Are Politicians More Likely to Back Climate Policy After Disasters?

Gagliarducci, Paserman, and Patacchini (2025) examine whether hurricanes influence legislators’ support for climate-related environmental policy. They analyze data on federal disaster declarations (1989–2020) matched to U.S. House districts, combined with records of sponsorship and cosponsorship of “green” bills. They find that representatives from hurricane-hit districts are more likely to support environmental legislation, especially in the year after a storm. The effect is concentrated among electorally secure members. The increase amounts to roughly 0.17–0.22 additional green bills per member, a rise of more than 15% relative to the mean. The response fades after about two years, suggesting a salience-driven mechanism.

Why This Article Was Selected for The Policy Scientist

This study addresses a policy issue of sustained global significance: whether extreme climate-related events alter legislative behavior. The topic extends beyond hurricanes or environmental bills, speaking to how democratic institutions process long-term risks that carry short-term political costs. The article is timely given rising disaster frequency, intensifying climate policy debates, and increasing scrutiny of electoral incentives. The authors have an established research agenda on political economy and legislative behavior, and this paper fits coherently within that body of work. The dataset is unusually strong, combining three decades of FEMA disaster declarations with detailed congressional sponsorship records. The design leverages plausibly exogenous storm paths, representing a credible causal inference strategy. The statistical framework is rigorous, and the findings are potentially generalizable to other disaster-prone democracies.

Full Citation and Link to Article

Gagliarducci, S., Paserman, M. D., & Patacchini, E. (Forthcoming 2025). Hurricanes, climate change policies and electoral accountability. American Economic Journal: Economic Policy. https://doi.org/10.1257/pol.20230070 

Central Research Question

This article investigates whether exposure to hurricanes alters the policy behavior of elected officials, specifically their support for climate change–related environmental legislation. The authors ask whether representatives whose congressional districts are directly affected by hurricanes become more likely to sponsor or cosponsor “green” bills, and whether this response varies systematically with electoral incentives. The question is situated within a broader theoretical tension in political economy: legislators must balance re-election motives against the pursuit of policies that generate long-term benefits but impose short-term political or economic costs. Climate policy represents a particularly suitable domain for this inquiry because mitigation policies often produce diffuse, delayed benefits while concentrating adjustment costs in the present. The study therefore evaluates whether extreme weather events function as exogenous shocks that increase issue salience and induce measurable shifts in legislative agendas.

Previous Literature

The study builds on three intersecting strands of research. First, a large body of work examines how natural disasters influence political outcomes, including voter behavior, incumbent punishment, and government responsiveness. Prior studies show that severe weather events can affect electoral support and that politicians may be rewarded or penalized depending on relief spending and disaster response. However, much of this literature focuses on short-term distributive policies, such as emergency aid or reconstruction funding, rather than long-term regulatory policy.

Second, research on climate change and political behavior suggests that unusual local weather conditions can increase public concern about climate risks and modestly influence legislative voting patterns. Studies using temperature anomalies or disaster fatalities indicate that environmental attitudes and roll-call votes may respond to weather-related salience. Yet roll-call analyses are limited by the fact that relatively few climate-related bills reach the floor, raising concerns about selection effects and party discipline.

Third, theoretical models of representative democracy emphasize the role of electoral incentives in shaping policy choices. Downsian frameworks predict that politicians closely track voter preferences, while ideological models stress stable policy commitments. This article contributes by examining a setting where politicians face incentives to adopt forward-looking policies that may be unpopular in the short term. By focusing on sponsorship and cosponsorship activity rather than floor voting, the study offers a complementary measure of policy engagement less constrained by leadership agenda control.

Data

The authors assemble a comprehensive dataset linking meteorological shocks to legislative behavior over three decades. Hurricane exposure is measured using the universe of federal disaster declarations from the Federal Emergency Management Agency (FEMA) covering 1989–2020. Counties listed in FEMA hurricane-related declarations are classified as “hit,” and these designations are mapped to congressional districts using Census Bureau district–county relationship files. This procedure produces district-level indicators of hurricane incidence, including contemporaneous and lagged measures.

Legislative activity is drawn from Library of Congress records for the U.S. House of Representatives spanning the 101st through 116th Congresses. The authors identify climate-related “green” bills using Congressional Bills Project classifications, restricting attention to legislation addressing air pollution, global warming, and alternative or renewable energy. Manual text analysis excludes relief bills, noise pollution legislation, and anti-environmental proposals. The final bill sample includes 1,056 green bills.

The dataset is enriched with electoral, demographic, and economic controls. These include prior election margins, district partisanship, unemployment rates from the Bureau of Labor Statistics, state-level fossil fuel production, demographic characteristics from the Decennial Census, committee memberships, tenure, age, gender, and environmental voting records from the League of Conservation Voters (LCV). The resulting panel contains 13,685 district-year observations representing 1,574 members of Congress and 1,732 district-decade combinations.

Methods

The empirical strategy exploits the plausibly exogenous timing and trajectory of hurricanes. Because hurricane paths are largely determined by atmospheric dynamics rather than local political conditions, within-district variation over time provides a credible basis for causal inference. The primary estimating equation regresses the number of green bills sponsored or cosponsored by a legislator on indicators of hurricane exposure, controlling for district characteristics, legislator attributes, and fixed effects.

The preferred specifications include year fixed effects and either district or legislator fixed effects, isolating deviations from a district’s or member’s typical legislative behavior. Additional robustness checks incorporate state-specific trends and alternative clustering methods. Placebo tests examine whether future hurricane exposure predicts current legislative activity, providing a falsification exercise against spurious correlations.

To address recent concerns about two-way fixed effects models under staggered treatment adoption, the authors implement a stacked difference-in-differences event-study design. This approach avoids problematic comparisons between early- and late-treated units and allows estimation of dynamic treatment effects. The study also conducts heterogeneity analyses by electoral safety, party affiliation, tenure, district unemployment, environmental record, coastal exposure, and post-2007 awareness following the IPCC Fourth Assessment Report.

Finally, the authors perform a supervised text analysis of congressional speeches. A manually labeled training set identifies language referencing short-run economic or political costs. A dictionary-based classifier is then applied to environmental speeches, enabling measurement of whether legislators who promote green bills explicitly acknowledge near-term costs.

Findings/Size Effects

The results show that legislators from hurricane-affected districts increase their support for green legislation, particularly in the year following a storm. The estimated effect ranges from approximately 0.17 to 0.22 additional green bills sponsored or cosponsored per member. Relative to the sample mean of about 1.05 green bills, this corresponds to an increase exceeding 15 percent. Contemporaneous effects are smaller, consistent with hurricanes typically occurring later in the calendar year.

The response is highly heterogeneous. The increase in green legislative activity is concentrated among representatives in electorally safe districts. Legislators facing competitive re-election contests exhibit muted or statistically insignificant reactions. Similarly, longer-serving members—another proxy for electoral security—display stronger responses. Party differences are pronounced: Democrats account for most of the observed shift, while Republicans show little systematic change.

Temporal dynamics indicate that the effect emerges immediately after hurricane exposure and dissipates within roughly two years. This pattern aligns with an issue-salience mechanism rather than a permanent updating of policy preferences. Geographic spillovers are limited; adjacent districts or other districts within the same state do not exhibit comparable increases.

The study also documents agenda reallocation. Increased sponsorship of green bills is accompanied by a decline in non-environmental legislative activity, suggesting displacement rather than simple expansion. Hurricanes predict higher numbers of relief bills, as expected. Importantly, the surge in green sponsorship translates into modest increases in enacted legislation, indicating that the response is not purely symbolic.

The speech analysis reveals that legislators who sponsor or cosponsor green bills are more likely to reference short-run costs in environmental speeches. This suggests awareness of potential economic or political trade-offs rather than naïve optimism about policy benefits.

Conclusion

The article concludes that extreme weather events influence legislative behavior, but only under specific political conditions. Hurricanes function as salient shocks that temporarily increase attention to climate risks. However, electoral incentives constrain policymakers’ willingness to support potentially unpopular long-term regulation. Legislators insulated from electoral pressure appear more willing to adjust their agendas toward climate mitigation.

These findings contribute to understanding how democratic institutions respond to long-horizon challenges. The evidence suggests that political reactions to climate-related disasters are real but conditional, shaped by both issue salience and re-election constraints. The study highlights the importance of electoral security, party alignment, and temporal attention cycles in determining whether environmental shocks translate into legislative change.