Please find my CV here.
"Better the Devil You Know: Selective Exposure Alleviates Polarization in an Online Field Experiment" [Revise and Resubmit, Journal of Public Economics]
This paper empirically investigates the causal link between news consumers’ self-selective exposure to like-minded partisan media and polarization. Through an online experiment, I find a lower likelihood of extreme policy views—an alleviation of polarization—among the subjects who were allowed to choose their like-minded partisan media than those who were not. I begin the paper by presenting a parsimonious model to formalize a traditionally neglected channel through which self-selective media choice leads to reduced polarization. The predictions of this model are supported by experimental evidence collected from a South Korean mobile news application that I created. Users of the app were given access to curated articles on key political issues and were regularly asked about their views on those issues. Some randomly selected users were allowed to choose the news sources from which to read articles; others were given randomly selected articles. The users who selected their news sources showed larger changes in their policy views and were less likely to have radical policy views—an alleviation of polarization—in comparison with those who read randomly provided articles. In support of the main mechanism in the model, an exogenous increase in familiarity enables news consumers to better adjust for the biases of news articles, and this in turn leads to more learning and less extreme policy views.
“Customs and Border Protection (CBP) Activities Mobilize Hispanic Voters” (with Cory Smith and David Lazer) [New Draft!]
Do activities by immigration enforcement agencies suppress or mobilize Hispanic voters? To answer this question, we exploit the sharp discontinuity in the legal authority of United States Customs and Border Protection (CBP) at the boundary of the 100-mile interior US border zone. We find that CBP activities increased Hispanic voter registration and turnout in the 2016 US general election by 1.8pp and 1.5pp, respectively. We suggest that the main mechanism is via personal experiences—observing or hearing about CBP activities—as opposed to elite-driven campaigns that are unlikely to be spatially discontinuous. We also estimate the electoral consequences of CBP activities through a simulation exercise and find a small increase in the Democratic Party's representation in the US Congress.
Many legislative chambers are segregated along party lines, limiting cross-party interaction.Would there be less polarization if politicians were physically integrated? We tackle this question by exploiting random seating in Iceland’s national Parliament. We find null general effects of cross-party proximity – MPs sitting next to other-party MPs are no more likely to vote similarly to other-party MPs, nor do they co-sponsor more frequently with other-party MPs. At the pair-level we find evidence of small and temporary social influence – the likelihood of voting the same way increases by about one percentage point when two MPs from different parties are sitting together. This effect is larger when one MP in the pair is randomly sat on a corner seat,with no escape from their other-party neighbor. These results together suggest that political integration has only weak impacts on bipartisan behavior.
Research in Progress
“Candidate Information and Polarization of US Voters” (with Cory Smith)
“Online Identity and Political Speech in Social Media” (with Hye Young You)
“The Power of Money: Evidence from a Bartering Platform” (with Suhas Vijaykumar, Michael Wong)