Conference Agenda

There is mixed evidence on whether individuals adapt to extreme climate events by relocating away from vulnerable areas. Using comprehensive French individual-level administrative data, we examine heterogeneous migration responses to flooding. We find substantial post-flood migration, with pronounced heterogeneity: private renters are significantly more likely to move than homeowners or social housing tenants, and migration responses are weaker among individuals in the bottom and top income quintiles relative to those in the middle.

Exploring potential mechanisms, we find no significant effects on individual income or local housing prices. Instead, we observe an increase in housing transactions outside flood-prone areas, consistent with flooding operating as a disamenity or triggering an updating of perceived flood risk. At the municipality level, we find no average treatment effect on population inflows or outflows, highlighting the importance of granular, micro-level data. Nevertheless, flooding alters the composition of population flows: outflows from flooded municipalities contain a lower share of individuals from the bottom quartile of the national income distribution, pointing to distributional consequences of flood-induced mobility.

Many cities face a double threat: a lack of affordability and climate change. In this paper, we study how rent control, a policy that aims to make housing more affordable, interacts with climate risks. We show that rent control can induce a lock-in effect that increases exposure to these risks. In particular, renters may move less due to rent control and consequently respond less to energy and heat stress shocks. We document these effects using a dataset of individual apartments from a real estate company in Switzerland, a country with low homeownership rates and a federal rent control policy.

Mandating flood insurance at subsidized prices is increasingly advocated as a solution to the persistent and growing flood insurance protection gap, yet its effectiveness relative to existing schemes remains underexplored. This paper provides the first empirical evaluation of such a policy on construction and house prices, and compares it with alternative systems. I study the case of France where flood insurance is mandatory and premiums are independent of exposure to risk. Using fine-grained dwelling-level data from France and Belgium, used as a control group, I estimate the impact of mandated insurance pooling on construction and property prices. I find large behavioral responses, but a relatively small effect of 3% on total flooding costs between 1981 and 2020. I then simulate the implementation of mandatory insurance with risk-reflective pricing, which raises dramatically living costs in floodplains. Trading-off between redistributive preferences and housing supply responses, I determine the optimal level of risk-pricing using a location choice model combined with a social insurance framework. I also simulate how integrating land-use regulations and taxes on new constructions in flood-prone areas could enhance welfare. Finally, augmenting the model with an insurance demand component, I compare the welfare effects of three policy scenarios: no government intervention, partial risk-sharing (similar to the US system), and an insurance mandate with optimal transfers. This study offers policymakers new empirical evidence and counterfactual policy options for improving flood insurance coverage.

Extreme climate events are more likely, more intense and require adaptation. We examine intended and realized migration behavior by exploiting a recent, devastating and salient event in France as a natural experiment. Using administrative and tracked real estate platform data, we show that the wildfire triggered a substantial increase in housing supply, coupled with a temporary slowdown in demand intentions. This gap translates to no adjustments in residential mobility, and a reallocation of properties toward rental use. Transaction prices remain stable suggesting limited reassessment of long-term value. These patterns reveal that climate shocks can disrupt local housing dynamics and reconfigure markets while preventing intended adaptation behavior.