Local Adaptation to Climate Risks in the Rental Housing Market

How to cite

Houde, Sebastien; Fanghella, Valeria. 2026. "Local Adaptation to Climate Risks in the Rental Housing Market." AEA RCT Registry. AEARCTR-0019106. https://doi.org/10.1257/rct.19106-1.0

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

Sebastien Houde* , Valeria Fanghella

* Primary investigator

Registry ID
AEARCTR-0019106
DOI
https://doi.org/10.1257/rct.19106-1.0
First registered
Last updated
Status
In development
Trial dates
2026-07-09 – 2028-09-30
Countries
Private
JEL codes
Q54, Q51, R31, R23, C93
Registry record
http://www.socialscienceregistry.org/trials/19106

Abstract

Where households choose to live is central to how they adapt to climate risks, yet little is known about adaptation in the rental market, where policy must work through regulation, prices, and information. We study heat-stress adaptation among renters in Paris—a city with pronounced urban-heat-island exposure and neighborhood-based rent control—through a framed field experiment. Renters who intend to move complete a two-part study: a neighborhood-choice module and a real apartment search on a live listing portal, tracked via a browser extension. Within subjects, we deliver information treatments about heat exposure (projected tropical nights) and rent-control ceilings at two levels of aggregation, the neighborhood and the apartment, and elicit incentivized moving intentions before and after each treatment. Between subjects, we randomize incentivized versus hypothetical elicitation, the on-listing salience of the information, and the size of a moving-cost voucher. Using this experimental variation, we estimate a dynamic neighborhood–apartment choice model and recover the willingness to pay to avoid heat at the neighborhood versus apartment level, the disutility of rent, and moving costs. The estimates let us simulate counterfactual information, rent-control, and place- versus building-based adaptation policies, and their welfare consequences.

Intervention

Participants are renters who intend to move in the Paris region. In an online study, they report where they would consider moving and how likely they are to move to each option, and they complete a real apartment search on a live listing portal. The intervention consists of information provided during the study: participants are shown objective, science-based information about two neighborhood and apartment attributes—exposure to heat (projected number of "tropical nights") and the applicable rent-control ceiling. This information is delivered at two levels of aggregation, the neighborhood and the individual apartment, and participants report their moving intentions before and after receiving it. Additional randomized elements govern whether the intention questions carry a monetary incentive for accuracy, whether the heat and rent information is displayed directly on listings during the apartment search, and the size of a hypothetical moving-cost voucher.

Experimental design

The study uses a within-subject design in two parts. In the first part, participants consider neighborhoods in Paris and report how likely they are to move to each; in the second part, they conduct a real apartment search online and report their intentions over specific apartments. In both parts, participants report their intentions before and after receiving information about local attributes, so that each participant serves as their own comparison. Across participants, we additionally vary a small number of features of the task—including whether the intention questions carry an accuracy-based reward, how the information is presented during the apartment search, and the size of a moving-cost scenario. Moving intentions are the primary outcomes.

Design details

Randomization method
The randomization of the information treatment (salience or not at the begining of Part 2) will be done via Qualtrics. The winners of the rewards will be selected after the completion by all participants. The winners will be selected with a randomization done in office by a computer. The question that will be incentivized will also be done with a randomization done in office by a computer.
Randomization unit
Participant-level
Clusters
NA
Observations
1500 potential renters
Arms
We will randomize half the participants in the treatment arm with salience information at the beginning of Part 2 of the study.
Primary outcomes
Intention to move, which we elicit several times during the survey. We use those to conduct a reduced-form and structural estimation.
Secondary outcomes
Our design also allows us to estimate other structural parameters. Ultimately, our goal is to have a fully specified model, for which we estimate the moving costs and preference heterogeneity for apartment attributes. Our deisgn also allows to estimate heterogeneous parameters, which we will investigate across demographics.

Cited by

No citation data available for this registration yet.

Keywords