River flooding and housing values - an economic assessment of environmental risk

The purpose of this article is to explore the determinants of the implicit price of the risk of flooding. We carry out a meta-analysis on estimates resulting from the application of hedonic price models. The variable under scrutiny is the relative change in the price of a house located in a floodplain when compared to a house located outside a floodplain, per level of risk (the reference level being the 100-year floodplain). It appears that the choice of specification of hedonic price models affects the variability between estimates. Factors related to the exploitation of spatial characteristics of the data do also play a role. Besides, a higher level of income goes together with a lower implicit price of risk, as it can be a proxy for a lower vulnerability to material loss. Finally it appears that specific attention has to be paid to the perceived level of risk; elements affecting the perception of individuals on the effective risk of flooding and allowing them to update their perception of the level of risk explain variability between estimates. In areas where a flood recently took place, the relative change in the price of a house due to flood risk is larger than in areas not affected.