Hanno Hilbig

I am an Assistant Professor of Political Science at the University of California, Davis. My research asks how voters and governments respond to economic change and climate change exposure, with a focus on state and local policy. My work covers three areas: housing and local budgets, the energy transition and climate adaptation, as well as place-based and industrial policy in France, Germany, and the United States. I often use causal research designs that involve administrative and electoral data, as well as text data, much of which I have built and made public.

Previously, I was a Postdoctoral Fellow at the Center for the Study of Democratic Politics at Princeton University. I graduated with a PhD from the Department of Government at Harvard University in 2022.

Publications

1. Biased Party Nominations as a Source of Women's Electoral Underperformance (with Pia Raffler and Thomas Fujiwara). Forthcoming, American Journal of Political Science. [Abstract]
   What accounts for differences in electoral success between male and female candidates? We argue that parties systematically nominate female candidates to districts where the party is less popular, making it harder for women to get elected. Our empirical strategy uses the German mixed electoral system to create counterfactual gender vote gaps. These gaps represent the scenario where male and female candidates are nominated in districts where their respective parties have equal popularity. Using data on all candidates for the German Bundestag across eleven elections, we document that female underperformance, and its variation across parties and election years, is explained almost entirely by women running in districts where their party is less popular. In contrast, we find no evidence that voter bias or candidate characteristics play a substantial role. Our argument highlights gendered party gatekeeping that increases in district strength as an important driver of female underrepresentation.
   
   • [Media: NPR]
2. How Budget Tradeoffs Undermine Electoral Incentives to Build Public Housing (with Andreas Wiedemann). 2026. American Journal of Political Science, 70 (1): 202-220. [Abstract] [Preprint]
   Housing shortages and rising rents have increased demands for affordable housing. In this paper, we examine whether electoral constraints can undermine local politicians' incentives to build public housing. Empirically, we draw on the full-count census of all housing built in Germany, data on 19,685 local elections between 1989 and 2011, and an original survey. Using a difference-in-differences design, we demonstrate that incumbents are not rewarded, but rather experience moderate electoral losses after constructing new public housing. We then show that these losses are not primarily driven by homeowner opposition or native–foreigner competition. Instead, electoral punishment is largest in economically disadvantaged municipalities with relatively affordable housing, as voters prioritize spending in other local policy areas that are crowded out by public housing. Survey evidence demonstrates that electoral constraints emerge when voters' short-term spending preferences conflict with municipalities' long-term goals to provide affordable housing.
3. Does Rent Control Turn Tenants Into NIMBYs? (with Anselm Hager and Robert Vief). 2026. Journal of Politics, online ahead of print. [Abstract] [Preprint]
   Affordable housing is a key challenge of the 21st century. A pivotal driver of growing housing prices is residents' opposition to construction, a phenomenon known as NIMBYism ("Not In My Backyard"). To make housing more affordable, city governments are increasingly implementing rent control policies. Does rent control—by making tenants more likely to stay in their apartments—spark NIMBYism and thus exacerbate the housing crisis? We study the case of Berlin, which recently passed a sweeping rent control law. Leveraging two discontinuities in the policy, we show that rent control made tenants less NIMBY. Specifically, tenants in rent-controlled apartments became more likely to approve of local-level construction and immigration, compared to tenants in non-rent-controlled apartments. We argue that the decline in NIMBYism is likely due to an economic channel. Tenants in urban centers associate construction and immigration with displacement pressures and gentrification. Rent control alleviates these concerns by providing financial and residential security.
   
   • [Media: The Atlantic]
4. The Green Transition and Political Polarization Along Occupational Lines (with Vincent Heddesheimer and Erik Voeten). 2026. American Political Science Review, 120 (3): 853–875. [Abstract] [Preprint]
   Green transition policies set long-term targets to reduce carbon emissions and other pollutants, posing a threat to workers in polluting occupations and communities reliant on these occupations. Can far-right parties attract voters who anticipate losing from the green transition? We explore this in Germany, which has ambitious green policies and a large workforce in polluting occupations. The far-right AfD started campaigning as the only party opposing green transition policies in 2016. Using a difference-in-differences design, we show AfD support increased in counties with more polluting jobs after this platform change. A panel survey further demonstrates that individuals in these occupations also shifted towards the AfD. Probing mechanisms, we find suggestive evidence that growing far-right support is due to changing perceptions of social stigma and lower status. Our results highlight the need for a new research agenda on a backlash against the normative dimension of the green transition.
5. Do Autocrats Respond to Citizen Demands? Petitions and Housing Construction in the GDR (with Hans Lueders and Sascha Riaz). 2026. Comparative Political Studies, 59 (4): 703-738. [Abstract] [Preprint]
   Citizens in authoritarian regimes frequently communicate grievances to the government. While there is some evidence that governments respond to such petitions, little is known about the nature of this responsiveness: can petitions yield tangible improvements to citizens' livelihoods? To answer this question, we assemble a novel panel of housing-related petitions to the government of the German Democratic Republic (GDR) and all housing constructed between 1945-1989. Exploiting the timing of the largest housing program in 1971, we employ a difference-in-differences design to show that construction was targeted at regions with higher rates of petitioning. We then use a variance decomposition method to benchmark the importance of petitions against objective indicators of housing need. Our results suggest that petitions allow citizens to meaningfully influence the allocation of public resources. The paper contributes to nascent scholarship on responsiveness in non-democratic regimes and shows that responsiveness leads to tangible improvements in citizens' livelihoods.
6. Estimating controlled direct effects with panel data: an application to reducing support for discriminatory policies (with Matthew Blackwell, Adam Glynn, and Connor Phillips). 2026. Journal of the Royal Statistical Society: Series A, 189 (2): 1070-1089. [Abstract]
   Recent experimental studies in the social sciences have demonstrated that perspective-taking conversations are effective at reducing prejudicial attitudes and support for discriminatory policies. We ask if such interventions can directly affect policy views without changing prejudice. Unfortunately, the identification of the controlled direct effect—the natural causal quantity of interest for this question—has required strong selection-on-observables assumptions for any mediator. We leverage a recent experimental study with multiple survey waves of follow-up to identify and estimate the controlled direct effect using the changes in the outcome and mediator over time assuming parallel trends in the potential outcomes. This design allows us to weaken the identification assumptions to allow for linear and time-constant unmeasured confounding between the mediator and the outcome. We develop a semiparametrically efficient and doubly robust estimator for these quantities along with a sensitivity analysis for the key identifying assumption of parallel trends. Contrary to what traditional methods find, our approach estimates a controlled direct effect of perspective-taking conversations when subjective feelings are neutral but not positive or negative, and this result is robust to moderate departures from parallel trends.
7. GERDA: The German Election Database (with Vincent Heddesheimer, Florian Sichart, and Andreas Wiedemann). 2025. Scientific Data, 12: 618. [Abstract] [Preprint]
   Elections are the key mechanism through which voters hold elected officials accountable. The partisan composition of local, state, and federal governments, in turn, shapes policy choices and public goods provision. Yet studying representation, government responsiveness, and partisan politics across multiple levels of government—especially at the local level—has been difficult due to inconsistently reported, incomplete, or insufficiently harmonized election data at small geographic scales. This paper introduces GERDA (https://www.german-elections.com/), a panel dataset of local, state, and federal election results in Germany at the municipality level spanning the past three decades. GERDA includes turnout and vote shares for all major parties and resolves challenges arising from municipal boundary changes and joint mail-in voting districts, yielding a consistent panel of municipalities in their 2021 boundaries. We also provide municipal and county boundary shapefiles to facilitate spatial analyses. Our dataset enables new research on partisan politics, policy responsiveness, and political representation at fine-grained geographic scales and over time.
8. The Gendered Persistence of Authoritarian Indoctrination (with Nourhan A. Elsayed, Sascha Riaz, and Daniel Ziblatt). 2025. British Journal of Political Science, 55: e124. [Abstract] [Preprint]
   A large literature has studied the effects of socialization under authoritarianism on political attitudes. In this research note, we extend this literature by demonstrating striking gender disparities in the post-transition persistence of these effects. We study the case of authoritarian indoctrination in the German Democratic Republic (GDR) using a regression kink design for causal identification. First, we draw on a unique survey fielded right before reunification to show that education under authoritarianism substantially reduced support for democratic capitalism and reunification with the West. In the second step, we triangulate multiple contemporary data sources to trace the persistence of these effects over time. Three decades after the fall of the GDR, the attitudinal effects of authoritarian indoctrination persist only among men, but not women. Our results highlight considerable heterogeneity in the persistence of authoritarian legacies, raising critical questions about post-authoritarian "re-socialization" and gendered adaptability.
9. Natural Disasters and Green Party Support (with Sascha Riaz). 2024. Journal of Politics, 86 (1): 241-256. [Abstract] [Preprint]
   A growing literature shows that extreme weather events induce pro-environment attitudes. We examine the political effects of a severe flood shortly before the 2021 German federal election. Drawing on about 600,000 survey responses and electoral data, we assess how flooding affected (i) the perceived salience of climate change, (ii) self-reported Green Party support, and (iii) Green Party voting in federal elections. We find that even severe local flooding had little to no effect on these outcomes. Additional evidence supports two mechanisms underlying this finding: nationwide rather than local effects of severe disasters, and voter demands for disaster relief rather than climate change prevention. We test the former mechanism using a regression discontinuity design and find that the flood increased nationwide Green Party support, although this effect persists for only two weeks. Our results shed new light on the precise duration and geographic scope of the political effects of natural disasters.
   
   • [Media: Bloomberg · Handelsblatt]
10. Government Spending and Voting Behavior (with Anselm Hager). 2024. World Politics, 76 (1): 88-124. [Abstract]
    Does government spending on public goods affect the vote choice of citizens? On the one hand, voters have been characterized as "fiscal conservatives" who may turn toward conservative parties when government spending goes up. On the other hand, increased spending may signal that the economy is doing well, which makes progressive parties a more viable option. To adjudicate between both hypotheses, this paper draws on a natural experiment, which created exogenous variation in government spending. A discontinuity in the 2011 German census meant that some municipalities saw an unforeseen increase in budgets. Using a regression discontinuity design, we show that the increase in budgets and subsequent spending on public goods benefited left-leaning parties, but had no effect on incumbent support. To parse out the causal channel, we rely on panel evidence and demonstrate that treated residents viewed their economic situation more favorably, which led them to espouse progressive parties.
11. Wealth of Tongues: Why Peripheral Regions Vote for the Radical Right in Germany (with Daniel Bischof and Daniel Ziblatt). 2024. American Political Science Review, 118 (3): 1480–1496. [Abstract] [Preprint]
    Why is support for the radical right higher in some geographic locations than others? This paper argues that what is frequently classified as the "rural" bases of radical right support in previous research is in part the result of something different: communities that were in the historical "periphery" in the center-periphery conflicts of modern nation-state formation. Inspired by a classic state-building literature that emphasizes the prevalence of a "wealth of tongues" (Weber 1976)—or nonstandard linguistic dialects in a region—as a definition of the periphery, we use data from more than 725,000 geo-coded responses in a linguistic survey in Germany to show that voters from historically peripheral geographic communities are more likely to vote for the radical right today.
12. Local Newspaper Decline and Political Polarization: Evidence from a Multi-Party Setting (with Fabio Ellger, Sascha Riaz, and Philipp Tillmann). 2024. British Journal of Political Science, 54 (4): 1256-1275. [Abstract] [Preprint]
    How does the decline of local news affect political polarization? We provide novel panel evidence on this question in a multi-party setting. In particular, we shed light on the link between newspaper exits, media consumption, and ultimately electoral behavior. To study the aggregate relationship between local news exits and polarization, we rely on a unique panel of all German local newspapers between 1980 and 2009. In addition, we precisely trace individual-level mechanisms by drawing on an annual media consumption survey of more than 670,000 respondents over three decades. Using a difference-in-differences design, we demonstrate that local newspaper exits increase electoral polarization, which aligns with evidence from the American context. Going beyond prior work, we then document that local news exits increase polarization because affected constituents substitute local news with national tabloid news. Finally, we show that local news exits increase politicization and partisanship at the individual level.
13. Refugee Labor Market Access Increases Support for Immigration (with Anselm Hager and Sascha Riaz). 2024. Comparative Political Studies, 57 (5): 749–777. [Abstract]
    Does the economic integration of refugees affect public attitudes toward migration? We assess this pertinent question by examining a policy change in Germany, where the government significantly eased labor market access for refugees in the majority of the country. Using administrative employment data, we show that the policy led to a substantial increase in refugee employment, while natives' wages and employment rates remained unaffected. The policy also had a positive effect on natives' attitudes toward migration. Voters exposed to more refugees in the labor market were two percentage points more likely to vote for pro-migration parties across both state and federal elections. Additional survey analyses suggest that our results are driven by positive native-refugee interactions in the workplace.
14. Local News Monopolies Increase Misperceptions about Immigration (with Sascha Riaz). 2023. Journal of Ethnic and Migration Studies, 49(17): 4536-4558. [Abstract] [Preprint]
    We examine how local news monopolies affect misperceptions about the size of the immigrant population in Germany. We propose a theoretical framework in which heterogeneous information from different local news outlets diffuses through social interactions. We posit that indirect exposure to information from multiple sources leads to more accurate beliefs in competitive markets. To causally identify the effect of local news monopolies on misperceptions, we exploit overlapping newspaper coverage areas as a source of exogenous variation in the number of available outlets. We estimate that local news monopolies increase misperceptions about the size of the local immigrant population by about four percentage points. We demonstrate that the effect of media monopolies hinges on social interactions. For individuals with fewer close social contacts, misperceptions remain unaffected by local news monopolies. Our results suggest that consolidation in the market for news decreases constituents' knowledge about critical policy issues.
15. Freedom of Movement Restrictions Inhibit the Psychological Integration of Refugees (with Sascha Riaz). 2022. Journal of Politics, 84(4): 2288-2293. [Abstract] [Preprint]
    How do freedom of movement restrictions affect refugee integration? While a growing body of research studies the initial spatial allocation of refugees, there is little causal evidence on subsequent policies that restrict residential mobility. We study a contentious law in Germany, which barred refugees from moving to a location different from the one they were randomly assigned to. To identify the causal effect of the movement restriction on integration, we utilize a sharp date cutoff that governs whether refugees are affected by the policy. We demonstrate that restricting freedom of movement had pronounced negative effects on refugees' sense of belonging in Germany while increasing identification with their home countries. In addition, the policy decreased engagement in a variety of social activities. Our findings suggest that discriminatory policies send a negative signal about the inclusiveness of the host society and thereby reduce the psychological integration of refugees.
16. Locked Out of College: When Admissions Bureaucrats Do and Do Not Discriminate (with Jacob Brown). 2022. British Journal of Political Science, 52 (3): 1436-1446. [Abstract] [Preprint]
    How does a criminal record shape interactions with the State and society? We present evidence from a nationwide field experiment, showing that prospective applicants with criminal records are about five percentage points less likely to receive information from college admission offices. However, we demonstrate that bias does not extend to race. There is no difference in response rates to Black and White applicants. We further show that bias is all but absent in public bureaucracies, as discrimination against formerly incarcerated applicants is driven by private schools. Examining why bias is stronger for private colleges, we demonstrate that the private-public difference persists even after accounting for college selectivity, socio-economic composition, and school finances. Moving beyond the measurement of bias, we evaluate an intervention aimed at reducing discrimination: whether an email from an advocate mitigates bias associated with a criminal record. However, we find no evidence that advocate endorsements decrease bureaucratic bias.
17. Does Public Opinion Affect Political Speech? (with Anselm Hager). 2020. American Journal of Political Science, 64 (4): 921-937. [Abstract]
    Does public opinion affect political speech? Of particular interest is whether public opinion affects (i) what topics politicians address and (ii) what positions they endorse. We present evidence from Germany where the government was recently forced to declassify its public opinion research, allowing us to link the content of the research to subsequent speeches. Our causal identification strategy exploits the exogenous timing of the research's dissemination to cabinet members within a window of a few days. We find that exposure to public opinion research leads politicians to markedly change their speech. First, we show that linguistic similarity between political speech and public opinion research increases significantly after reports are passed on to the cabinet, suggesting that politicians change the topics they address. Second, we demonstrate that exposure to public opinion research alters politicians' substantive positions in the direction of majority opinion.
18. Do Inheritance Customs Affect Political and Social Inequality? (with Anselm Hager). 2019. American Journal of Political Science, 63 (4): 758-773. [Abstract]
    Why are some societies more unequal than others? The French revolutionaries believed unequal inheritances among siblings to be responsible for the strict hierarchies of the ancien regime. To achieve equality, the revolutionaries therefore enforced equal inheritance rights. Their goal was to empower women and to disenfranchise the noble class. But do equal inheritances succeed in leveling the societal playing field? We study Germany—a country with pronounced local-level variation in inheritance customs—and find that municipalities that historically equally apportioned wealth, to this day, elect more women into political councils and have fewer aristocrats in the social elite. Using historic data, we point to two mechanisms: wealth equality and pro-egalitarian preferences. In a final step, we also show that, counterintuitively, equitable inheritance customs positively predict income inequality. We interpret this finding to mean that equitable inheritances level the playing field by rewarding talent, not status.

Public Engagement and Policy Reports

1. Navigating California's Wildfire Insurance Challenge: Community Demographics and the FAIR Plan (with L. Hen and M. Macaraeg). 2025. California Policy Lab, University of California.
2. Can Regional Policy Stabilize Structural Transformation? Evidence on the Political Effects of Regional Development Policy in Germany (with V. Heddesheimer, D. Posch, and J. Südekum). 2026. Bertelsmann Foundation.
   • German version
   • [Media: Der Spiegel, FAZ, profil, Social Europe]
3. Why Do Mainstream Parties Lose Either Way on Accommodating the Far Right? 2026. Guest essay, Popular by Design.

Data

1. GERDA: The German Election Database (with Vincent Heddesheimer, Florian Sichart, and Andreas Wiedemann). 2025. Harmonized German election results for municipalities, counties, and constituencies, available as data downloads and packages for R and Python.
2. CLARA: Communication of Local German Party Associations, Recorded Annually (2015–2026) 2026. A chapter-year panel of 2,635 local German party organizations, including leadership, issue emphasis, and 278,638 archived news articles in the 2015–2026 panel.
3. U.S. House (2018–2024) and Senate (2002–2024) Candidate Websites 2026. Archived campaign website text for 7,353 candidate-years, covering 799,058 Wayback Machine pages.
4. Political Science LLM Benchmark 2026. Matched evaluation of 31 open-weight and commercial language models on 33 political science text-coding tasks, with cost and generation speed.
5. German Municipal Broadband Availability (2005–2021) 2026. A harmonized panel of broadband availability for German municipalities, standardized to 2021 boundaries.

Working Papers & Work in Progress

1. Voter Responses to Climate Adaptation in High-Risk Communities (with Christian Baehr and António Valentim). R&R, American Journal of Political Science. [Abstract]
   As climate impacts intensify, governments increasingly adopt adaptation policies, yet their electoral consequences remain poorly understood. We study electoral effects of moratoria on homeowners insurance nonrenewals after California wildfires. Using a difference-in-differences design exploiting eligibility rules, we find null effects on support for incumbents. To explain, we develop a framework of experience-based accountability, arguing that the electoral effect of adaptation is path-dependent: prior crisis experience sets the standard for what voters perceive as a sufficient policy. Consistent with this, we find modest electoral rewards where prior insurance disruption was limited, but none where it was chronic. Survey evidence indicates that, prior to the policy, residents of high-disruption areas already held more negative and crystallized views of the governor. We offer a framework for the politics of adaptation, showing that credit depends on perceived sufficiency: a standard set not only by the policy, but also by voters' prior experience with the crisis.
2. Reindustrialization and Political Change (with Vincent Heddesheimer and Marc Sabatier Hvidkjær). (draft available upon request) [Abstract]
   Can reindustrialization reverse the political consequences of industrial decline? We study this question in France, where deindustrialization has been a common explanation for far-right support and where recent governments have made the return of production a policy priority. We link establishment-level records of large manufacturing site openings to commune-level manufacturing employment as well as elections between 2002 and 2026. Using staggered difference-in-differences designs, we show that new production sites generate sizable gains in local manufacturing employment. The political response to these gains is asymmetric. Incumbent mayors of treated communes become substantially more likely to stand for reelection, behaving as if openings are electorally valuable. Voters respond selectively: far right support does not decline in any election type we study. Far left support falls modestly, concentrated where openings are not knowledge intensive. Governing parties receive no robust average reward. In the 2022 cohort, Macron gains support in declining industrial areas and after trade and craft openings. The pattern is consistent with coarse retrospection over an issue that lacks a partisan owner: visible renewal produces limited and uneven incumbent gains without reducing far-right support. Economic repair does not undo the political consequences of economic loss.
3. Place-Based Policies, Local Responses, and Electoral Behavior (with Vincent Heddesheimer and Andreas Wiedemann). [Abstract]
   Place-based economic policies are increasingly seen as instruments to counter political discontent in economically-depressed regions. Do regional investment subsidies affect electoral behavior? We evaluate this question in the context of Germany's largest regional investment program. Our identification strategy leverages a 2014 subsidy rate cut for manufacturing firms that—due to EU rules—was exogenous to local economic trends. The subsidy cut reduced voting for the far-right AfD and increased support for the incumbent Christian Democrats. We then demonstrate that subsidy cuts trigger two counterreactions: local firms invest more in human capital and local governments invest more in infrastructure and public goods. We argue that these dynamics signal to voters that local governments are responsive and care about their constituents, undermining populist appeals. Our findings suggest that political consequences of place-based policies—and austerity more broadly—cannot be understood without considering counteractions by local firms and governments.
4. Industrial Policy and the Politics of Delivery: Evidence from WWII War Plants (draft available upon request)
5. Long-run Political Change after the Great Recession (with Vincent Heddesheimer). [Abstract]
   What are the long-run political consequences of economic crises? We study the Great Recession in the United States, exploiting geographic variation in unemployment shocks. Harder-hit counties shifted toward Democratic candidates. House gains persisted through 2022, after the shock's employment effects had dissipated, while presidential gains attenuated after 2016. House gains did not track local recovery or Republican incumbency, which weighs against simple retrospective voting accounts. Harder-hit places also showed no shift toward right-wing populist candidates or movements. Instead, survey and ballot measure evidence points to economic scarring and demand for protection: residents of harder-hit places remained pessimistic about their economic prospects, even as personal finances recovered, and became more supportive of social protection. Taken together, the findings suggest that a crisis can durably reshape what affected communities demand from government after acute local labor market distress has subsided.
6. Salient Shock, Imperceptible Impacts: Germans' Immigration Attitudes and the 2022–2023 Energy Price Spike (with Niloufar Davis, W. M. Halm, and Daniel J. Hopkins). [Abstract]
   Prior research finds that personal economic experiences typically have limited political impacts but can shape attitudes under certain conditions. Here, we explore a likely case: in Germany, Russia's 2022 invasion of Ukraine led to a dramatic spike in energy prices and an influx of Ukrainian refugees, both widely covered by news media. We use an online panel of German citizens (2022–2024) to assess the energy shock's impacts on immigration attitudes. Living in areas with larger natural gas price increases (and so higher residential heating costs) is associated with heightened economic distress. Yet we find no systematic evidence that these shocks shifted immigration attitudes over the subsequent 20 months, with minimum detectable effects of 0.10 to 0.18 standard deviations across our five attitude indices. Although extant research documents sizable political responses to some economic shocks, even a large, salient economic shock need not generate exclusionary immigration attitudes.
7. Bureaucratic Inertia or Legal Responsiveness? College Admissions Officers' Behavior Before and After the Affirmative Action Ban (with Jacob Brown and Hunter Rendleman). R&R, Perspectives on Politics. [Abstract]
   How do street-level bureaucrats respond when courts reshape public policy? The 2023 Supreme Court decision striking down affirmative action altered the legal framework around college admissions, prompting speculation about its implications for administrative behavior and minority representation on campus. Yet existing work offers limited evidence on whether -- and how -- bureaucrats adapt to judicial interventions. We investigate whether college admissions officers shift behavior in response to the ruling. We report results from two original field experiments with over 3,000 U.S. admissions offices. In Study 1, we recontact institutions from a 2018 audit testing responsiveness to Black and White applicants, enabling a pre-post comparison. In Study 2, we randomly vary applicant race (Asian, Black, White) and references to the Court's decision. Across both studies, we find no evidence of racial bias, even when the ruling is salient. Together, these null results suggest limits to the judiciary's influence on bureaucratic behavior.
8. Electoral Responses to Local Cooperations with Radical-Right Parties (with Philipp Heyna, Violeta I. Haas, and Tim Wappenhans). (draft available upon request) [Abstract]
   What are the electoral consequences when mainstream parties break the cordon sanitaire against the radical right? Across Europe, this norm is increasingly under pressure. Initial breaches typically occur at the local level. While existing research focuses on national-level accommodation strategies, we know almost nothing about the effects of local cooperation. Drawing on newly collected data on municipal-level cooperations between the AfD and mainstream parties in East Germany (2019–2024), we study the electoral effects of joint legislative voting, committee appointments, and similar forms of collaboration. Using two-period and synthetic difference-in-differences designs, we find that cooperations are associated with substantively meaningful electoral losses for the AfD. Penalties are largest when cooperation is highly visible. To shed light on the mechanism, we use local newspaper data to measure issue salience. The evidence is most consistent with cooperation shifting public discourse toward political pragmatism. These results suggest that when the Brandmauer erodes locally, the radical right may pay a short-term electoral price, even as local breaches risk normalizing cooperation at higher levels of government.
9. Foreign Policy Debates Shape Refugees' Psychological Integration (with Florian Sichart and Georgiy Syunyaev). [Abstract]
   Refugee integration is a central policy challenge. Expanding on prior research on policy debates targeting immigrants, we examine whether host country debates about military aid to refugees' homeland shape their psychological integration. In a survey experiment with 2,631 Ukrainian refugees in Germany, participants viewed authentic statements by German politicians expressing varying levels of support for military aid. On average, these messages had no detectable effect. Yet, this null masks important heterogeneity. By linking our experiment to daily variation in national news coverage, we find that when ambient media salience is low, exposure to any political statement on military aid---supportive or opposing---reduces psychological integration. These effects vanish when the issue is already prominent in the media. We argue that this pattern reflects a contestation mechanism: refugees interpret the very visibility of political debate, rather than its specific content, as a signal that host-country support is contested and potentially less secure.
10. Open-Weight LLMs Are Often Competitive with Commercial APIs for Political Science Text Classification [Abstract]
    Can researchers use local open-weight models instead of commercial APIs for LLM text classification? Local models avoid marginal API charges, keep data on the researcher's machine, and make exact model versions easier to preserve. I benchmark five local models against four commercial API models on 34 political science classification tasks. Local models are often competitive, especially on simpler tasks. In a task-specific oracle comparison, local models match or exceed API performance on 9 tasks; on average, the best API model exceeds the best local model by 0.015 F1. The four strongest observed model means fall within 0.021 F1. API models have their clearest edge on complex tasks with many labels or multiple outputs per item. Batching several items in one prompt usually reduces local runtime per item, but specific model-task pairs can return invalid response formats or labels. Taken together, the results make local open-weight models a practical candidate alternative for many political science classification tasks, provided researchers validate candidate models on task-specific labels and check batching reliability before scaling up.