Job Market Paper

1. Black by Popular Demand: Media, Competition and the Evolution of a Social Norm Latest Draft

What role do economic incentives play in social change? I study how market forces shaped newspapers’ adoption of a new identity label, exploiting a Civil Rights era campaign in the mid-1960s that sought to shift usage from "Negro" to "Black". I introduce an original dataset created by digitising 58 million newspaper articles from 450 Southern newspapers (1960–1973) and 1964 newspaper circulation audits, yielding measures of newspaper–county circulation flows and local market competition. I show that Southern newspapers began adopting "Black" almost immediately after the shock, and that its use brought more affirmative and agentic portrayals of Black individuals, consistent with the label’s intended meaning. Adoption patterns varied systematically: newspapers facing racially conservative audiences were slower to adjust, and this resistance was amplified in more competitive local markets, where outlets had stronger incentives to align with audience preferences. These effects are not explained by editorial ideology or ownership. Finally, in counties covered by Section 5 of the Voting Rights Act, earlier adoption of "Black” predicted faster growth in Black local officeholding between 1966 and 1974. The findings show how economic incentives in media markets can either accelerate or hinder norm change, depending on how they interact with underlying audience attitudes and market structure.

Working Papers

2. Vertical Governance of Online Speech: Evidence from Google’s Moderation Mandate Latest Draft

This paper shows that infrastructure providers can reshape online discourse by enforcing moderation through access-based leverage. Exploiting Google’s 2022 Play Store policy update in a triple-differences design across three alternative social media platforms, I find sharp and persistent declines in threatening content, especially among high-risk users. Politically sensitive narratives such as election denial also fell, highlighting how infrastructure-level enforcement can redefine the boundaries of online speech.

3. The Political Effects of Threats to the Nation: Evidence from the Cuban Missile Crisis

(joint with Tommaso Colussi)

This project examines how voters and policymakers respond to external threats, focusing on the political and military consequences of the 1962 Cuban Missile Crisis. Using novel measures of local media exposure and exogenous variation in threat salience, we estimate the crisis’s impact on electoral participation, partisan support, congressional behavior, and military service.

Work in Progress

Media & Political

4. From the Pulpit to the Polls: The Electoral Consequences of Christian Talk Radio

This paper studies the political effects of Christian-conservative radio expansion in the United States. Using variation in radio signal strength driven by topography and station acquisitions, I link exposure to county-level voting records and evangelical presence. Greater exposure increased Republican vote share, especially in areas with larger evangelical populations, highlighting how deregulation enabled partisan religious media to shape electoral outcomes.

5. Editorial Bias: Evidence from Historical Newspapers

This paper explores editorial bias showing systematic differences in language choices when newspapers published the same wire articles.

Climate

6. Climate on the Move: How Migration Imports Beliefs on Climate Change

(joint with Matteo Pograxha)

This paper examines how climate shock-induced migration shapes the diffusion of climate change beliefs across international borders.

7. Selective Science: How Media Bias Shapes Climate Change Narratives

(joint with Matteo Pograxha)

This paper examines whether newspapers selectively cover climate science in ways that align with their audiences’ political priors, independent of article quality, shedding light on how media shape public perceptions of climate change.

AI

8. The Introduction and Reduction of Estimation Bias in AI Assisted Research

This paper explores how biases can affect LLM-assisted research and shows how methodological design can improve the reliability of AI-based textual analysis in the social sciences.

Development

9. When to baseline - The impact of lottery allocation household outcomes in a highly vulnerable setting

(joint with C. M. Fernandez, A. Guariso, M. Holmlund, T. Mitchell, and C. Newman.)

This paper examines whether public lottery allocation in field experiments affects survey outcomes, using evidence from villages in Niger.