Exploring the link between flood risk perception and public support for funding on flood mitigation policies

Abstract

Obtaining the support of affected stakeholders is a crucial first step to successful implementation of any public policy, including flood mitigation policies. Public support for flood mitigation policies is largely influenced by flood risk perceptions and flood risk perceptions are, in turn, shaped by a multitude of factors. This paper explores the impact of the determinants of risk perception on willingness to pay for flood risk prevention in Dunkerque (France) using a contingent valuation survey. We find that whether or not respondents include their home within their perceived flood risk areas, trust in flood mitigation measures, environmental beliefs and socio-economic factors are strong predictors of public support for flood risk prevention, whereas actual distance of a respondent’s home to the flood source, knowledge of flood risk, prior experience and trust in local authorities have a limited influence. Policy implications and suggestions for future research are discussed.

Keywords:

• Flood risk perception
• contingent valuation
• willingness to pay

Acknowledgements

We would like to thank anonymous referees for valuable comments and suggestions. We would also like to thank Camilla Knudsen for providing remarks on a first draft that improved the paper and for the English language review. Any remaining errors are the responsibility of the authors alone.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

Notes

1 An industrial facility is classified as a SEVESO site if it displays major technological and environmental hazards.

2 A Seveso site is an establishment that displays major technological and environmental hazards.

3 Initially, a two-step point – interval regression model with selectivity was estimated to control for a potential sample selection bias after excluding protest bidders from the analysis. A probit model estimated the probability of protesting. The parameter estimates were then used to calculate the inverse Mills’ ratio, which was included as an additional explanatory variable in the WTP function. The inverse Mills’ ratio appeared to be non-significant, suggesting that sample selection bias was not present in our dataset.