(iPoster) The Paradox of Censorship: Journalist Assassinations and News about Corruption

• In Event: Poster Session: Crime, Violence, Institutions in Comparative Politics

Abstract

Every year, thousands of journalists are murdered while investigating corruption, especially in countries where the rule of law is weak. These assassinations, intended to suppress information, raise critical questions about their actual effects on the news media. Do they create a chilling effect that silences reporting, or do they provoke a rally-round-the-flag response that amplifies scrutiny? Understanding this dynamic is crucial for comprehending how violence influences information exchange, civic organization, and the political accountability of corrupt actors.
This study examines the impact of journalist assassinations on media coverage of municipal corruption in Mexico, one of the most dangerous countries for investigative journalism. Drawing on data from all journalist murders between 2006 and 2021, we employ a sharp Regression Discontinuity Design (RDD) to estimate the causal effect of 54 assassinations on news coverage of the corruption cases those journalists were investigating. Our findings consistently show that assassinations increase the amount of media attention devoted to these corruption cases, contradicting the expectation that violence provokes self-censorship.
To validate and extend these results, we address potential limitations stemming from the reliance on human intervention in the data collection process. The initial analysis, while thorough, was constrained by the simplicity of the search dictionary and the need for manual classification of results. Corruption-related news often lacks clear indicators in headlines, and despite efforts to blind the research assistant to the study’s hypotheses, unconscious biases could have influenced classifications. To mitigate these challenges, we incorporated a novel approach in a second analysis: using automated web scraping to collect data directly from the Google News aggregator. This methodology reduces human error, increases reliability, and allows for a more exhaustive exploration of corruption-related reporting. Regardless of the data collection method, functional assumptions about time trends, or the distribution of the dependent variable, our results consistently demonstrate that assassinations increase attention to governmental corruption.
Further analysis reveals that this increase in coverage is driven by a “militant reaction” from the press, wherein journalists amplify attention to corruption cases as an act of defiance and solidarity with their murdered colleagues. Interestingly, the effect persists even after controlling for direct references to the murders themselves, indicating a broader shift in media focus toward governmental misconduct following these violent events.
These findings challenge conventional views on the coercive power of violence over the press, instead highlighting the potential for backlash effects that intensify scrutiny of corrupt actors. This study contributes to our understanding of the interplay between violence, information flows, and political accountability, offering critical insights into how the press navigates environments of coercion and impunity.
By examining the resilience of media institutions in the face of violence, this work aligns with the conference theme, “Reimagining Politics, Power, and Peoplehood in Crisis Times.” It highlights the adaptive strategies of civic actors in unsettled conditions and advances broader conversations about the limits of power and the dynamics of information exchange in fragile democracies.

Authors

• Daniel Zizumbo-Colunga, CONAHCYT

• Maria del Pilar Fuerte, Centro GEO