Does Good News Wash out the Bad? Belief Updating and Climate Risk Information

How to cite

Mulder, Philip; Conell-Price, Lynn. 2025. "Does Good News Wash out the Bad? Belief Updating and Climate Risk Information." AEA RCT Registry. AEARCTR-0015373. https://doi.org/10.1257/rct.15373-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Philip Mulder* , Lynn Conell-Price

* Primary investigator

Registry ID
AEARCTR-0015373
DOI
https://doi.org/10.1257/rct.15373-1.0
First registered
Last updated
Status
Completed
Trial dates
2021-03-25 – 2021-05-09
Countries
United States of America ()
JEL codes
Q54, G52, C93, D83
Registry record
http://www.socialscienceregistry.org/trials/15373

Abstract

While property owners around the world have access to an increasing array of information about climate risks, it is unclear how such information affects their beliefs and participation in disaster insurance markets. This paper shows results from the first field experiment providing detailed, property-level flood risk information to U.S. homeowners. Our experimental design compares changes in risk beliefs and implied insurance demand between a treatment group who received individualized flood risk information from the First Street Foundation Flood Model, and control arms who received no individualized information. Our results show that while homeowners who overestimated their risk relative to the model revised their beliefs downward in response to information, homeowners who underestimated their risk showed little updating. Although underestimating participants are more confident about their pre-treatment risk beliefs, we do not find evidence that these results are driven by differences in private information or adaptation. These findings imply economically significant changes in flood insurance demand, reducing predicted voluntary take-up by more than five percentage points among homeowners at lower risk and increasing adverse selection.

Experimental design

After completing the initial survey section, participants proceeded to a section implementing the information provision experiment by randomly assigning participants to one of three variants of the survey flow. The first, "Control - No Information" condition skipped this block entirely and proceeded to follow-up questions. The second, “Control – Generic Information” treatment, did not provide any participant-specific risk information and simply displayed a headline reading “Your Flood Risk Summary” above an accurate but generic message indicating that “your property has a chance of flooding each year” and that any such flooding "can cause significant damage". Finally, the third, “Individualized Risk Information” Treatment displayed the same headline as the Control - Generic Information arm along with a detailed flood risk assessment specific to the respondent’s property. This information treatment directly embedded and displayed the respondent’s property-specific FloodFactor Flood Risk Assessment based on the First Street Foundation Flood Model (FSF-FM) and made freely available on the FSF website at the time. To ensure engagement with the additional information, this screen also prevented respondents from proceeding to the next page for at least 20 seconds.

Design details

Randomization method
Random number generator in Qualtrics
Randomization unit
Individual
Clusters
884 Individuals
Observations
884 Individuals
Arms
227 - no information control group 243 - generic information control group 414 - individualized information treatment group
Primary outcomes
Flood risk beliefs across six measures, voluntary flood insurance demand

Cited by

No citation data available for this registration yet.

Keywords