Behavioral Economic Analysis of Framing Effects on Residential Safe Room Demand

Abstract

Tornadoes cause widespread financial and infrastructural damages and pose a hazard to human well-being. Much of the US Midwest and Southeast endure tornadic events and potentially tornadic storms annually, yet in-house safe rooms and publicly accessible shelters remain quite limited creating avoidable risk. The purpose of this study was to use a simulated operant behavioral economic framework to model how reimbursement and warning-text language presented in a community-wide FEMA hazard mitigation grant influence participants’ willingness to purchase a residential safe room. Participants indicated willingness to purchase a hypothetical safe room as a function of increasing per unit prices with paired reimbursement. We examined framing effects in a 2×2 arrangement, where all recipients viewed reimbursement pricing for a safe room as a dollar amount or percentage of the overall unit costs. All participants had three discrete-choice options on the simulated purchase task. We also evaluated FEMA-recommended warning text through a between-subjects comparison. Application of field-standard demand modeling resulted in orderly operant demand curve measures for each group. Visual analysis and group-level demand metrics indicated that participant demand for a residential safe room was highest in the condition with no warning text and reimbursements framed as the dollar amount. Qualitative data (i.e., written feedback) generally supported this claim and suggested that many participants were only willing to purchase safe rooms when they were priced within a narrow range that notably did not include free price—a “too cheap, too expensive” effect. Data broadly suggest that baseline demand for residential safe rooms is low despite language manipulations. Informational campaigns and outreach efforts should be prioritized to increase baseline demand for residential safe rooms, particularly in high-risk locations.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 0 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.