EPG Seminar Series

Environmental Politics and Governance (EPG) Online seminar series at the Department of Social Policy, LSE.

We will being holding in-person viewings at LSE, with some presentations also happening in-person. If you would like to attend these, please indicate interest here.

For any questions related to the seminar series, please email the academic organiser Dr Liam Beiser-McGrath.

2025/26 Seminar Series Schedule

20 November, 4.30pm-5.30pm, OLD 2.21

Technocratic or Political? The Effect of Partisanship on Drought Responses in France
Rens Chazottes with Nina Lopez-Uroz (EPFL)

The increasing frequency of natural disasters due to climate change has intensified pressures on societal well-being. In such times, understanding the institutional features that enable efficient, objective, and neutral disaster recovery is crucial. Recent studies have highlighted the severity of government oversight in disaster relief, often favoring co-partisan groups in developing and clientelistic countries. Disaster recovery systems are particularly vulnerable to the politics of post-disaster fund allocation. However, scholars have suggested that institutional design can counteract these dynamics, with France's mandatory disaster insurance system frequently cited as a model. In this study, we assess the extent to which France's mandatory disaster insurance system has been manipulated for electoral gain during presidential and municipal elections. Utilizing data from the CatNat national repository and municipal elections from 1980 to 2024, we employ a regression discontinuity design to examine how partisanship alignment between local and national governments affects both the demand and the supply side of disaster recognition and the response time. Our preliminary findings indicate that partisan alignment correlates with a higher demand for disaster recognition. However, the French institutional system appears effective in mitigating political distortions, as we find no significant evidence of partisanship influencing disaster relief. This article sheds light on the effectiveness of institutional design in reducing political distortions during the disaster recovery phase.

Climate Change and Electoral Behaviour: Evidence from Extreme Weather Events in Costa Rica
Alvaro Zuniga-Cordero (University of Namur & World Inequality Lab)

This paper examines the political implications of climate change by analysing how extreme weather events affect electoral outcomes in Costa Rica. Leveraging rich administrative data—including electoral results, individual-level turnout records, employer-employee information, disaster inventories, and geo-localised weather registries—the study integrates high-resolution environmental and electoral data to examine one central question: How do extreme weather events affect voting behaviour? The analysis compares regions exposed to climate shocks over time using a difference-in-differences framework with staggered treatment and event-study designs. Preliminary findings suggest that direct experiences with climate change shape voter attitudes and participation, with responses varying by socioeconomic status and political predispositions. We observe that in progressive-voting areas, extreme weather increases support for environmental policies, while in conservative regions, such events boost backing for populist alternatives. To explain this divergence, the study examines evangelical churches' timing and spatial distribution, revealing that extreme weather strengthens existing evangelical networks, mobilising electoral support for evangelical populist candidates.

4 December, 4.30pm-5.30pm, OLD 2.21

Performing Decarbonization
Anthony Calacino (University of Oxford)

Fossil fuel firms as climate-forcing asset holders face an existential contradiction: their profitability mostly relies on continued hydrocarbon production, yet, increasingly, they are under pressures to decarbonize and go green. Existing accounts often portray corporate greenwashing as reputation management aimed at consumers and investors. This paper argues that climate-forcing firms also deploy environmental messaging as a political strategy to shape the media and policy arenas. Using an original dataset of nearly 5,000 YouTube videos released by fourteen of the world's largest oil companies between 2008 and 2025, we apply text-as-data methods to examine when and why oil firms emphasize environmental messaging. We explore such messaging at key political junctures. The first juncture is when national climate mitigation legislation is under debate, consistent with the view that firms seek to preempt more ambitious mitigation policies. Second, messaging changes when ambitious climate candidates are competitive in national elections. The third juncture is in the lead-up to UNFCCC climate summits, which reflects the growing influence of climate-forcing asset holders at these global meetings. Moreover, we find different rates of environmental messaging between Global North and South firms, as well as between private, traded and state-owned. These findings reposition greenwashing as a form of political influence rather than only consumer persuasion, contributing to theories of corporate power, disinformation, and the political economy of climate change.

When do Congressmembers Break the Party Line on Climate Policy? The Significant Impact of Employment in High-Emitting Industries
Peter Wyckoff (LSE)

Climate policy is a highly polarized issue in the United States, yet Congressmembers repeatedly deviate from their party's position on votes on climate bills. Between 2009 and 2022, 234 House members and 65 Senators voted at least twice against their party's position on climate legislation. This paper argues that small but distinctly uneven variation in employment shares exposed to climate policy across districts can help explain these deviations. Using industry emission intensity rates and granular employment data, this paper finds that a 4-percentage point difference in exposed employment share roughly doubles the likelihood a Democrat will deviate from their party line, and halves the likelihood for a Republican. The paper presents two potential mechanisms for the outsized impact of these small employment shares: employment multipliers, and local industrial undiversification.