Abstract
This paper presents a micro-location model of public investment in pedestrian safety capital. A special case of the model predicts that economies of scale in safety capital can offset the effect of rising population density on the pedestrian fatality rate. Using countylevel data we confirm this prediction empirically and measure the elasticity of the fatality rate with respect to civil time of sunrise and sunset, sales at bars, highway lane miles, income, climate, and tourism. Pedestrian fatalities on interstate highways are shown to differ from those elsewhere. Other accidents are shown to be the best pedestrian exposure measure on interstates. (Author/publisher)