I am a PhD candidate at the Hertie School and a Research Associate at the Hertie School Centre for Sustainability.
From January to April 2025 I was a Visiting Research Student at the Grantham Research Centre on Climate Change and the Environment at
the London School of Economics and Political Science. I hold a First Class BA
in History and English from the University of Oxford and a Master of Public Policy with Distinction from the Hertie School.
My research examines the politics of climate change in high-income democracies. I am focused on the political barriers to (and enablers of) ambitious climate policy action, including the role of party politics, institutions, compensation, and public opinion.
I employ both quantitative and qualitative empirical methods.
- Introducing and applying the Climate Institutions Analysis Framework. A comparative analysis of climate institutions in Germany, the United Kingdom, Sweden, and Australia using the CIAF
With Jacob Edenhofer and Christian Flachsland
OSF Link
Abstract
Climate institutions – including framework legislation and advisory bodies – have proliferated globally as part of countries’ responses to climate change. They have received growing attention amid a broader 'institutional turn’ in
the study of climate politics. We lack conceptual tools, however, to analyse these meso-level institutions and disentangle the mechanisms driving their effects on climate policymaking. To fill this gap, we develop the
Climate Institutions Analysis Framework (CIAF) and use it to qualitatively analyse climate institutions in Germany, the United Kingdom, Sweden, and Australia. We demonstrate how cross-country variation in effects stems
from interactions between institutions and ’intervening variables’ specific to countries’ contexts. Across our sample, most institutions address agenda-setting and knowledge-related strategic challenges; few deliver ex-ante
accountability and none compensation. Our framework is relevant for comparative political economy analyses of meso-level political institutions, and, alongside our results, offers lessons for policymakers seeking to create or
improve climate institutions.
-
When do green parties matter? Green party bargaining leverage and climate policy stringency
With Christian Flachsland and Mark Kayser
Draft available upon request
Abstract
Political parties are often overlooked in studies of net zero transitions. Despite our growing understanding of the influence of public opinion, interest groups, and institutions, we know little about the role of parties in
determining climate policy. This paper addresses a specific puzzle: when do Green Parties (GPs) deliver stringent climate reforms? Theoretically, the success of ‘green’ special interest parties should lead to better climate policy
outcomes. Yet evidence of the effect of GP representation in parliament and government is mixed. We argue coalition bargaining leverage — the capacity for a GP to demand its preferred outcome in negotiations based on an expanded set
of coalition options — is a missing link in the relationship between GPs and policy change. Building on a novel, empirical measure of leverage developed by Kayser et al. (2023), we test whether multiparty
systems are more likely to implement (more) stringent climate policies when GPs increase their leverage. We measure changes in climate policy stringency using a new harmonised cross-country database from the OECD. Our results show GP leverage is associated with greater climate policy stringency and has a stronger effect than a GP being in government or increasing its support in the polls. We only identify this pattern, however, in the electricity sector. GP leverage appears to have had little effect (so far) on policies in the industrial, buildings, and transport sectors. Our findings motivate theoretical and empirical study of the political economies of different sectors of climate policy.
- The electoral consequences of climate ambition in fossil fuel communities
Draft available upon request
Abstract
Government policies to mitigate climate change threaten communities reliant on the fossil fuel industry for their livelihoods and can provoke electoral backlash. But how do voters in these communities respond prospectively to climate policies, the effects of which will only materialise in the future? Building on studies which highlight the role of uncertainty in shaping climate policy preferences, I argue that uncertainty about the nature and consequences of climate reforms leads voters in fossil fuel communities to punish the party proposing them. I study this question in the context of Australia, a country characterised by regional variation in its domestic and export-oriented fossil fuel industries. I leverage the 2019 federal election -- in which climate change rose to record levels of salience and split the major parties -- as a treatment event, to estimate a difference-in-differences. Assembling a novel panel of vote shares at 4,125 polling stations across Australia, I find voters in fossil fuel communities rejected the pro-climate party, before any climate policies had been implemented. To highlight the role of uncertainty, I compare how voters responded in communities that differed in their anticipated exposure to climate policies and the degree of uncertainty in the 2019 election. I find that effects are stronger in export-oriented communities, which were not exposed to domestic climate policies, but nevertheless faced substantial uncertainty in the 2019 election campaign. These results provide new evidence of prospective voting on climate policy and offer insights for parties developing electoral strategies for climate change mitigation.
-
Economic booms and the electoral success of green parties: cross border evidence from Switzerland's neighbours
With Tom Arend and Mark Kayser
Draft available upon request
Abstract
Scholars have long known that the the economy influences the electoral success of incumbent parties. We argue here that the business cycle also influences the support for a specific type of party -- Green parties -- independent of incumbent status. When Switzerland liberalized its labor market in 2002, permitting EU citizens to work in the country, they induced an exogenous income shock in the bordering regions of the neighboring countries. We leverage this natural experiment to test for causal income effects on green party vote share with difference-in-differences models in municipalities near the Swiss border. Individual-level DiD estimates in France confirm a mechanism, showing increased income and support for environmental movements in border areas. Support for green parties has long been stronger among the affluent. We demonstrate a different phenomenon: that short-run income shocks increase support for the Greens, the party family that is most strongly associated with non-material policies.
-
How, or how much? Preferences for institutional design for compensation. (Pre Analysis Plan)
With Jacob Edenhofer
Draft available upon request
Abstract
Ambitious climate policy often imposes concentrated, short‐term costs on particular constituencies—workers and firms in carbon‐intensive sectors, households facing higher energy bills, and communities dependent on fossil‐fuel industries. To mitigate political backlash, governments offer climate compensation (or “just transition”) packages. Yet empirical studies reveal mixed evidence on compensation’s ability to build or sustain public support. We argue that this inconsistency stems from three distinct sources of uncertainty: (1) time‐inconsistency (will future governments honour promises?), (2) delivery risk (will funds actually reach intended beneficiaries?), and (3) representation gaps (do individuals trust that their interests are represented in allocation decisions?). Our theoretical framework lays out, in narrative form, how each uncertainty channel affects the compensation threshold a voter requires, and how different institutional features—legislative lock‐ins, central versus regional delivery, local advisory panels—trade off these uncertainties. We then outline a rating‐based conjoint experiment that randomises within each profile: commitment mechanism, delivery agent, representation mode, and compensation amount. Finally, we specify our preregistered analyses, including tests of how perceived representation gaps moderate preferences over commitment features.
