discuss implications for the timing of efforts to build consensus on climate action.Publications:
The Impact of Online Competition on Local Newspapers: Evidence from the Introduction of Craigslist
(with Ruben Durante and Greg Martin)
Review of Economic Studies, May 2024
abstract
How does competition from online platforms affect the organization, performance, and editorial choices of newspapers? What are the implications of these changes for the information voters are exposed to and for their political choices? We study these questions using the staggered introduction of Craigslist (CL)—the world’s largest online platform for classified advertising—across U.S. counties between 1995 and 2009. This setting allows us to separate the effect of competition for classified advertising from other changes brought about by the Internet, and to compare newspapers that relied more or less heavily on classified ads ex ante. We find that, following the entry of CL, local newspapers reliant on classified ads experienced a significant decline in the number of management and newsroom staff, including in the number of editors covering politics. These organizational changes led to a reduction in news coverage of politics and resulted in a decline in newspaper readership, particularly among readers with high political interest. Finally, we document that reduced exposure to local political news was associated with an increase in partisan voting and increased entry and success of ideologically extreme candidates in congressional elections. Taken together, our findings shed light on the determinants of the decline of print media and on its broader implications for democratic politics.Media Slant and Public Policy Views
(with Ruben Durante, Elliot Motte and Eleonora Patacchini)
American Economic Association Papers and Proceedings, May 2024
abstract
We study how exposure to partisan news channels (Fox News and MSNBC) affects individual views on four policy issues: climate change, gun rights, abortion, and immigration. First, using GPT to annotate news transcripts, we document large differences in the way the two networks cover these issues. Second, exploiting exogenous variation in viewership due to channels' positions in cable lineups, we show that exposure to Fox News (MSNBC) is associated with more conservative (progressive) views, even when controlling for self-reported ideology and party affiliation. Our findings indicate that partisan media contribute to the rise of political polarization in the United States.Persuasion through Slanted Language: Evidence from the Media Coverage of Immigration
American Economic Review, March 2023
Econonimate, AEA Highlights
abstract
I study the persuasive effects of slanted language, exploiting a ban on the politically charged term "illegal immigrant" by the Associated Press (AP) news wire. My empirical strategy combines the timing of the ban with variation across media outlets in their baseline reliance on AP copy. I document sizable diffusion of the ban from AP copy to media outlets. Moreover, individuals exposed to the ban through local media show significantly lower support for restrictive immigration policies. This effect is more pronounced for moderates and in locations with fewer immigrants, and does not transfer to views on issues other than immigration.Media Attention and Strategic Timing in Politics: Evidence from Presidential Executive Orders
(with Ruben Durante)
American Journal of Political Science, December 2021
abstract
Do politicians tend to adopt unpopular policies when the media and the public are distracted by other events? We examine this question by analyzing the timing of executive orders signed by U.S. presidents over the past four decades. We find robust evidence that executive orders are more likely to be signed on the eve of days when the news is dominated by other important stories that can crowd out coverage of executive orders. This relationship only holds in periods of divided government when unilateral presidential actions are more likely to be criticized by Congress. The effect is driven by executive orders that are more likely to make the news and to attract negative publicity, particularly those on topics on which president and Congress disagree. Finally, the timing of executive orders appears to be related to predictable news but not unpredictable ones, which suggests it results from a deliberate and forward-looking PR strategy.The Great Recession and Social Preferences: Evidence from Ukraine
(with Ralph De Haas and Elena Nikolova)
Journal of Comparative Economics, February 2016