The Effectiveness of a Large-scale Flood Risk Awareness Campaign – Evidence from Two Panel Data Sets

In the context of integrated flood risk management, governments and public agencies aim to strengthen flood risk reduction and insurance at the household level. They often employ awareness campaigns in order to educate the public about flood risk and possible coping strategies. However, the effectiveness of these awareness campaigns has hardly been examined empirically, and the available analyses come to ambiguous conclusions. Evaluations based on longitudinal data of actual household behavior are missing. In this study, we perform two independent empirical analyses to assess whether a large-scale awareness campaign in Germany affected private flood protection or insurance behavior. We thereby exploit the fact that different federal states initiated the campaign at different points in time between 2009 and 2017. In the first analysis, we use a longitudinal data set of a national sample of 6,729 household heads in the years 2012 and 2014, and focus on three federal states that launched the campaign in 2013. In the second analysis, we use flood damage and insurance penetration data at the federal state level from 2002 to 2018 (N=256). Based on a fixed effects difference-in-differences estimator, we do not find that the campaign had a significant effect on households’ behaviors, damage, or insurance penetration. The results show that large-scale flood risk awareness campaigns, as they are currently conducted in many countries, have a limited effectiveness in terms of strengthening actual flood protection or insurance behavior of households.

Osberghaus, Daniel and Hendrik Hinrichs (2021), The Effectiveness of a Large-scale Flood Risk Awareness Campaign – Evidence from Two Panel Data Sets, Risk Analysis 41(6) , 944-957