A decomposition procedure was used to estimate the percentage contributions of exposure, risk of collision and fatality to explain the decreases in pedestrian collision death rates observed in Spain from 1993 to 2011. Information was obtained from the Spanish National Institute of Statistics and the Spanish Register of Road Crashes with Victims. A quasi-induced method was used to obtain estimates of annual pedestrian exposure. Poisson regression was used to obtain age-adjusted and sex-adjusted estimates of the mean annual percentage decrease in pedestrian death rates attributable to exposure, collision and fatality; these values were +2.28%, -45.86% and -51.86%, respectively. The results suggest that the decrease in fatality rates and, to a lesser extent, collision rates were the most important determinants of the reduction in pedestrian collision death rates, whereas the contribution of exposure was irrelevant.