Valuation of environmental pollution in the city of Madrid: an application with hedonic models and spatial quantile regression

Air pollution and urban noise are major concerns in big cities. This paper aims at evaluating how they impact transaction prices in the city of Madrid. For that purpose, we incorporate some measures of air pollution and noise, joint with other variables available for individual characteristics, accessibility and socio-economic data, and we use spatial quantile models with a sample of 5080 houses in the municipality of Madrid. Quantile regressions are a good instrument to deal with common problems in hedonic models, such as non-normality due to asymmetries and outliers, since it allows for different relationships in the middle and the tails of the price distribution. As it is well-known, valuations at the mean can be misleading in terms of welfare calculation. In this sense, quantile regressions estimates monetary valuation and welfare benefit of pollution at different income-level populations. In fact, we find that air-pollution have no clear significant effect except in the wealthier neighborhoods, where it is clearly a disamenity. Additionally, we find a counter-intuitive positive sign for the effect of noise only for the group of the most exclusive properties, possibly because of their owners? high `sense of place', joint to the existence of some other related positive associated externalities, such as upscale shops and historical or cultural areas.