Traffic crash: Are they cause of changes in driver behavior?

C. Bolancé, M. Guillen Estany, A. M. Pérez Marín

This study estimates the causal effects of traffic crashes on subsequent driving behaviour using weekly data from an insurtech company. It focuses on weekly kilometres driven, speeding behaviour and the share of urban and nighttime kilometres driving. A difference-in-differences quasi-experimental design is used to estimate causal effects of a crash. The treatment group consists of individuals with an at-fault crash during the observation period. The treatment period corresponds to the week in which the crash occurs, preceded by a 12-week pre-treatment period and followed by a 12-week post-treatment period. The control group consists of individuals without crashes. The causal effects are estimated using parametric models, and robustness is checked through a semiparametric estimation based on inverse probability weighting. Our main result is that an at-fault crash has a statistically significant negative effect on the four driving outcomes, that gradually approaches zero.

Keywords: at-fault crash, quasi-experimental design, difference-in-differences, insurtech

GT Análisis de Riesgos
September 2, 2026  5:40 PM
Aula 20

L. Bermúdez Morata, I. Morillo Lopez, A. Salazar Belver

I. El Ferjani, M. Santolino Prieto

J. M. Sarabia, M. Guillen Estany, J. Navarro, E. Gómez-Déniz

J. S. Yanez Pinto, M. Guillen Estany, P. Roszkowska, J. P. Nielsen