[Published at World Development 136 (2020): 105111]

Co-author(s): Santiago Medina, Gustavo Yamada, Pablo Lavado, Miguel Núñez, Hugo Alatrista, Juandiego Morzán

Abstract: Crime perception has increased in Peru, as in other developing and developed countries, in spite of the reduction in crime victimization figures. Our hypothesis is that the news industry is partially responsible for such opposing trends. As Peruvians are great consumers of written news, we focus on the written press. Using a unique database of written news, we georeference the location of each reported crime to identify short-term deviations from trend in the coverage of crime news at the province level and estimate their effect on crime perception. We measure coverage as the area an article occupies in cm2. We find that a spike of negative crime news increases people’s perception about the probability of being a crime victim. The effect of positive news is opposite. However, the effect per cm2 of negative news is almost three times larger than the effect of positive news in absolute value, signaling a potential asymmetry in the revision of people’s expectations. The effect of the written press is stronger for men and non-victims. Moreover, perception changes are mostly driven by increases in the fear of house and car theft and common street crime, rather than more violent crimes like kidnapping or sexual abuse. Finally, we delve into the possible consequences of worsening the mismatch between crime perception and crime victimization.