Journal of Development Economics · 2026 · [{"name": "José A. Jurado", "affiliation": ["Arizona State University"]}, {"name": "Juan S. Morales", "affiliation": ["Wilfrid Laurier University"]}]
Abstract
This paper studies how attacks against journalists affect news media activity. We combine data on content shared by Mexican media outlets on Twitter with records of journalist assassinations between 2010 and 2020, and use difference-in-differences methods that are robust to staggered treatment and heterogeneous effects. Our analysis shows that reporting activity of affiliated outlets declines significantly in the months after journalists are killed, relative to outlets not yet exposed to a killing. Additionally, we observe subtle shifts in the tone of news coverage and a significant increase in grammatical errors. Using census data, we also document that the number of active journalists declines in places where killings occur, and that their demographic characteristics shift: remaining journalists tend to be younger, less educated, less likely to be married, and less likely to have children.