Sunset Long Shadows : Time, Crime, and Safety Perception

I investigate the long run effect of daylight on crime and fear of crime. Forty years of reforms shifted the boundaries between Russian eleven time zones. Differently from daylight saving time transitions, these changes are permanent and allow for identification of their long run effect. I find that a switch to a later sunset leads to a two year long decrease in robbery, and the lower level of robbery persists afterwards. The magnitude of each post-reform year's marginal effect is similar to the regression discontinuity estimates from other countries at daylight saving time transitions. I confront this finding with individual perception of safety. Longitudinal survey data shows that women's feeling of safety ``remembers'' the daylight regime many years after it changed. However, men's feeling of safety is only affected by the two preceding years