Can property values capture changes in environmental health risks? : evidence from a stated preference study in Italy and the UK

Hedonic property value models are often used to place a value on localized amenities and disamenities. In practice, however, results may be affected by (i) omitted variable bias and (ii) whether homebuyers and sellers are aware of, and respond to, the assumed environmental measure. In this paper we undertake an alternative stated preference (SP) approach that eliminates the potential for unobserved confounders and where the measure of environmental quality is explicitly presented to respondents. We examine how homeowners in the United Kingdom and Italy value mortality risk reductions by asking them to choose among hypothetical variants of their home that differ in terms of mortality risks from air pollution and price. To our knowledge this is the first stated preference study examining respondents' willingness to pay for properties using a quantitative and clearly specified measure of health risks. We find that Italian homeowners hold a value of a statistical life (VSL) of about €6.4 million, but UK homeowners tend to hold a much lower VSL (€2.1 million). This may be due to the fact that respondents in the UK do not perceive air pollution where they live to be as threatening, and actually live in cities with relatively low air pollution levels. Exploiting part of our experimental design, we find that Italian homeowners value a reduction in the risk of dying from cancer more than from other causes, but UK respondents do not hold such a premium. We also find that those who face higher baseline risks, due to higher air pollution levels where they live, hold a higher VSL, especially in the UK. In both countries, the VSL is twice as large among individuals who perceive air pollution where they live as relatively high.