We present an empirical estimation of the distribution of WTP for effective speed reduction via implementation of local traffic calming schemes. Random samples are drawn from the populations of households of three villages intersected by main trunk roads with varying through traffic conditions. We retrieve the underlying WTP distributions from discrete-choice responses to site-specific contingent valuation studies accounting for zero-bidders. We then test the hypothesis of different distributions across villages. The statistical analysis is first conducted by means of a parametric specification and then by a totally non-parametric one. Stated welfare changes for effective speed reduction are found to be sizeable and the parameters of the random utility models are related to differences in objective speed measures.