Elena Leuschner

Privileged citizens are more likely to participate in politics and to be better politically represented. Does unequal participation drive unequal representation? Or are politicians biased toward privileged citizens, regardless of their political activity? While politicians’ biases toward high-status citizens have been documented, the role of political participation therein remains empirically under-researched. Using nearly 40 years of uniquely comprehensive survey data on Swedish politicians and citizens, we provide the most robust observational evidence to date that politicians’ perceptions of public opinion are more consistently aligned with the opinions of the politically active. Motivated by this background evidence, a novel pre-registered survey experiment with Swedish politicians will test whether politicians give greater weight to an identical political activity based solely on the participator’s status. Our integration of observational and experimental findings will provide new answers to the important question of whether low-status citizens could increase their political influence by participating more.

Do voters from different partisan camps understand democracy in the same way? We analyze more than 25,000 open-ended survey responses from representative samples in 14 countries, classifying answers with a fine-tuned language model and sentence embeddings. Citizens most often define democracy through individual and civil rights; multidimensional definitions are rare, and elections are mentioned by only 17% of respondents. Yet shared vocabulary masks partisan divergence. Far-right voters and non-voters are less likely to mention civil rights, elections, and horizontal accountability, while far-right voters more often invoke majoritarian principles. Among respondents mentioning rights, left voters connect rights to equality and citizenship, whereas right voters emphasize freedom of speech. Democratic contestation thus unfolds within democratic language, creating openings for selective elite reframing.

Incumbents often make unpopular policy decisions. But can they remedy their negative electoral consequences? We extend the wide literature concerning retrospective voting to the case of unpopular policies and examine whether voters reward a responsive withdrawal of an unpopular policy proposal or punish the disclosed policy intention despite the withdrawal. To test this, we use granular data on Swedish local election results from 2002 to 2018 and the case of widely unpopular school closure proposals, some of which were implemented and others not. We exploit within municipality variation in voting over time to causally estimate the consequences for incumbents in the neighborhood surrounding the schools. Our results confirm that even if a school remains open, voters punish the incumbent and consider the initial proposal as informative for their vote. Our findings have implications for the understanding of democratic accountability and which information voters take into account when casting their vote.

Governments spend hundreds of millions on evaluations to assess the performance of public organizations. In this article, we scrutinize whether variation in the institutional design of evaluation systems leads to biases in evaluation findings. Biases may emerge because influence over evaluation processes could enable the bureaucracy to present its work in a more positive way. We study evaluation reports published by nine international organizations (IOs) of the United Nations system. We use deep learning to measure the share of positive assessments at the sentence level per evaluation report as a proxy for the positivity of evaluation results. Analyzing 1082 evaluation reports, we find that reports commissioned by operative units, as compared to central evaluation units, systematically contain more positive assessments. Theoretically, this link between institutional design choices and evaluation outcomes may explain why policymakers perceive similar tools for evidence‐based policymaking as functional in some organizations, and politicized in others.

Autocrats typically respond with coercion when citizens take to the streets demanding political reform. Sometimes, however, they tolerate mass protests and even give in to protesters’ demands. While the effect of coercion on mobilization is well-studied, we know less about the role of concession-making. We argue that accommodating demands is rarely an effective strategy in demobilizing opposition movements. Authoritarian rulers are usually neither willing nor able to fully address protesters’ dynamic demands, nor can they offer credible commitments. We conduct a quantitative analysis using multiple cross-national data sets to empirically assess the relationship between concessions by the government and subsequent mass mobilization. By analyzing protest events in temporal and spatial proximity, we estimate the effect of making concessions on protest mobilization at the subnational level in 18 autocracies from 1991 to 2012. Our results indicate that concessions are associated with a significant and substantive increase in subsequent protest activity.

International organizations (IOs) of the United Nations (UN) system publish around 750 evaluation reports per year, offering insights on their performance across project, program, institutional, and thematic activities. So far, it was not feasible to extract quantitative performance measures from these text-based reports. Using deep learning, this article presents a novel text-based performance metric: We classify individual sentences as containing a negative, positive, or neutral assessment of the evaluated IO activity and then compute the share of positive sentences per report. Content validation yields that the measure adequately reflects the underlying concept of performance; convergent validation finds high correlation with human-provided performance scores by the World Bank; and construct validation shows that our measure has theoretically expected results. Based on this, we present a novel dataset with performance measures for 1,082 evaluated activities implemented by nine UN system IOs and discuss avenues for further research.