Natural Disasters and Insurance
Paper Session
Sunday, Jan. 3, 2027 8:00 AM - 10:00 AM (EST)
- Chair: Joakim Weill, Federal Reserve Board
Distributional impacts of updating flood risk information in US housing markets
Abstract
Information about residential properties’ exposure to flood risk has historically been underprovided and difficult to access, resulting in uninformed purchasing decisions. Improving provision of this information may lead to efficiency gains, but could also have heterogeneous distributional effects. We evaluate the distribution of price capitalization effects associated with flood risk informational updates based on the sociodemographic characteristics of incumbent homeowners. We also estimate the effects of informational updates on residential sorting over flood risk.Our informational treatments are based on staggered revisions to FEMA’s Flood Insurance Rate Maps and the implementation of state-level laws requiring the disclosure of properties’ exposure to flood risk during real estate transactions. We identify treatment effects using a nationwide repeat sales model, comparing sale prices and homebuyer sociodemographic characteristics for individual property transactions before and after they are subject to these updates.
We find that the negative capitalization effects of remapping properties into federally designated flood zones and flood risk disclosure requirements have regressive distributional impacts among incumbent homeowners. However, following these information updates, new homebuyers have higher income and are less likely to be Black or Hispanic. These changes in residential sorting could be driven by informational asymmetries among prospective homebuyers prior to treatment or adoption of different mortgage lending practices associated with a property’s flood zone status.
Our results demonstrate that improving access to information through map updates and disclosure requirements are viable policy tools for helping to alleviate disparities in flood risk exposure across race and income groups; however, the repricing of exposure to flood risk is expected to disproportionately harm lower-income households.
Climate Risk and Insurer Adaptation: Evidence from the 2017–2018 California Wildfires
Abstract
Climate change poses growing challenges to the stability of insurance markets. This paper examines whether property insurers are adapting to climate risk by changing their underwriting behavior in response to large, unexpected losses, and the broader implications of insurer climate risk management on the functioning of insurance markets. Using firm-specific catastrophic losses from the 2017–2018 California wildfire seasons as a one-time shock, we find that insurers who suffered greater wildfire losses reduced their underwriting in high fire hazard areas within California, with stronger retreat among those with higher self-disclosed quality in climate risk management. These insurers also curtailed underwriting in other high fire hazard states and in hurricane-prone areas in Florida, suggesting both geographic and cross-peril spillovers. The selective retreat of climate-conscious insurers from high-risk areas raises concerns about the concentration of solvency risk and coverage reliability in these regions. Meanwhile, these firms tend to offer larger premium discounts and provide stronger incentives for hazard mitigation, highlighting policy opportunities to enhance both insurance supply and climate adaptation.The Value of Public Disaster Insurance
Abstract
Disaster insurance markets face a central tension: catastrophic losses are rare, but when they occur they are spatially correlated. Insurers exposed to geographically concentrated risks must hold costly capital reserves to remain solvent, raising premiums and reducing coverage.This paper develops a tractable model to quantify the value of spatial risk pooling in disaster insurance markets. The model features risk-averse households, demand frictions, heterogeneous regional disaster risk, and insurers that face costly capital requirements tied to aggregate losses in their portfolios. We compare welfare under two market structures: a single national insurer that pools losses across regions, and decentralized regional insurers that bear local catastrophic risk separately.
The model quantifies how national pooling reduces required reserves, lowers the cost of capital embedded in premiums, and increases insurance take-up. These gains are largest when risks are highly correlated within regions but weakly correlated across regions. We bring the model to data using granular measures of property-level flood risk, flood insurance take-up, and historical flood losses to quantify the value of the National Flood Insurance Program (NFIP), the only federal disaster insurance program in the United States. We estimate the welfare consequences of replacing the current national program with state-level insurance systems and show that, for a range of plausible parameter values, the cost of insuring catastrophic risk locally is high enough for insurance markets to fail to exist in many states.
These results highlight a fundamental benefit of federal disaster insurance. While prior work has emphasized the costs and implementation challenges of the NFIP (including moral hazard, adverse selection, and outdated risk models), we provide a tractable framework for quantifying its national risk-pooling benefits.
Discussant(s)
Jeffrey Shrader
,
Columbia University
Laura Bakkensen
,
University of Oregon
Philip Mulder
,
University of Wisconsin-Madison
Benjamin Collier
,
University of Wisconsin-Madison
JEL Classifications
- Q5 - Environmental Economics
- G5 - Household Finance