Working Papers

This article examines how the reception of EU place-based policies is associated with politicians’ sentiment towards the EU. Existing research has paid a lot of attention to the effects of EU place-based policies on the electorate, however, there is little knowledge of the effects on politicians. Relying on a large-N Twitter database with up to 5 million observations by German election candidates in a time period from 2017-2021 I estimate how the regional dependency of EU money is impacting the salience and the sentiment of language towards the EU. For this, I match this corpus with up to 300,000 EU cofinanced projects on the voting district level in Germany. Additionally, I utilize a supervised classification algorithm that identifies eurosceptic tweets. I argue that a higher share of EU funding will lead to a higher and more positive sentiment among candidates. Furthermore, I hypothesize that this effect will differ between parties and is conditional to the party line on EU matters. I find that receiving more EU financial support makes the EU more salient in candidates’ posts. However, there is no change in EU sentiment. In fact, for candidates of the AfD, receiving EU funding is associated with a higher share of EU skeptic posts.

To counter regionally concentrated political frustration, place-based economic policies targeting disadvantaged regions have become increasingly popular among policymakers. This paper examines the effectiveness of place-based policies in mitigating political frustration and populism with a focus on their distributional effects within regions. We argue that place-based policies are unlikely to benefit the most disadvantaged voters due to a skill bias. Empirically, we study the world’s largest place-based policy in the context of the European Union (EU) using a natural experiment and randomized survey experiments. We construct a 1980-2020 dataset with subnational geocodes at the individual level (N=1.4 million) across the EU and leverage a discontinuity in regional funding eligibility in an RD design. In experiments embedded in a two-wave survey (N=2,046), we expose respondents to information and encouragement treatments related to the policy. Both analyses show that place-based policies reduce political discontent only among the highly skilled and the rich, while political attitudes of other groups are not affected. These results highlight that the effectiveness of place-based policies in addressing political discontent is severely limited by their distributional effects.

The growing electoral success of extremist parties has sparked debate over whether whole parties or candidates should be legally banned from running for office. Yet these bans may alienate voters and fuel anti-system sentiment. This paper examines how temporary candidate bans declared by courts affect parties’ and candidates’ electoral strategy and electoral support as well as voter confidence in their democratic system. At the micro level, we empirically study French court decisions to declare candidates ineligible for running for office based on non-compliance with campaign finance rules. We compile a dataset of voting behaviour, candidate manifestos, the national commission for campaign finance’s (CNCCFP) reports, and Constitutional Court rulings. This allows us to trace the whole pipeline from candidacy over the campaign to potential sanction. For causal identification we employ a staggered difference-in-differences design. At the macro level, we exploit court rulings, declaring prominent political candidates ineligible in embezzlement cases. Since the decisions occurred during field work for surveys, we can leverage unexpected events during survey designs strategies to causally estimate the effect of the court ruling on voter confidence in democracy.

Place-based policies are often depicted as a means of decreasing resentment by improving economic conditions in left-behind places. Yet, little attention had been paid to the heterogeneity of funds and their respective ends. Relying on 335,000 projects by the European Cohesion Policy for the funding period 2014-2020 in Germany, I investigate how different funding types impact the vote share for the eurosceptic party AfD and support for the EU. First, for the electoral analysis, I use a fine-grained municipality level-dataset (n = 48,045). Second, to further investigate the mechanism of my argument, I harmonized two GLES election datasets (n = 12,000) for examining individual-level data. I find that public good projects and direct firm support are especially potent in improving firm tax revenues. Nevertheless, it is projects dedicated to individual support or public goods that significantly decrease AfD vote share. Lastly, I show that the effect on project type on EU support is depending on the income-level of citizens.

This article presents a supervised classification model for measuring mentions and sentiments about the European Union. This research provides new ways to study and explain variation in Euroscepticism. Specifically, this research presents a binary classification algorithm for detecting speech about the European Union. In addition, it introduces a second classifier consisting of a multi-class classification model for EU praise, hard eurosceptic sentiment, policy scepticism and demand for further EU integration. These are widely used concepts in Euroscepticism research. Based on a pre-trained BERT-based model, I fine tune my classification models on 2000 human annotated tweets of German election candidates between 2007-2021. I am further refining this by training additional models on parliamentary speeches, in order to expand the potential texts to be analysed with these models. All models are available in German and English. This new classification application can expand the possibilities for researchers to study how politicians write and speak about the EU.