The Crime Elasticity of Demand: Do Viral Crime Events in the New York City Subway Depress Subway Demand?

Abstract

This paper investigates whether viral crime events in the New York City subway system affect ridership patterns to affected stations and how these effects may vary by time and demographic characteristics of one’s origin neighborhood. Leveraging a unique data set that combines MTA origin-destination trip data from 2021-2024 with a novel measure of crime virality derived from Google Trends, this paper employs a Difference-in-Differences fixed effects framework to test this relationship. The analysis asks four research questions: (1) Do viral crime events depress ridership at affected stations? (2) Do off-peak and weekend ridership patterns react differently to viral crime events compared to rush-hour trips? (3) Are the impacts of viral crime events heterogeneous across origin neighborhoods differing in median income, racial composition, female ratio, and transit reliance? (4) Does the ”virality” of a viral crime event alter its effect on ridership? The paper finds no evidence of a conclusive trend between viral crime events and subway ridership patterns, suggesting that subway demand seems to be inelastic with respect to these events. The paper provides a novel offering to a niche literature at the intersection of behavioral economics and urban policy in the digital age, and provides a starting point for future research in using large-scale urban data sets.