CAP 26 — Program
Wednesday, 17 June
1.1 · Agenda Dynamics & Political Attention I
Democracy and Media Attention Dynamics
Abstract
Recent comparative scholarship on punctuated equilibrium shows that governing conditions and regime types shape the dynamics of legislative and budgetary change, with authoritarian-leaning regimes exhibiting more disjointed and episodic policy activity (Baumgartner et al. 2018, Lam and Chan 2014, Frantzeskaki and Bech Seeberg 2022). In this paper, we extend this literature by assessing the dynamics of policy related attention in news media and test the extent to which variations in democracy shape inputs to the policy process. We expect that more democratic regimes, as a result of civil society capacity, citizen mobilization, and government information transparency, will more smoothly attend to domestic policy issues over time. We leverage new data from the Machine Learning for Peace project, spanning 1m+ news articles reporting in 5 countries (two imperfect democracies, two autocracies, and one backsliding democracy) and implement Sebők et. al. (2024)’s fine-tuned, multi-lingual LLM to classify articles in the CAP topic scheme. We estimate month-country panel models (N=780) to test our hypothesis.
Agenda Diversity and Policy Entrepreneurship in the Federal Bureaucracy
Abstract
Why do some agencies successfully assert jurisdiction over emerging policy domains while others remain confined to their traditional mandates? Existing research focuses on how legislative directives shape regulatory expansion, but fails to explain how agencies claim authority in the absence of new laws. I argue that diverse policy portfolios provide agencies with more flexible interpretive frameworks that enable them to credibly frame emerging problems as extensions of existing authority. To test this theory, I analyze an original dataset of 959,325 regulations enacted across 105 federal agencies in the US bureaucracy from 1970 to 2024, coded by major topic using the Policy Agendas Project classification system. I measure agenda diversity using the Shannon entropy of an agency’s regulatory portfolio and track expansion by identifying the introduction of novel tokens within specific policy topics over time. I find that agencies with a broader issue portfolio are more likely to expand their authority into novel issues. These findings suggest that the structure of the bureaucracy shapes how governments adapt to new problems.
Personal or Partisan Agendas? Strategic Issue Selection in Parliamentary Written Questions
Abstract
Most parliaments allow legislators to submit written questions to the government, a key instrument for monitoring executive action and ensuring accountability. Existing research shows that legislators use written questions to advance personal agendas, including the representation of constituents or specific social groups. Yet we still know little about how legislators strategically select policy topics when using this instrument, and how personal and partisan incentives interact in shaping this behavior. In this article, we develop a framework that links written questions to legislators’ career motivations—reselection, reelection, and advancement—by distinguishing between partisan priorities and individual-level sources of expertise and electoral incentives. We argue that MPs are more likely to engage with policy areas that are salient to their party and electoral district, but that this partisan logic is conditioned by individual characteristics such as pre-parliamentary professional experience and committee membership, which lower information costs and foster issue specialization. We test these expectations using an original dataset of nearly 940,000 CAP-coded written questions submitted in the Spanish Congreso de los Diputados between 1996 and 2023. The paper highlights written questions as a strategic venue through which legislators combine collective party goals with personal career incentives, contributing to broader debates on agenda-setting, parliamentary behavior, and legislative representation.
Talking to Whom, How, and When? Health Regulatory Communication under Political Pressure in Latin America
Abstract
Governments set policy agendas under conditions of bounded rationality, allocating limited attention to a small number of issues and actors while filtering out most alternatives. Yet we know surprisingly little about how governments prioritize actors when observable capacity cannot guide attention. This challenge becomes especially acute during periods of polycrisis—such as the COVID-19 pandemic—when overlapping pressures across policy domains compress decision time and constrain fiscal, technical, and organizational resources. This paper introduces projected capacity as a novel agenda-setting mechanism that operates under such conditions. Projected capacity captures the state’s anticipatory attribution of future competence to specific actors when material capacity cannot yet be observed, verified, or mobilized within the decision-making timeframe. Rather than relying on demonstrated performance, governments draw on pre-crisis signals to prioritize attention under urgency and uncertainty. Empirically, projected capacity is operationalized through a pre-crisis, actor-level index combining three observable dimensions: institutional embeddedness (documented prior interactions with the state, such as contracts or participation in public programs), cognitive availability (actors’ visibility in official ministerial communications on Twitter/X prior to the crisis), and publicly articulated expectations of future capacity (official statements or policy documents identifying actors as strategically relevant). All indicators are measured independently from agenda outcomes and aggregated into a standardized index. The framework is tested using Argentina as a case of a lower-middle-income country that is analytically revealing. The analysis draws on an original dataset of legislative texts—including bills, resolutions, and legislative initiatives—produced during the polycrisis period of 2020–2022. Policy attention is measured as actor-level legislative attention, captured through the frequency and concentration of actor references across policy domains. The findings show that actors with higher projected capacity receive a disproportionately larger share of legislative attention before material capacity can be verified or mobilized. These results identify a mechanism of agenda-setting that helps explain how governments allocate attention under extreme constraints, with implications for understanding agenda dynamics across policy domains in crisis contexts.
1.2 · Media, Communication & Public Opinion I
Tweets from the Gallery: Covering Congress in the Digital Age
Abstract
Congress isn’t your typical workplace — unless your office also has TV cameras in the hallways and reporters shouting questions at you. For the reporters whose job it is to ask those questions, Congress is a beat unlike any other where access is unrivaled, Twitter/X remains relevant, and storylines are unending. The story of Congress is told by reporters whose daily work in the press galleries and on Twitter shapes what the nation sees as politics and technology change the rules. This book considers how the congressional reporters that drive the deluge of information have adapted to meet the rapid pace of news and digital information. Covering Congress is a constant effort to stay professionally relevant, both in person and online. I explain how congressional reporting is a careful balancing act between guarding the gates and getting through them, drawing on interviews with current and former journalists and observations of Capitol Hill to show how reporters navigate the competing demands of access, speed, and accuracy in a media environment shaped by partisan politics and relentless technological change.
After Hours, Off Message: How Modern Politicians Communicates in Two Distinct Modes
Abstract
The modern communication environment enables politicians to communicate directly to the public around the clock — yet little is known about how politicians communicate when they are not at work. In response, this article explores the temporal dynamics of political communication, arguing that politicians exhibit dual communication behaviors depending on whether they are on or off-duty. Drawing on a novel dataset of over 5.5 million tweets from MPs in UK and Denmark (2015–2022), I find that issue ownership—the tendency for parties and politicians to emphasize issues the party is deemed most competent in handling—is temporally contingent. Specifically, politicians are significantly more likely to emphasize party-owned issues during work hours and on weekdays, whereas off-duty communication, such as weekends and evenings, sees a marked decline in alignment with party issue ownership. Furthermore, a similar picture appears in device use: tweets sent from laptops align more closely with party-owned issues than those sent from mobile devices. These patterns suggest the modern politician has developed two versions of themselves: a disciplined, party-aligned self, active during official hours, and a more autonomous self after hours. Consequently, the rise of a communication environment in which politicians are never off-duty has profound consequences not only for the timing of their communication, but also for what they communicate about. Until recently, these dynamics remained both unobservable and largely irrelevant, as politicians lacked the means to communicate around the clock. This article, however, opens up entirely new angles of understanding political communication in a time where politicians are always “on,” allowing us to grasp how the modern politician has developed two versions of themselves.
The Gray Lady and Policy Agendas: Introducing the New York Times Data Set
Abstract
It has long been evident that the news media can influence attitudes of electorates, effectively “shaping political reality” through various channels (McCombs and Shaw 1972). Through its coverage, news media helps both to set the policy agendas of governments and raises the public salience of issues both during and after electoral campaigns. As such, it is critical to capture the policy content of media coverage, as it might have downstream effects on political outcomes of interest. We expand the US Policy Agendas Project New York Times dataset, drastically increasing the time period under study and expanding the CAP media coding addendum originally developed by Boydstun (2014). The New York Times MediaText Dataset is a comprehensive dataset of all New York Times articles spanning 1851 to 2025 — over 15 million articles retrieved from the New York Times API and third-party API services. We code articles using a supervised machine learning approach, classifying cases into Comparative Agendas Project major topic categories and a series of new codes designed to better capture non-policy content. This significant extension of the existing PAP media dataset will enable the CAP community and researchers generally to conduct systematic analyses of media coverage, agenda-setting, and journalistic attention over more than half of American history.
Channels of Power: Presidential Communication and Agenda Dynamics in Russia 2014-2025
Abstract
The political calendar in Russia is traditionally dominated by several events: the Annual Presidential Address delivered to the Federal Assembly, the yearly press conference with the media, and the “Direct Line” with the President, a televised question-and-answer event with the general public. The Annual Presidential Address to the Federal Assembly, typically delivered at the start or end of each year, plays a significant role in shaping the nation’s policy direction despite lacking formal legal authority. The “Direct Line,” structured as a question-and-answer event with de facto predefined questions, enables the President to communicate his views and policies to a broad audience in a talk show format. The yearly press conference is the main communication event for the media representatives. In 2024, the press conference and the “Direct Line” were merged into a single event, and with the Presidential Address once again omitted in 2025, these alternative communication formats have become even more important for understanding presidential agenda-setting in Russia. Our paper focuses on these diverse and often overlooked sources to explore the dynamics of the presidential agenda in Russia over the last decade, covering the period from 2014 to 2024.Through a detailed analysis of these communications, we investigate the evolution of the presidential agenda, targeting different audiences. It allows us to trace the presidential policy priorities in the context of Russia’s increasing isolation from Western values and partners, culminating in the initiation of a military operation in Ukraine in 2022. A preliminary analysis of the Presidential Addresses reveals a shift in policy priorities from economic development and social policies to an increasingly belligerent defense policy, providing insights into the broader political dynamics at play in Russia during this period. We use manual content analysis based on the Comparative Agendas Project coding scheme, adapted for Russia, to track changes in policy attention.
1.3 · Parties, Elections & Representation I
Favorable news Media coverage, issue ownership and party support
Abstract
At the core of the issue-ownership theory is the proposition that political parties benefit electorally from public attention to issues they own. Current research lacks empirical examinations of this macro level proposition, leading to uncertainty regarding the electoral issue ownership effect. This paucity likely reflects the resource-intensive data collection required to capture the distribution of issue attention in national political debates, across countries and over time. Employing a corpus of more than five million news articles in four countries over nearly two decades, this paper offers the most extensive empirical examination of the core issue ownership effect to date. Our conclusion is that positive issue reputations matter to party support, aligning with theories of issue ownership and media priming. When issues rise on the news agenda, the parties that own these issues attract more voters. This effect is remarkably stable across time, countries, and different categories of parties.
Radical-right parties, government-seeking, and welfare politics
Abstract
Whereas radical-right parties carved their way into the political landscape through their distinguished anti-immigration stances, many have recognized that their continuous electoral success could not have been without the substantive working-class support they have garnered. Consequently, this growing working-class presence has been accompanied by more attention to and representation on welfare issues by these parties. Yet, radical-right parties’ ways to power have thus far been through cooperation and coalitions with mainstream right-wing parties, with rather opposite economic agendas. We currently know very little about to what extent such cooperation influences the welfare politics of radical-right parties. While it has been suggested that radical-right parties adapt their welfare positions to that of their respective mainstream right-wing party, others have argued that the populist profile of these parties and their followers come with a policy purist feature, making concessions less likely due to its relatively high costs. In this paper, I seek to study how and when government inclusion chances affect radical-right parties’ welfare politics. I do so by focusing on party competition and argue that welfare accommodation is contingent on the relative immigration policy distance between a radical-right party and the mainstream right-wing party. To test this, I employ a novel dataset covering 30 Western democracies over three decades.
Unpinning Expertise
Abstract
Policy experts and “expert-based information” play an important role in policy processes (Weible 2008). Policymakers rely on information from policy experts in order to learn, legitimize political decisions, and problem solve. Yet, recent research suggests that trust in policy experts is polarized and Republicans increasingly disagree that scientists share their values (Evans and Hargittai 2020). I theorize that the definition of an “expert” and the heuristics used in gauging expertise varies amongst American citizens. While previous research often pins expertise on educational qualifications, the chasm between liberal and conservative acceptance of higher education and partisan sorting by college degrees suggests that the broader population does not agree. Further, partisan sorting confounds whether citizens (dis)trust policy experts because of their ideology or because of demographic (in)congruence. Using a single-profile conjoint experiment, I present survey respondents with hypothetical witnesses who have been invited to testify and examine the respondents’ use of heuristics in gauging expertise. I field three survey experiments, each in a unique policy area, to test whether perceptions of expertise are sensitive to the issue at hand. I also use observational data from the American National Election Studies to undergird my experimental findings. Analyses for this project will be complete before the Comparative Agendas Project conference.
2.1 · Agenda Dynamics & Political Attention II
How parties invent new issues: The case of digitalisation policy
Abstract
Issues are integral to party competition and agenda setting. However, the concept of “issues” is rarely defined, hindering research into the changing structure of both. This lack becomes problematic over time, as challenges to CAP & MARPOR categories indicate: New or expanding problems and policies blur established delineations between issues for political actors and researchers alike.
This paper proposes a remedial issue concept and theorises vote- & policy-seeking parties as agents setting delineations between issues. Constrained by established delineations, they use discursive opportunities to shift them in their favour, based on electoral, administrative and intra-party incentives. Whenever salient problems and policies do not fit easily into an established issue, parties choose to (a) integrate them into several existing issues, (a) expand a single existing issue or (c) form a new issue.
This framework is empirically applied to the emerging issue of digitalisation policy to explore how new and cross-cutting problems and policies are formed into issues that can be selectively emphasized and put on the agenda. Results speak to the agenda setting and issue competition as well as the policy integration literature. Additionally, the issue concept might provide a theoretical footing for future LLM-/classifier-based issue coding.
From state domains to issue areas: Government activities in the long-term
Abstract
Policy areas constitute the basic areas in which modern governments allocate attention, organize expertise, and formulate and implement government policy. Yet the origins and long-term development of these areas remain poorly understood. Much research in public policy treats policy domains as analytically given or emerging directly from democratic politics, while historical scholarship on state formation rarely conceptualizes these issue domains. This paper advances a process-oriented account that links the evolution of policy areas to the dynamics of agenda setting, jurisdictional expansion, and administrative development and consolidation.
We argue that policy areas emerge through successive waves of state attention to policy issue, driven by the gradual extension of state jurisdiction, growing societal differentiation, and the construction of public problems. Stabilization occurs when repeated agenda engagement prompts administrative self-reference, organizational specialization, and sustained policy subsystems. Rather than viewing democracy as the primary source of policy areas, we conceptualize democratization as a transformative shift that alters how policy agendas develop. Democratic institutions broaden participation in agenda setting, reshape state selectivity, intensify cross-domain competition, and accelerate subsystem formation, thereby modifying rather than initiating the construction of policy areas. Accordingly, the longitudinal pathways in democratization matter for the emergence of policy areas.
Empirically, the paper draws on a novel longitudinal quantitative dataset covering all formal ministerial units in Prussia and Germany, France, Japan, Norway, and the United Kingdom from the early 19th century until the 1960s. By tracing this emergence, stabilization, and reorganization of state activities across countries and issue areas with two different coding schemes (CAP and a novel coding informed by the denomination of ministerial units), we identify distinct trajectories in the formation and stabilization of issue areas. Our findings demonstrate that issue areas may originate in pre-democratic contexts, with apparent consequences on their shape later on, while democratic regimes facilitate the emergence of novel issue areas. Furthermore, we identify critical junctures and path dependencies across issue areas and countries. Accordingly, we broaden our current understanding in agenda research towards longitudinal comparisons that arguably affect also current dynamics.
The Policy Content of Oversight: GAO Activity Through the Comparative Agendas Lens
Abstract
This paper introduces a new dataset that links the U.S. Government Accountability Office’s (GAO) oversight activity to policy attention using the Comparative Agendas Project (CAP) coding framework. Leveraging more than 50,000 GAO reports and over 70,000 agency-directed recommendations, the study used the Babel Machine (2025) to code each GAO product and recommendation to a CAP major topic and subtopic, enabling systematic analysis of how oversight attention is distributed across policy domains. The dataset allows scholars to examine whether GAO oversight mirrors, amplifies, or diverges from congressional policy agendas, and to assess variation in the generation and implementation of recommendations across issue areas. Using descriptive and exploratory analyses, the paper demonstrates substantial cross-policy heterogeneity in GAO activity, with some policy domains characterized by dense, recurrent oversight and others by episodic or limited engagement. By integrating GAO oversight data with the CAP framework, this paper provides a scalable tool for studying agenda setting, bureaucratic accountability, and the role of audit institutions in the U.S. policy process.
The Hierarchy of Needs and the Public Agenda: How Crises and the Economy Drive the ‘Most Important Issue’
Abstract
What determines the issues that make it onto the public agenda and when? Despite the importance of this question to the study of agenda-setting and substantial interest in the ‘most important problem’ in public opinion, there have been few attempts to understand in general terms the conditions under which certain issues come to be salient for the mass public. In this paper, we develop a theory of when and how issues ‘crowd out’ attention to other issues, driven by a ‘hierarchy of needs’ where physical safety and economic concerns supersede concern about social organization and post-material issues and other policy issues. Specifically, we argue that when the economy and other major crises become salient in public opinion they reduce attention to other issues – even when those issues remain important to citizens. As a result, economic downturns or moments of societal crisis (such as the Covid-19 pandemic) can lead to sudden and unexpected contraction of the range of issues that voters collectively consider important.
To test our theory, we make use of a monthly time series of survey data on the “Most Important Issue” in the UK, with 458 monthly observations, from 1977 to 2024. We employ dynamic correlation between issues, and with the overall dispersion of the public agenda. Our findings demonstrate that while the economy generally ‘crowds out’ public attention to other issues, this is not always the case. It also shows a strengthening of attention with trade. Just as importantly, while the general pattern is that increased attention to the economy tends to lead to concentration of the public agenda, this is not always the case (most notably during the Covid-19 crisis).
Emotions and policy punctuations: an exploration using large language model analysis
Abstract
The Punctuated Equilibrium Theory (PET) posits that policy images combining evidence, values and emotions influence the understanding and interpretation of policy problems. Policy images have the potential to increase the importance of a policy issue and contribute to policy change. In the PET framework policy entrepreneurs have an important role in promoting new policy images, breaking policy monopolies, and putting issues on the agenda. Maor argues that some policy actors should be considered ‘emotional policy entrepreneurs’ who primarily focus on manipulating emotions—altering the affective experiences, expressions, and behavior of others—in pursuit of their policy goals. Besides policy entrepreneurs, the role of tragic events like school shootings or natural disasters should be underscored as they can serve as emotionally loaded ‘focusing events’ opening policy windows for learning and reform. While PET implies that intense emotional reactions—for instance, fear, moral outrage, or hope—can reframe policy problems, focus public and elite attention, and enable learning that results in policy shifts, the role of emotions hasn’t been in the focus of PET-related empirical research so far. Our paper engages into exploratory research in this field, using an AI-powered tool to identify emotions in policy discourse. Analyzing data from the Hungarian Comparative Agendas Project we address questions like: Does the distribution of emotions also follow a leptokurtic distribution? Can we link emotional storms to external focusing events? Can we identify ‘emotional policy entrepreneurs’ through policy discourses? Does issue saliency increase the emotionalization of policy discourse? Is there an association between policy punctuations and emotional intensity of different policy agendas?
2.2 · Media, Communication & Public Opinion II
Citizens and politicians hold shared perceptions of power and deservingness of needy groups
Abstract
Social construction of a target population (Schneider and Ingram 2005) as well as work on deservingness (van Oorschot 2006, Petersen et al. 2011) highlight that welfare preferences are explained by political values and perceptions of deservingness of recipients. Citizens and politicians perceive some groups as powerful and some groups as deserving. A shared understanding of societal groups and electoral accountability leads to similar views between citizens and politicians about deservingness and power. Welfare benefits are directed toward groups that appear to be powerful and deserving (such as the elderly), but target group characterized as undeserving and powerless (such as the unemployed) receive only contingent benefits. In a second step, we inquire into the mechanisms triggering these beliefs about recipients. Building on moral judgement theory (Haidt and Joseph 2004), citizens and politicians distinguish between fairness considerations and belonging to an ingroup when assigning benefits. We survey citizens and nationally elected politicians in Canada, Germany, Luxembourg and Norway in 2022 in order to test our expectations. First, participants rate six groups (elderly, drug addicts, people with disabilities, unemployed, investment bankers and farmers) in terms of deservingness and power. A model-based cluster analysis groups these responses and compares them between citizens and politicians. Second, in an experimental design, we test if social benefits are assigned based on fairness and/or ingroup considerations. Overall, the study advances our understanding about the treatment of welfare recipients and show the congruence in perceptions of deservingness between citizens and politicians as well the common moral judgement about these choices.
Does the dog watch? When and how the media reports on political elites’ repositioning
Abstract
Traditional media remains central to how citizens learn about politics. When political elites change their minds about a policy issue, media coverage of these shifts is essential for informing the general public. Despite growing attention to elite position shifts in the party competition literature, little research has examined the role of the media. Here, we present the first study on when and how the media report on issue position changes of political elites. To this end, we create a novel dataset using a large corpus of Swiss news articles from the past 20 years, in order to identify factors associated with the intensity and tonality of media coverage of political elites’ position changes. Through inductive, manual coding, as well as drawing on the CAP codebook to identify different issues, we develop a typology of issue position changes which we apply to the entire corpus using a transfer learning approach. In general, we expect a negative tonality in news articles about position changes. Drawing mainly on news value theory, we hypothesize that the intensity and tonality of media coverage depend on the characteristics of the position change and of the political elite involved in repositioning. Notably, we hypothesize that a higher degree of unexpectedness, e.g., when elites change their positions on one of their core issues, and a greater position inconsistency results in more negative news coverage, particularly among elites from radical parties and those in more prominent political roles. Our findings contribute to our understanding of how the media amplifies or diminishes attention and frames political elites’ issue repositioning. These dynamics are crucial for understanding how political media coverage influences political competition and political behavior in advanced Western democracies.
Explaining Media Attention to New Zealand Supreme Court Cases
Abstract
How do journalists decide which judicial decisions merit attention, particularly in political systems where courts lack strong constitutional powers? This article examines case-level characteristics that drive media attention to decisions of the New Zealand Supreme Court (NZSC), a relatively new apex court established in 2004 following the abolition of appeals to the Privy Council. Drawing on cue theory and scholarship on news values, we analyze whether and how case characteristics and case participants shape both the likelihood and duration of newspaper coverage. We draw on two original datasets—the New Zealand Supreme Court Database and a comprehensive dataset of New Zealand Herald articles from 2004 to 2017. In previous work, we evaluated the effects of a set of variables thought by the literature to affect a case’s newsworthiness: 1) Court-issued press releases, 2) the readability of those releases, 3) Attorney-General participation, 4) civil society participation, 5) amicus curiae participation, and 6) the presence of dissenting opinions. We found press release readability and intervener participation to be the most consistent and powerful predictors of media coverage. Decisions accompanied by more readable Court press releases were significantly more likely to receive coverage, while cases involving interveners were both more likely to be reported and to remain in the news cycle longer. By contrast, participation by civil society organizations as original parties did not increase media attention, and Attorney-General participation did not independently predict coverage. Dissenting opinions increased coverage only when they arose in policy areas not traditionally associated with judicial decision-making. In this paper, we employ additional independent variables, measurement, and modeling strategies to enhance the analysis. Additional case-characteristic variables include: 1) whether the case concerned the Treaty of Waitangi, the foundational document between Māori and the British Crown; 2) whether the case concerned New Zealand’s statutory Bill of Rights; 3) and the type of policy area implicated in the case. In addition to examining the effects of these new variables, we construct a cumulative independent variable that accounts for the presence of multiple newsworthy indicators in order to assess if the presence of multiple variables in a case influence a case’s newsworthiness. We ultimately seek to produce more scholarly information about the way media outlets respond to, report on, and by extension – legitimize (or de-legitimize) a legal-political institution.
From media discourses to policy: The imprint of group constructions in the media on policy choices affecting them
Abstract
Existing research suggests that social constructions shape policy choices and group representation, yet this effect remains difficult to test empirically. This paper (1) develops a theoretical argument on how media framing and representation drive policy change via public opinion and (2) draws on recent advances in natural language processing to provide an empirical test of this relationship. Social constructions of groups are measured using natural language inference (NLI) models trained to identify group mentions and to capture framing and tone in French and German newspaper coverage from 1995 to 2023. Policy change is operationalized through group-specific scales measuring the restrictiveness or supportiveness of policies over time. The analysis focuses on three groups, women, immigrants, and investors, each associated with distinct media narratives and policy contexts. Regression models test the impact of media coverage on policy outputs while controlling for public opinion trends, economic indicators, and political context. The findings provide new evidence on how media narratives shape public perceptions and translate into policy change.
Playing the Crowd : Linguistic strategies of politicians
Abstract
Playing the Crowd - Political communication involves much more than the direct message; meaning emerges from the interaction between message, speaker, audience, and delivery. This paper explores one underexamined aspect of that dynamic: how politicians adapt their language to their audiences. How do they consciously or unconsciously adjust their speech, and what motivates these choices? Drawing on in-depth interviews with 82 politicians from Flanders, Belgium, the study investigates how political actors describe their own language adaptation strategies across various communicative contexts. Special attention is given to their use of regional accents, dialects, and informal registers, and to how they perceive citizens’ reactions to these choices. The paper maps the range of approaches politicians adopt toward linguistic flexibility and explores how differing motivations—fostering closeness, authenticity, and professionalism—shape these strategies. It also shows how politicians, regardless of their approach, feel judged by citizens on how they speak as much as on what they say, how their mastery of different language registers constrains the strategies available to them, and how the region they represent changes this dynamic. In doing so, it offers new insights into how linguistic strategy mediates the relationship between politicians and citizens.
2.3 · Policy Content & Specific Domains I
Questioning Defence Policy in the French Parliament: A Study of Parliamentary Questions on Military Operations (2012–2025)
Abstract
This paper examines how French Members of Parliament (MPs) engage with defence policy through parliamentary questions in the Assemblée nationale between 2012 and 2025. Combining manual coding with automated data extraction from the French parliamentary database, the study covers a period marked by major external military operations, including interventions in Mali and Libya, as well as a renewed focus on high-intensity conflict following Russia’s full-scale invasion of Ukraine in 2022. Building on earlier research on the 2001-2012 period - which identified a notable increase in questions concerning military expenditure from 2002 onward - this paper analyses the evolving patterns and functions of parliamentary scrutiny in the field of defence.
Preliminary findings reveal a significant rise in defence-related questions between 2012 and 2017, particularly those addressing the costs of external operations. More recent developments, notably the war in Ukraine, provide an important backdrop for understanding renewed parliamentary attention to defence spending, military capabilities, and international commitments. This trend prompts a closer examination of MPs’ motivations and the substantive focus of their inquiries. Specifically, the paper asks: when MPs address external military operations, which dimensions attract the most attention? How central are financial considerations in these questions? To what extent do parliamentary questions function as instruments of democratic oversight, as opposed to serving alternative purposes such as signalling expertise, constituency advocacy, or articulating military demands?
More broadly, the paper aims at assessing to what extent the French parliament constitutes a key arena for agenda-setting and political control in defence policy. By analysing parliamentary questions, it sheds light on how civilian oversight of the military is exercised in practice and how defence priorities are negotiated under increasing strategic and budgetary constraints in a security environment reshaped by the war in Ukraine. The paper thus contributes to debates on legislative oversight, civil–military relations, and defence policymaking.
AI Policy Advice Systems: Accessibility, Equity, and Multidisciplinary Collaboration
Abstract
This paper aims to examine the conditions under which AI policy advice systems (AIPAs) are more likely to adopt a multidisciplinary composition and collaborative framework. Advanced democracies have developed AIPAs to guide the responsible development, deployment, and use of artificial intelligence technologies. These governance structures serve a range of essential functions, including establishing regulatory standards, monitoring technological advancements, addressing ethical concerns, ensuring equitable access to AI’s benefits, and mitigating risks associated with AI misuse. However, AIPAs vary in their accessibility, openness, equity, and alignment with guiding principles and goals. Accessibility refers to how technical and practical knowledge is made available through public services, think tanks, NGOs, or private consultants. The openness of channels highlights the inclusivity of advisory systems, ensuring that a wide array of stakeholders can contribute their insights. Equity in access emphasizes creating equal opportunities for all actors to offer their perspectives and have their advice considered in policymaking. Alignment with principles, goals, and instruments underscores how AIPAs are shaped by the broader policy subsystem. Building on frameworks from Policy Advice Systems (PAS) and governance approaches, this paper explores how advice is structured, delivered, and utilized in AI decision-making processes. Using a mixed-methods approach, we compare 73 AIPAs through both quantitative and qualitative analyses. We explore their accessibility, the alignment of policy ideas, and other institutional characteristics. The paper reveals significant differences in AIPAs, with notable variations in the number and type of actors involved, the functions they perform, and the resources they have at their disposal. Accessibility within PAS not only facilitates technically sound advice but also fosters inclusiveness, usability, and alignment with the realities of governance. Furthermore, collaborative frameworks ensure that diverse perspectives shape decision-making, enabling governance systems to balance innovation with accountability and equity. This comparative analysis highlights the importance of designing AI policy advice systems that are both comprehensive and inclusive, ensuring effective and equitable policymaking in the rapidly evolving field of artificial intelligence.
US Education Policy and Party Politics
Abstract
Twenty-five years ago, the United States was on the cusp of a major expansion of the federal government’s role in K-12 education policy. The No Child Left Behind legislation had bipartisan support and established standards and accountability as strategies to improve education. Until roughly 2013, reading and math scores did improve; however, in the years since, student achievement on test scores has gone down, and this trend accelerated during the COVID-19 pandemic. In the last decade, with new and rising challenges for educational outcomes, how has the politics of education at the federal level changed? What happened to bipartisan support for standards-based accountability, and what has replaced it? We code and analyze longitudinal data on changes in party positions, congressional hearings from the U.S. Policy Agendas Project, and executive action on education. Our results show that attention to education policy has receded on the congressional agenda, after a significant policy punctuation with the adoption of No Child Left Behind. We go on to closely assess the aftermath of this punctuation, and we examine how growing attention from the executive branch and increasing partisan polarization have shifted the policy image of education away from standards and accountability and into a culture war battleground.
Our results show that both major political parties backed away from standards and accountability in education, with Republicans putting greater emphasis recently on discipline, morality, and patriotic education, while Democrats emphasize policies impacting targeted groups. Furthermore, presidents have become more active in education policy. Administrative actions during the Biden and Trump presidencies have steered education further from accountability policies by issuing waivers to states from federal requirements and focusing on sharply contrasting partisan approaches to issues such as racial/ethnic and gender identities as well as history and civics curriculum. While most of the policy structures that created standards and accountability in education remain in place, the politics of education in the U.S. have shifted dramatically. Meanwhile, far right parties in Europe are also pursuing culturally conservative and nationalist agendas in education, suggesting that an investigation of partisan agenda dynamics in education politics across U.S. and European contexts could be fruitful for future research.
The Policy Agenda of Social Scientific Research: Assessing Institutional Alignment, Funding Bias, and Disparities Across Disciplines
Abstract
This paper examines the policy topics explored by social scientists in peer reviewed journals since 1945. Following a study of policy process scholarship found in Policy Studies Journal (Fagan et al. 2024), we introduce a new dataset of more than 12 million abstracts from peer reviewed journals across all social science disciplines, coded for their policy content using the Comparative Agendas Project classification system. We leverage the “CAP Babel Machine” fine-tuned large language model and validate a sample of observations using human classification (Sebők et al. 2025). We explore how topics are distributed across disciplines, the role of funders in biasing the topics explored by academics, and the relationship between the academic agenda and the policy agendas of formal institutions and think tanks in the US.
Plenary - Book Projects
Welcome address
Of Promises and Power. How Political Influence Shapes Mandate Representation
Abstract
Do elections bind governments to what they promised, or do constraints and organized power fundamentally limit the scope for mandate representation? Of Promises and Power revisits this foundational question by analyzing governing executives’ pledges in France from 1995 to 2022. Electoral platforms are often read as mandates: voters choose, winners govern, and their promises should translate into policy. Yet governing is widely understood to be shaped by institutional, budgetary, and political constraints, implying that the imprint of platforms on policy is, at best, limited. This book shows that both views are partly right, but incomplete. Promises do matter: French executives deliver a substantial share of their pledges. But fulfilment is patterned by political influence, above all by the power of the groups that promises target. The book argues that campaigns incentivize leaders to promise broadly and ambitiously in order to signal responsiveness, while governing forces prioritization. Organized, resource-rich, and electorally pivotal groups are then better able to keep ‘their’ promises on the agenda, shape feasible policy designs, and insulate reforms against opposition. By contrast, pledges made to diffuse or weakly organized constituencies are more easily postponed, redefined, or symbolically satisfied. Combining large-N evidence with process-tracing case studies of emblematic reforms, the book rethinks mandate representation as a real but structurally unequal mechanism: elections open policy windows, but group power determines which promises ultimately move through them. It makes us rethink what elections can realistically hold governments to and highlights how unequal influence can be reproduced through the very mechanism meant to ensure democratic accountability.
What Matters: Public and Private Agendas in Cross-National Perspective
Abstract
This book develops a picture of the policy agendas of the public, interest groups, and the government across four countries—Germany, the Netherlands, the United Kingdom, and the United States. We draw together original data from nearly 400 interviews with interest groups, original public opinion surveys in each country, and analysis of national government budgets, legislation, and executive speeches by policy area using the coding scheme of the Comparative Agendas Project. We find that, in general, interest groups play a mostly benign role in setting governmental agendas, which they mostly react to rather than generate. However, we also uncover evidence that certain advantaged groups have an outsize ability to get their preferred policy proposals onto the agenda—and, perhaps more importantly, to keep their unwanted policies off.
Understanding Policy Changes. The Mixture Model of Politics
Abstract
Policy change is important, and a very large number of actors and events contribute to a given policy change. The list of contributions includes government change, public opinion, public debate, lobbying, expert advice, events, and media coverage. Many of these contributing factors are interdependent, their effect on policy is variable, and their contributions are often difficult to observe. This resembles the features of a complex system, and it calls for an approach to the study of policy change that takes this complexity seriously. Understanding Policy Changes: The Mixture Model of Politics offers a new theoretical and methodological approach to the study of policy changes. The approach draws inspiration from research on physical systems such as earthquakes and waves, and it utilizes mathematics to improve the precision of the theoretical arguments and the empirical applications. Theoretically, the book develops a model of politics that explains policy changes as a function of variation in the number of contributions. Methodologically, the book presents the comparative distributional method as a statistical tool to analyse policy change. Using case studies, and data on public spending across a large number of countries and issues, the book demonstrates how the mixture model accounts for general patterns in the distributions of policy changes and how institutional differences, particularly democratic institutions, are associated with variation in these distributions.
Thursday, 18 June
3.1 · Agenda Dynamics & Political Attention III
The Reflected Attention Gradients Index (RAGI): Toward an Integrated Strategy for Quantifying Attention to Public Policies
Abstract
This article addresses a reflexive conception of the institutional agenda as an expression – of variable density – of categories such as the capacity of policies regarding the effective enforcement of their rules over time, and the behavior of this enforcement in terms of (dis)continuities. Starting from the assumption that these categories are directly related to specific amounts of attention devoted to the respective issues and policies by decision-makers, and that such attention is already usually observed within different political systems through the behavior of isolated variables – such as budgetary allocations, biases in the discourse of political authorities, legislative activity, the content and volume of media coverage of issues and policies, public opinion surveys, political party priorities, and government programs, among others (Baumgartner; Jones; Wilkerson, 2011; Baumgartner; Green-Pedersen; Jones, 2006) – it is argued that this set of variables, in theory, delineates a dynamic and multifactorial system that both conditions and determines the resulting attention devoted – especially by government leaders – to each of the various policies over time. This system itself also constitutes a means of expression of these aggregates of attention. By aggregates of attention, we therefore designate the sum of expressions – individually measurable over time within each of the aforementioned variables – that can be interpreted as evidence describing the movement of a given policy within the institutional agenda. Thus, the paper proposes a strategy for the integrated quantification of these aggregates of attention through an initial exercise in modeling a synthetic index based on a subset of the aforementioned variables: the Reflected Attention Gradients Index (RAGI). The selection of the variables and respective weights that comprise the index is mainly based on that idea of agenda density related to policies’ rules enforcement through time, notably: (1) the occurrence and type of a policy’s formal authorization act; (2) the flow of ongoing administrative acts of preparation or implementation; (3) the policy’s own budget execution; (4) the policy’s budget execution relative to the federative entity’s overall budget; and (5) the occurrence of discourse related to the policy or issue. This study is presented as a complementary theoretical and methodological contribution to comparative agenda research.
REFERENCES
BAUMGARTNER, Frank R; GREEN-PEDERSEN, Christoffer E JONES, Bryan D. “Comparative Studies of Policy Agendas”. Journal of European Public Policy,13:7, 2006 (pp. 959-974).
BAUMGARTNER, Frank R; JONES, Bryan D. e WILKERSON, John D. “Comparative Studies of Policy Dynamics”. Comparative Political Studies, 44 (8), 2011 (pp. 947-972).
Is scientific expertise at the service of public policy inertia, while emotions act as drivers of change?
Abstract
Punctuated equilibrium theory predicts a certain degree of policy inertia, except at critical moments that give rise to rapid and profound change. In their work on punctuated equilibrium, Baumgartner and Jones argue that emotions play an important role in such changes. At the same time, in their work on the politics of information, these same scholars identify expertise as a factor contributing to policy inertia. Theoretical reflections on this topic have indeed tended to treat expertise and emotions separately. Expertise is assumed to encourage deliberative reasoning, whereas emotions are seen as more spontaneous and more directly rooted in individuals’ values. In a presentation at CAP last year, Montpetit suggested that expertise and emotion may in fact be more closely intertwined. Experts not only experience emotions themselves, but may also trigger emotions among decision-makers.
For CAP 2026, we propose an empirical test inspired by this idea. More specifically, we ask whether expertise and emotion may be closely associated during both periods of policy inertia and periods of policy change, while the emotions triggered by experts differ in nature across these two types of periods.
We therefore propose to compare periods of stability and periods of change in legislative activity in Quebec since the Second World War, distinguishing among the policy topics identified by CAP. Using classifiers (mDeverta-v3) trained to recognize CAP policy categories and references to scientific evidence in parliamentary debates at the sentence level, we will test the hypothesis that expertise is mobilized to trigger positive emotions during periods of inertia, whereas it is mobilized to generate negative emotions during periods of change. We argue that stability requires reassuring expertise, whereas change requires expertise that creates a sense of urgency. This paper will help clarify the role of expertise and emotions in our understanding of punctuated equilibria.
Substitute venues? Policy production beyond organizational responsibilities
Abstract
Ministerial bureaucracies are purposefully designed to aid governments in producing and designing public policies. Despite this, several issues that fall outside or in between the formal structures and tasks of ministries. Governments do of course still have to produce policy on such issues. Though some governments address this by organizational reform, other issues must be dealt with within the existing structures of the bureaucracy. The purpose of this article is to examine how such policies are assigned. Or rather, to what extent does the personal traits of cabinet ministers explain the production of policy bills outside of the ministry’s core tasks?
Producing bills outside of formal organizational structure may grant the cabinet minister increased leeway in policy design, influence on the institutionalization of emerging issues and serve as a way to signal competence. Several hypotheses based on the political experience of cabinet ministers are tested by examining the substantive policy content of parliamentary bills using the Comparative Agendas Project (CAP) codebook. Using a dataset of the policy responsibilities of ministries, all bills are then classified as either within or outside of the ministries core tasks. Several binomial regression models are run on a dataset containing all parliamentary bills presented to the Norwegian Storting, 1985-2024.
The Who, What, When and Where of Municipal Policy Attention in California Cities
Abstract
City governments are widely acknowledged to play an important role in the policymaking activities of the United States’ governing apparatus, frequently making important decisions about public service provision and serving as the most frequent point of contact for citizens. Despite this, the study of policymaking at the municipal level has long been hindered by the lack of a centralized source of municipal government activity. I introduce a new database of every city council agenda item from every incorporated city in the state of California, stretching back as far as records are made publicly available. This dataset includes agenda items from over 200,000 meetings, with each agenda item extracted from individual PDFs of each meeting record. Alongside the agenda item text, I provide the roll call vote for each agenda item when available as well as code each agenda item by substantive topic. Both the roll call data and agenda items are validated using alternative data sources. This new dataset is the most comprehensive source of data for any geographic region in the United States. As such, it opens up a wide variety of new opportunities to study the policymaking activities of city councils from a comparative perspective.
3.2 · Media, Communication & Public Opinion III
Public Attention & Interest Group Lobbying: A Time Series VAR Study of Citizen and Business Mobilization in the United States
Abstract
How do interest groups respond to changes in public attention over time? This paper examines the conditional mobilization of U.S. citizen and business interest groups using time series data across issues and years. I compile a dataset from 1998 to 2023 combining lobbying disclosure reports, MIP survey responses, and media coverage to track shifts in public salience and lobbying behavior across 20 issues. Using Vector Autoregression (VAR) modeling and Granger causality tests, I assess the temporal relationship between public attention and lobbying activity. I expect that increases in public attention Granger-cause increases in lobbying, particularly by citizen groups, but also business groups. This effect is likely strongest on highly visible issues such as healthcare, immigration, and gun policy. Group-level and institutional controls, including campaign contributions and congressional hearings, are incorporated to account for confounding factors. If verified, these findings would demonstrate that interest group lobbying is responsive to public salience and that citizen groups, in particular, follow public attention more closely than business groups. This research contributes to our understanding of dynamic interest group behavior and the role of public opinion in shaping organized advocacy.
When do politicians respond to interest group requests? A field experiment on politicians’ information processing
Abstract
Politicians constantly filter an overwhelming amount of request from citizens, stakeholders, and interest groups. Which asks do they attend to (/engage with)? This question has been central to much interest group research, and a burgeoning literature suggests that interest groups typically make private or public appeals to draw the attention of decision-makers. Findings on its impact on decision-makers’ attention are mixed. Furthermore, studies indicate that personal anecdotes evoke information processing more so than statistics. To move research closer to politicians’ actual acts of information processing, and to move research towards identifying the causal effect of appeals on attention, we leverage a novel, large-scale field experiment (pre-registered and ethically approved). We collaborate with a real interest group and measure in an email to 4,000 Danish politicians if they attend to its appeal. Such email reflects a common situation in the work of politicians, and politicians do not know that the email is part of a study. Using random allocation in a 2x2 factorial design (private-public, anecdotes-statistics), each email contains one of four types of appeals, and we estimate variation in the likelihood of attending to the appeal. The study helps to establish the micro-foundation of politicians’ information filtering and zooms in on an early but crucial stage of decision-making, namely who gets to decide which policy inputs politicians act on.
The agenda of political advisors in the parliament
Abstract
Political advisors became highly important in policymaking. They often play an important role in ministerial agenda-setting, as they are the ones pushing issues important to the minister onto the agenda of the cabinet or parliament. Existing research on political advisors focuses on their role within the ministry. This paper examines their interactions with the parliament and its members. Specifically, it seeks to examine the role of political advisors in the legislature’s agenda. The paper is based on a unique dataset of the appearances of political advisors before Israeli parliamentary committees from 2013 to 2024. It also builds on in-depth interviews with political advisors responsible for the relations between the executive and the parliament. Initial findings suggest differences among ministries in the number of appearances by political advisors on parliamentary committees. In addition, ministerial advisors’ appearances are often intended to signal to members of parliament the importance the minister attaches to the issue. Therefore, in many cases, the advisors attend the meeting but do not speak.
The Politics of Attention in U.S. Technology Regulation
Abstract
Policymakers face uncertainty, or lack of relevant information, about the nature of problems they are asked to address, the possible consequences of any alternatives they might select, and the possible consequences of agenda denial and inaction. Baumgartner and Jones (2014) posit that policymakers try to reduce their uncertainty in different ways depending on whether they are focused on problems (where they expand their information search) or alternatives (where they focus on expert information). Yet uncertainty is not the only condition policymakers face when regulating emerging technologies; they also must weigh risk and ambiguity, which can influence information search and prioritization. Policymakers are not only uncertain about how new and emerging technologies work but in some cases face competing risks and ambiguity over which problem dimensions are relevant. Emerging technologies can “disrupt” not only the fit between existing regulatory frameworks and new activity, but also jurisdictional arrangements and the allocation of authority. In this paper I argue that emerging technologies and their related policy problems can be placed into three categories based on how those technologies interact with economic sectors and governing arrangements, which then generate three different patterns of attention and information generation. Data from U.S. congressional hearings and Congressional Research Service reports illustrate the differences in attention dynamics across different types of technologies within multiple issue areas including energy, food and nutrition, and telecommunications.
From silence to storm: Media agenda dynamics of political failures
Abstract
This contribution investigates media attention dynamics in response to political failures, as scrutinizing politics is one of the media’s democratic core functions (Esser & Neuberger, 2018). Specifically, we build on the concept of media storms (Boydstun et al., 2014) to understand why attention surges abruptly, following punctuated equilibrium dynamics (Baumgartner & Jones, 2009). However, existing research often suffers from selection bias by focusing solely on successful storms. We argue that to truly understand attention thresholds, we must analyze the “silent” cases (i.e., issues that fail to gain traction). These cases can be considered “miscalibrated” based on Entman’s (2012) scandal theory. The study analyzes 474 cases of political failure in German-speaking Switzerland (April 2019 – April 2020). This period covers a federal election campaign followed by the onset of the COVID-19 pandemic. Cases were identified through a qualitative screening of 1,357 articles from three leading newspapers (NZZ, Blick, Der Bund) and range from minor administrative errors to potential systemic failures. We test contrasting hypotheses regarding competition. Regarding political competition, we hypothesize that elections and referendums amplify coverage of failure. Conversely, concerning issue competition, we expect the onset of COVID-19 to crowd out other issues. Finally, the paper assesses miscalibration (i.e., mismatch between newsworthiness and media attention). We hypothesize that issues with high newsworthiness are likely to be underreported (miscalibrated) if they lack personalization (i.e., are systemic) and/or are overshadowed by competing issues on the media agenda. Conversely, low-relevance issues are expected to trigger miscalibrated reporting if they are highly personalized and occur during periods of low issue competition, or if they are driven by political competition. By mapping these dynamics, this study contributes to a finer understanding of how news value and contextual factors interact to shape the political media agenda.
3.3 · Budgeting & Public Spending I
Budget Amendments as a Channel of Radical Party Influence
Abstract
Radical parties have changed electoral competition across Europe, but we know much less about how much they influence actual policy-making. While existing research shows that these parties affect elections and push mainstream parties to adjust their positions, it is unclear whether they shape concrete policy outcomes. This study measures whether and how radical parties influence fiscal policy through budget amendments in France, Sweden, and Denmark. Budget amendments are an important but often overlooked way for opposition parties to affect policy priorities. I analyze how many budget amendments radical parties propose, what issues they focus on, and how often their proposals are reflected in final budgets. Using text analysis, I compare radical party amendments with enacted budget bills to assess how much of their policy agenda is adopted. By comparing countries with different institutional rules and party systems, the study identifies the conditions under which radical parties can exert meaningful influence beyond elections. This contributes to research on legislative power, budget politics, and the policy effects of fragmented party systems in parliamentary democracies.
Ideology and Defense: An analysis of budgets and troop deployments in 28 European countries from 1995-2024
Abstract
The role of partisanship and ideology in defense spending and security policies is receiving increasing scholarly attention, especially since the Russian invasion of Ukraine. It is, hereby, often assumed that right-wing parties are in favor of a strong military and inclined to engage in military actions abroad more easily than left-wing governments. However, this assumption is largely based on anecdotal evidence and has seldom been examined systematically across a wide range of countries and over longer periods of time. Introducing a new dataset of defense spending and troop deployments for 28 EU Member States from 1995-2024, we examine whether the ideology of governments influence their defense policies, and how.
Budgeting Across Regimes: Two and a Half Centuries of Maltese Public Spending
Abstract
The theory of punctuated equilibrium (PET) was developed to explain patterns of stability and change in democratic policymaking: long periods of incrementalism interrupted by bursts of disproportionate adjustment. Over the past decade or so, several studies found evidence of budgetary punctuations in non-democratic regimes, indicating that PET’s logic travels across regime types and administrative systems, capturing a more general property of information processing and institutional friction. This paper contributes to this scholarship analyzing a unique historical dataset of Maltese public budgets covering roughly 250 years and several regime configurations – from the late confederal theocracy rule of the Knights of the Order of St John (ca. 1778) through British colonial administration (1800), independence and democratization (1964), and EU membership (2004), up to 2020. The dataset includes annual expenditure data by function of government, reconstructed from the ancient books recording the expenditures of the Knights, the Blue Books recording expenses under the colonial regime, and contemporary budget documents. It therefore spans multiple transitions in sovereignty and regime type, each associated with different modes of governance — from patrimonial rule to imperial bureaucracy to democratic accountability. The analysis first traces the evolution of the functions of government across these eras – from such categories as falconry, nobilities, and purchase of slaves during the Knights, to more modern functions during British rule, to contemporary COFOG categories. It then compares the statistical distribution of annual budget changes across these periods to test the extent to which punctuated dynamics persist – and how they vary – under very different historical, institutional and informational conditions.
Information Processing, Regime Change, and Institution-Building in Brazil (1964–2024)
Abstract
Scholars of the policy process have identified systematic differences in how authoritarian and democratic regimes process information. Existing research shows that policy change tends to be more efficient in democratic regimes than in authoritarian systems (Baumgartner et al. 2017; Chan and Fan 2020; Chan and Zhao 2016; Jones, Epp, and Baumgartner 2019; Lam and Chan 2015; Or 2019; Sebők and Berki 2018; Sebők, M. Balázs, and Molnár 2022; Van Den Dool and Li 2023). While there is meaningful variation among established democracies in terms of information-processing efficiency (Breunig and Koski 2006; Fagan, Jones, and Wlezien 2017; Jones et al. 2009; Robinson 2004), the gap between democratic and authoritarian regimes remains substantially larger. These differences are commonly attributed to the informational disadvantages that authoritarian regimes face relative to liberal democracies (Chan and Zhao 2016). Liberal democratic systems rely on institutional mechanisms that support problem detection, the allocation of governmental attention, and policy formulation, including a free press, freedoms of expression and association, the rule of law, and expertise embedded in relatively autonomous bureaucracies, universities, and think tanks. However, these mechanisms do not operate uniformly across democratic contexts. Transitional or weak democracies often lack the institutional and bureaucratic capacities required to sustain efficient information processing (Fagan 2023). Even so, comparative research consistently finds that patterns of budgetary change differ markedly between democratic and authoritarian regimes (Baumgartner et al. 2017; Fagan, Jones, and Wlezien 2017). This article focuses on the contrast between authoritarian and democratic periods in Brazil from 1964 to 2024. Brazil was governed by an authoritarian military regime from 1964 to 1985 and subsequently underwent a prolonged transition to democracy. Although the country became an electoral democracy in the late 1980s, the consolidation of key policymaking institutions unfolded gradually over subsequent decades. Using new data on federal budgetary allocations coded by policy content with the Comparative Agendas Project classification system, the article examines how regime change is associated with differences in information-processing capacity, as reflected in the distribution of budgetary changes over time.
Punctuated Equilibrium and Budgetary Dynamics in Latin America
Abstract
This paper presents the first systematic application of Punctuated Equilibrium Theory (PET) to the comparative analysis of public budgets in Latin America, as part of a broader research program examining the institutional determinants of budgetary punctuation across the region. Drawing on annual central government expenditure data from 14 countries over approximately three decades, classified under the COFOG framework, the analysis establishes that budget changes exhibit the leptokurtic distributions predicted by PET across all cases, regardless of regime stability or fiscal institutional quality. Sectoral variation follows theoretically consistent patterns: housing, economic affairs, and environmental protection display the highest punctuation rates, while education and public order remain comparatively insulated. Cross-national differences in punctuation frequency align with broader variation in fiscal governance, suggesting that institutional configurations shape both the frequency and intensity of budgetary change. These descriptive findings provide the empirical foundation for a research agenda that moves from distributional mapping toward causal explanation, engaging the questions of when, where, and why punctuation occurs under conditions of institutional heterogeneity. By applying PET to a diverse and underrepresented regional context, the paper responds to recent calls to globalize policy process theories and contributes to an empirically grounded comparative understanding of fiscal dynamics in the Global South.
4.1 · Agenda Dynamics & Political Attention IV
Human Judgment in Hybrid Approaches to Large-Scale Text Classification
Abstract
Automated text classification has made it possible to code large corpora at a scale far beyond what can be accomplished with traditional hand coding alone. At the same time, even highly sophisticated models struggle to fully capture the judgment calls and decision rules embedded in complex coding schemes such as the Comparative Agendas Project (CAP) codebook. Using a large corpus of bill summaries from the U.S. Congress, this paper examines how automated classification performs across topics and how targeted human input can improve those results. Rather than treating human and machine coding as substitutes, we employ a hybrid approach that deliberately combines the two. We show that focusing human coding efforts on the cases where automated models are least certain can substantially improve classification quality, reaching levels of reliability that are difficult to achieve with either approach on its own. Crucially, these gains do not require widespread hand coding: relatively modest amounts of targeted human input capture most of the achievable improvement. Together, these findings demonstrate how hybrid human-machine approaches can meaningfully improve the quality and reliability of large CAP-style datasets while illustrating how researchers can combine automation and expert judgment in ways that are both efficient and substantively credible.
Winnowing Washington Work: Individual and Institutional Determinants of Congress’s Substantive Agenda
Abstract
Congress must navigate numerous constraints, demands, and opportunities in determining which among the many bills drafted by members will receive favorable action. Beginning in the late 19th century, individual members assumed the leading role in bill sponsorship, drastically driving up the workload of its members and the committees who act as first movers on most legislation. How did committees identify which proposals merited attention, and what role did partisan and constituency considerations play in that process? We build on prior work on winnowing to develop theoretical expectations about committee activity and use original data on bills introduced in the House of Representatives between 1881 and 1931 to show the influence of factors like committee membership and partisan demand on bill reporting patterns. Our results illustrate the importance of sponsor cues and the varying degree to which partisan considerations impacted committee behavior. By looking at a period in which members’ sponsorship activities were wide-ranging, we illustrate the adaptiveness and responsiveness of legislative institutions to competing pressures in an era of rapid development.
Agenda Sensitivity and Political Information Processing in Comparative Perspective
Abstract
The process of agenda-setting lies at the heart of a rich scholarly tradition, pioneered by Baumgartner and Jones (2019, 2005, 1993), but with roots in earlier works (Cobb & Elder, 1971; Cook et al., 1983; Kingdon, 1984). A central tenet of this literature is that political agendas are the product of ongoing political struggles. The capacity of a political system to respond to new information is constrained by its ability to absorb and process external signals. This paper argues that the disproportionality in how political systems allocate attention—measured through the distributional characteristics of shifts in legislative activity—can serve as a key indicator of their responsiveness. Using comparative data on lawmaking across 25 democracies, we examine whether patterns of punctuated attention can capture differences in how political systems address emerging challenges. We propose a more systematic approach to measuring disproportionality, building on traditional metrics such as Shannon’s H and leptokurtosis. Preliminary findings reveal substantial variation among democratic regimes: some exhibit highly peaked (leptokurtic) distributions, suggesting concentrated and punctuated responses, while others display flatter distributions, indicating incremental adjustments. We explore whether this variation correlates with institutional features, including Lijphart’s executive-parties dimension, regime durability, and electoral system characteristics. While bivariate analyses suggest modest relationships—with more majoritarian systems tentatively linked to higher kurtosis—these results raise important questions about the robustness of these measures and their theoretical interpretation. To further test these dynamics, we analyze the COVID-19 crisis as a critical case, assessing whether disproportionality acts as an asset or a liability in the face of sudden, acute shocks. This study contributes to the literature by refining the measurement of agenda flexibility and its implications for political responsiveness. By linking distributional patterns to institutional structures, we aim to deepen our understanding of how political systems adapt—or fail to adapt—to rapid change.
Party, Politician, Issue and Voter Determinants of Policy Congruence
Abstract
Democratic representation is, amongst other determinants, driven by policy congruence. If voters and their representatives share policy preferences, chances increase that policies come about that reflect the popular will. Our paper addresses a number of shortcomings in the existing literature on policy congruence. Based on the ongoing work of the POLPOP consortium we leverage a dataset spanning 14 countries and numerous policy issues, and systematically compare levels of policy congruence across countries and issues. Further, our data give us the opportunity, due to the many parties that operate in these polities, to leverage party variation and check to what extent party features matter for congruence. Moreover, we draw on extensive surveying of ten thousands of citizens’ policy preferences. This permits us to dig deep into the question who the citizens are who share their party’s stance most, and least. Finally, our novel dataset not only contains parties’ policy stances but also, and this is particularly exceptional, the policy stances of hundreds of individual politicians. So, our data allow us to descend to the level of the individual politician and examine why some politicians are more aligned with their voters than others. In sum, our paper tests the drivers of policy congruence and examines whether party, politician, country, issue or voter characteristics matter most.
Legislative Divisiveness and The US Supreme Court’s Policy Agenda
Abstract
The. US Supreme Court is a national policymaker and the institution’s willingness to consider divisive policy issues – especially those generally outside the Court’s usual policy agenda appears to be constrained by external political context that changes over time. In order to understand the current Court’s willingness to consider divisive cases, we argue its important to understand the period of protectionism displayed prior to the consolidation of power with the conservative voting block. Using data measuring attention to policy issues from 1948-2018, we consider we find (1) that the Court’s agenda is heavily concentrated in small number of topic areas, (2) that increased agenda control is associated with more engagement on issues that engender significant conflict in the U.S. Congress, and (3) that in the post-1988 period, the Court’s willingness to place legislatively divisive issues on its agenda is depressed when congressional and Court preferences diverge.
4.2 · Parties, Elections & Representation II
Generational conflict in parliament: Do narrowly winning young politicians respond more to the issue concerns of the youth?
Abstract
Old politicians populate modern parliaments. Does that diminish political issue responsiveness to young people’s issue concerns? Despite much research on unequal representation, we lack an answer to this question. The question is tricky to answer because it is difficult to isolate the effect of age among the politicians that get elected. We approach causal identification by leveraging a regression discontinuity design on the sample of close winners to the European Parliament, 2004-2023. We focus on electoral competition only between pairs of candidates of opposite age (young vs old). We take advantage of the as-if-random assumption that these narrowly winning politicians were elected by chance, and therefore, the research design approximates an experiment where constituencies (young people) get randomly allocated to a young or old politician. We test if narrowly winning young politicians ask more questions in the European Parliament than narrowly winning old politicians on issues that concern young voters. Amidst a rapidly growing literature using RDDs to examine effects of politician characteristics, this study is among the first to examine age, issue representation, and study the European Parliament.
Executive Attention and Gendered Surrogacy: U.S. First Ladies in Focusing Events
Abstract
U.S. presidents have been called the “agenda setter-in-chief” because they receive more media coverage than any other single figure in American government, and therefore are better positioned to get their issue priorities on other institutions’ agendas. But presidents have limited attention and also must respond to national and world events. Presidents often use “surrogates” or stand-ins to help them expand their attention capacity, with the first lady (the president’s spouse) as an important such surrogate. U.S. first ladies typically are more popular with the public than the president and are seen as less partisan. In this paper we investigate contemporary first ladies as surrogates during potential focusing events. Delegating attention to the first lady may suggest the White House acknowledges the focusing events’ importance, but not to the level of a presidential priority. We analyze which potential focusing events from 1993 to 2024 saw the first lady make public remarks, and when and how often in the event’s timeline they were used as a surrogate. We further compare how U.S. first ladies and presidents talked about the same event. We find evidence of variation in the potential focusing events that first ladies speak about and the timing of their involvement, which suggests that some potential focusing events are considered worth the president’s time and others less so. Conversely, some events are considered appropriate for first ladies to speak about publicly and others are not. We also find important differences in the framing and narratives presidents and first ladies use to discuss the same potential focusing events.
The role of the opposition’s legislative committee chairmanship in agenda-setting
Abstract
Legislative committees play a significant role in the law-making process. However, especially in European parliamentary regimes, this role is merely formal as the majority of committee memberships are composed of governing party MPs. On the contrary, in most countries, at least a handful of legislative committees are led by opposition MPs. In our paper, we focus on the contradiction between the dominance of government parties and the leadership of an opposition party representative. Our research question is: What impact does an opposition committee chairman have on legislative activity? We investigate whether committee chairmen are efficient tools of opposition parties for agenda-setting by analysing the influence of the opposition on the government’s legislative agenda regarding the policy portfolio of legislative committees led by their MPs, in contrast to other policy fields. We also investigate whether committee chairmen are efficient tools of opposition parties for influencing law-making by analysing the success of opposition bill initiatives on the policy portfolio of the legislative committees led by their MPs, in contrast to other policy fields. We analyse the influence of opposition committee chairmen on the legislative agenda and the opposition bill’s success in the Central European Visegrád Group (Czechia, Hungary, Poland, and Slovakia) for the period after the democratic transition around 1990 to 2023.
Retreating from Politics? Union Local Activity in an Era of Union Decline
Abstract
The decline of organized labor in the United States is well-documented. As union density has declined, comparatively less attention has been given to the political organizing activities of labor unions. Leveraging data on expenditures of union locals on political activity, I hypothesize that overall political activity from labor unions in the United States is declining. I argue that local levels of union density are central to this relationship, and that declines in political organizing activity should be concentrated in states that have seen the largest drop-offs in union membership. I also assess the impact of state-level political changes in union activity, such as changes in partisan control of state governments and passage and repeal of right-to-work laws. The results speak to the importance of changes in union power in assessing the political effects of the labor movement.
4.3 · Policy Content & Specific Domains II
Mapping AI Governance Portfolios: A Scalable Comparative Method for Classifying Policy Targets and Instruments
Abstract
This paper applies a policy portfolio approach to examine how artificial intelligence (AI) is governed across a diverse set of countries. We build a cross-national dataset of AI-related policies and classify each document along two core dimensions: policy targets (the objectives and issue areas that policies seek to shape) and policy instruments (the regulatory, programmatic, and governance tools they deploy).
Methodologically, the paper contributes a scalable workflow for policy mapping that combines text analysis with generative AI (Large Language Models, LLMs) to support consistent classification across heterogeneous policy documents. The approach is designed to travel: it can be extended to other policy sectors and applied at different levels of government, from national to regional and local.
Empirically, the resulting portfolio perspective makes it possible to identify patterns of convergence and divergence in AI governance—showing not only what countries prioritize, but also how they pursue these priorities through distinct mixes of instruments. The dataset and classification scheme provide an empirical foundation for Comparative Agendas–style analyses of agenda dynamics in emerging technology policy, and set up future work on how different policy portfolios may shape trajectories of technological development and innovation.
Descriptive Representation in Bruxelles? Ana Analysis of the Activities of Members of the European Parliament with a Migratory Background.
Abstract
he attention dedicated to migration in the European Parliament (EP) has significantly increased over the years, reflecting both the assembly’s growing policy powers and the issue’s increasing salience for European public opinion. A fundamental question remains: Is the voice of migrants truly heard in the European representative assembly? It is widely argued that members of disadvantaged and discriminated minorities achieve better substantive representation when elected officials share their social background. This argument rests on the premise that parliamentarians with direct experience of minority discrimination are assumed to be more likely to champion the interests of their group, suggesting that descriptive representation fosters substantive representation. In the European context, first- and second-generation migrants face frequent discrimination and are notably underrepresented in political institutions. Despite this, the number of Members of the European Parliament (MEPs) with a migratory background has incrementally increased over the last two legislative periods. Leveraging an original dataset compiled within the EUQuest project, this research investigateswhether the rise in MEPs with a migratory background has translated into greater advocacy for migrants’ rights in the EP. The analysis examines committee assignments and the content of parliamentary questions during the 8th and 9th legislative periods (2014–2024), utilizing a mixed-methods design that combines quantitative analysis with qualitative explorations. The findings demonstrate that MEPs with a migratory origin are significantly more likely than their colleagues to be assigned to the committee addressing migration and discrimination issues, consequently making them more likely to scrutinise the European Commission on these topics. However, qualitative analysis reveals a nuanced reality: speaking about migration is not synonymous with speaking for migrants. Ultimately, partisanship and personal experiences emerge as powerful factors shaping the political stance of MEPs, even those with a migratory background.
Supplying Issue Attention Without Eroding the Brand: How Parties Emphasize Issues Across Demographic Groups
Abstract
Research on issue competition has traditionally studied party communication through channels delivering uniform messages to all voters. However, the rise of social media advertising platforms has introduced a fundamental change in that they allow parties to target different audiences when distributing ads. In response, we investigate whether individual parties present consistent or differentiated agendas to various demographic segments through advertising on Meta platforms. Using a new cross-national dataset of 121,974 issue-based advertisements from 68 parties in eight Western European and North American democracies, we show that even with the capacity for microtargeting, individual parties present largely similar issue profiles across demographic groups, reflecting systemic pressures toward brand consistency. At the same time, individual parties modulate their issue emphasis across age and gender groups, following systematic patterns across countries. This modulation follows an inverse U-shape by age: individual parties present their most distinct issue profiles to younger and older audiences, while presenting more typical emphases to middle-aged groups. Issue-level analyses reveal that parties emphasize issues aligned with life-cycle and gendered policy concerns.
Does Government Participation Change Party Policy Agendas in Authoritarian Settings?
Abstract
Government membership has been found to reshape parties’ policy agendas in democracies, conditional on the governing party’s incentives and capacities. Whether these patterns hold under authoritarian regimes is less clear. This article examines the question through a within-system comparison centred on Morocco, using an original dataset of 346 legislative proposals submitted by eight parliamentary groups across three legislatures (2002–2016). Drawing on three-way fixed-effects regressions at the party-year-domain level, we find that government participation does change policy agendas, but only under specific conditions. The effect is concentrated in governments led by a prime minister from a traditional, already co-opted opposition party; technocratic and new opposition-led governments produce no comparable adjustment. Parties that cycle between government and opposition display a symmetric pattern of expansion and contraction in the diversity of their policy agendas, indicating a reaction to institutional status. These findings suggest that the mechanisms linking partisan status to policy agendas are not confined to democratic settings, but their activation depends on conditions, such as how the government is led in the first place, that additional research needs to uncover.
5.1 · Polarization, Inequality & Democratic Challenges I
Forms of polarization
Abstract
Despite the vastly increasing scientific attention for political polarisation, effective conceptualisations and measures of different forms of polarisation in parliamentary records and news media are still lacking. Furthermore, there is no knowledge on how levels of polarization in politics and the media influence each other. Employing a longitudinal manual content analysis of parliamentary questions, debates and newspaper articles on climate change, this study aims to accurately assess the development of polarisation in the Dutch political and media debate on climate change between 2010-2021, analysing three distinct forms of polarisation: ideological, affective, and factual belief. We find that the levels of ideological polarisation have remained largely stable across this period, while affective and factual belief polarisation remain low, though slightly increasing in the later years. In the political realm, these increases in climate change polarisation are mostly attributed to the increased presence of ecological and populist radical right parties, while the increases in the media realm are mostly observed in popular newspapers. Time-series analyses between the political and media realms show that levels of ideological polarization in parliament influences ideological polarization in the media, while we observe the reverse for affective polarization. Here levels of affective polarization in the media influence the levels of affective polarization observed in parliament. This shows that polarization levels between realms are connected and should be studied in align with one another.
When Partisanship Silences Inequality: Polarization and the Erosion of Income-Group Priorities
Abstract
An evolving body of research demonstrates that public issue priorities vary across groups divided by both income and partisan affiliation. Income-based inequalities generate distinct sets of policy concerns, while partisan identities structure how citizens rank and interpret those concerns. Some recent scholarship has highlighted the growing role of partisanship in shaping policy priorities. At the same time, other work has explored how growing inequality influences policy priorities and affects political responsiveness. What remains less understood, however, is how these two forces interact: does polarization suppress the political expression of inequality in public priorities, or do economic divisions persist even as partisan conflict deepens? To address this question, I analyze four decades of public priorities in the United States (1981–2020) and in Israel (1977–2023), two democracies that have experienced both persistent socioeconomic inequality and sharp partisan polarization. Using comparable longitudinal data, I examine the policy agendas of groups divided by income levels alongside the partisan distribution of priorities. I assess both the overall diversity of priorities and the within-group variation over time, focusing on whether socioeconomic distinctions in issue salience persist when partisan divisions intensify. The findings demonstrate that, under acute polarization, partisan affiliation prevails over income-based inequality in shaping the public agenda. As polarization intensifies, differences in policy priorities across income groups decrease, while partisan diversity in those priorities narrows. This dynamic has important implications: it reduces the visibility of socioeconomic concerns in the political arena, constrains political parties’ ability to represent the full range of their voters’ priorities, and raises broader questions about the responsiveness of democratic institutions. By comparing two cases that combine enduring inequality with deepening polarization, the paper highlights how partisan conflict can silence the policy voice of income-based groups, thereby limiting the capacity of democracies to translate inequality into political representation.
Does Salience Hurt in the Polarized Age?
Abstract
Traditionally, advocates want to raise the salience of their issue area and problem to move it up the agenda of policymakers. But in a polarized age, high salience can backfire. Indeed, high public and media attention in the United States can now bring polarized framing and zero-sum conflict. Canonical findings from policy agendas research show that the public agenda can influence the policymaking agenda and help bring about policy change. But recent findings show that media attention can slow policymaking, that involvement by presidents and governors can polarize votes in legislatures and bring gridlock, and that policymakers succeed by working behind the scenes and avoiding the spotlight. And public commentators have noted that the “Secret Congress” is the main place we are still seeing policy advancement. Under close party competition and thick institutional veto points, the attention-centric playbook often misfires. Many durable policy advances emerge from low-salience venues, expert subsystems, and cross-pressured coalitions; efforts to elevate visibility can harden partisan lines and slow enactment. I investigate the mechanisms for the quiet path to successful policymaking and argue that policy process theory needs updates to reflect when moving a problem higher on the agenda can hurt, rather than help, advance policy. Despite operating under intense polarization and narrow majorities, recent congresses enacted significant legislation on infrastructure, semiconductor manufacturing, gun safety, electoral reform, and climate change. A steady stream of substantive policy changes—from raising the smoking age to phasing out hydrofluorocarbons—passed with overwhelming bipartisan support and minimal public contestation. Rather than shattering iron triangles that slow change, expanding the scope of conflict or moving a problem up the public agenda may now inhibit policy change that can advance within narrow communities or via bipartisan dealmaking out of the spotlight. High salience now triggers “teamsmanship”, activating partisan gatekeeping and reducing the policy space to a binary choice. Under these conditions, attention functions not as a catalyst for change, but as a freezing mechanism. By analyzing data from the Policy Agendas Project and legislative histories of recent bipartisan enactments, this paper demonstrates that successful policy entrepreneurs today invert the traditional logic. Rather than expanding scope, they aggressively narrow it. They seek to maintain low salience to prevent the activation of partisan identity, allowing the old governing wisdom of negotiation, logrolling, and technical adjustment to function within protected spaces. I synthesize evidence across congressional procedure, state policy diffusion, elite communication, and crisis legislating to show when salience helps and when it backfires. Catastrophic events can still accelerate legislating but what passes in those windows is usually drawn from proposals already specified, vetted, and coalition-tested before the emergency. Durability then turns on design: programs that generate constituencies, administrative routines, or other positive feedbacks are more likely to persist; others lapse or are re-written when attention recedes. I then seek to revise traditional theories of punctuated equilibrium and independent streams for a polarized age. I propose friendly amendments and scope conditions. I find that problems can reach the policymaking agenda through narrower channels, such as industry and lobbying-focused publications, research communities, state policymaker networks, and insider policy entrepreneurs. I ask how far the American example should go to change general cross-national theories of attention and policy adoption. There are US particularities but also broaden lessons for policymaking in a polarized age.
5.2 · Parties, Elections & Representation III
What Members Do and What They Say: Agenda Divergence in Congress
Abstract
Democratic accountability rests on a basic expectation: that voters have a meaningful sense of their representatives’ policy priorities. Yet members of Congress often priortize different issues in their policymaking activity than in their communications with constituents, presenting a skewed picture of their legislative agendas to the public. This paper examines when and why members emphasize different issues in internal and constituent-facing communications, and what this divergence reveals about the relationship between public-facing and legislative agendas. To capture members’ public-facing agendas, I analyze a large corpus of members’ e-newsletters, which reveal how legislators choose to frame their priorities to their constituents. I compare these agendas to those presented in members’ Dear Colleague letters, which reflect their efforts to build legislative coalitions and advance policy inside Congress. The analysis shows that agenda divergence is widespread but uneven: while some members present closely aligned agendas across venues, others emphasize substantially different issue priorities depending on their audience. Taken together, these findings show that divergence between public and policymaking agendas is a systematic feature of congressional communication, with important implications for democratic accountability.
Greatest Hits: How Parties Use Social Media to Test and Select their Election Promises
Abstract
Election promises are central to democratic representation, yet we know surprisingly little about how parties develop them. Existing accounts typically emphasize elite-driven processes occurring within party organizations. We argue instead that parties increasingly use social media as a venue to pre-test policy proposals outside election campaigns and to gauge public reactions before deciding which proposals to elevate to formal election promises. Drawing on data from parties’ Facebook communication in Denmark, the United Kingdom, and Austria, we analyze how user engagement with policy-related posts—measured by the volume and tone of responses—shapes parties’ subsequent campaign promises. We show that policy proposals receiving more engagement, and especially more positive feedback, are significantly more likely to be repeated as election promises. Given that a substantial share of election promises are ultimately fulfilled, these findings suggest that social media feedback play a key role for future policy-making.
Bring in the cavalry: Solving CAP minor topic coding with AI agent-systems
Abstract
While automated classification of Comparative Agendas Project (CAP) major topics has achieved research-grade performance using fine-tuned transformer models, the classification of minor topics—comprising over 200 policy subcategories—remains a significant methodological challenge. This article introduces a novel multi-agent system built on LangGraph StateGraph architecture to address this complexity gap. The system deploys an orchestrator agent that routes classification tasks to specialized task agents, including model fine-tune and validator agents, leveraging an ensemble of generative and non-generative large language models. The architecture uses shared-state management for context persistence and a lightweight intent classifier for cost-efficient query routing. Accessible via Slack via natural language commands, the system enables sophisticated classification for researchers without computational expertise. Preliminary results demonstrate that agent-based ensemble approaches outperform single-model solutions on fine-grained classification tasks, particularly for underrepresented policy subcategories, extending automated coding solutions to the granular level required for detailed agenda-setting analyses.
Mapping national regulation of local governments
Abstract
In recent years, a double trend has been observed in many countries: A growing regulatory burden along-side increasing political decentralization. We have limited knowledge about how these two trends interact. This paper will present the results of a systematic mapping of national policy obligations for local councils. More particularly, it will content code all Danish national legislation (1989-2025) aimed at local councils focusing on how hard it regulates local politics. Important indicators are whether the law stipulates something the local council can (more freedom) or must (less freedom) do as well as changes in legal intensity (i.e. whether obligations are stricter or looser than before). The paper’s systematic mapping across all policy areas contributes with new and important insights to our understanding of regulatory burdens in the context of intergovernmental relationships. Denmark is an informative case with many tasks delegated to the local level of government, and with very good access to valid and systematic data on national regulation.
5.3 · Policy Content & Specific Domains III
Executive Agenda-Setting in Subnational Legislatures: Evidence from U.S. State Politics
Abstract
This paper examines how executive agenda-setting operates within subnational legislatures, using U.S. Illinois state politics as a comparative lens for understanding policymaking in decentralized systems. It analyzes whether governors can shape legislative priorities through major agenda-setting speeches, drawing on newly assembled data linking the policy content of State of the State addresses to all bills introduced in the Illinois General Assembly from 1996–2025. Bills and speeches are coded by the Illinois Policy Agendas Project, enabling systematic comparison across policy domains and over time. The research design compares the distribution of legislative topics introduced before and after each annual speech, holding the institutional context constant within a legislative session. We hypothesize that gubernatorial emphasis produces measurable shifts in legislative attention toward the policy areas highlighted in the speech. By situating executive and legislative dynamics at the subnational level, this study contributes to broader international debates on agenda control, political communication, and the capacity of executives to influence policy priorities in multi-level governance systems.
The emotionality of politicians’ social media posts as a predictor of parliamentary attention
Abstract
Political attention is scarce. Parliaments can address only so many issues at a time. Which issues make it onto the agenda and which ones are ignored? We shed new light on this perennial CAP question by focusing on individual politicians’ emotions as a driver of issue attention. Our assumption is straightforward: humans are hardwired to pay attention to things that arouse them emotionally, and as politicians are human, we hypothesize that will be more inclined to act upon issues about which they are more emotional. To test the hypothesis, we rely on a longitudinal dataset of Facebook posts by Belgian politicians. We coded the topic (CAP) of each post, as well as occurrence of emotion (anger, fear, hope,…) in the posts. We use time-series analyses (2010-2022) to test whether the degree of emotionality surrounding an issue at a certain point in time predicts the subsequent level of parliamentary attention for the issue—in parliamentary questions as well as bills. Doing so, we contribute to the growing body of work addressing the conditionality of agenda-setting, scrutinizing under which conditions politicians act on issues that are publicly debated.
The Popes’ Policy Priorities
Abstract
In this paper, I examine if the policy priorities of Pope Francis are different from his predecessors. I tease out the public policy priorities of the popes since the early 1960s through their annual messages to the Holy See’s diplomatic corps. When considered in their totality, I find that Francis’s addresses did not differ that much from those of John XXIII, Paul VI, John Paul II, and Benedict XVI. After considering the overall policy content of their addresses, I perform two finer-grained analyses in which Francis’s distinctive voice becomes clearer.
“Up and Down with the Ecology”: salience and the transformation of public opinion
Abstract
Abstract
This paper adopts a longitudinal perspective to assess changes in public opinion on environmental issues in Western Europe. It suggests that (1) the salience of environmental issues and the support for environmental protection have generally increased over time, albeit with fluctuations following the attention cycle; (2) both salience and support have become increasingly polarized along ideological lines. In other words, we consider that the sequence opened after the Covid-19 pandemic is a moderate downward shift following a historical upward trend that has too quickly been labelled an “ecological backlash”. These hypotheses are tested using a combination of survey data from the Eurobarometer, the European Values Survey, the European Social Survey, and the International Social Survey Program. The analysis relies on a classical measure of environmental issue salience, as well as an original environmental mood.
Proposal details
Puzzle While the use of opinion polls has long been criticized, along with the very concept of public opinion (Bourdieu 1973), the amount of evidence showing that public opinion influences policy-making is piling up (Burstein 2010; Erikson, MacKuen and Stimson 2002; Peters 2021; Soroka and Wlezien 2009). A clear understanding of how and why public opinion regarding the environment varies over time and space is key if one wants to understand advances and gridlocks in environmental policy-making – in the broad sense of policies that seek to keep political economies within the planetary boundaries (O’Neill et al. 2018). Moreover, the strong evidence for the existence of a so-called ecological backlash - or the absence thereof – are still sparse. Since Anthony Down’s seminal work on the cyclical salience of environmental issues, we know that public opinion goes “up and down with ecology” (Downs 1972). Long-term and recent developments confirm this tendency: while the attention to the environment started rising in the 1970s and 1980s (Dunlap 1991; Jacob 1999), it was followed by a decline in the 1990s (Sainteny 2000). Similarly, the peak of attention to the rejuvenated climate movement in the late 2010s (de Moor et al. 2020) was quick overpassed by issues related to the Covid-19 pandemic, and the economic consequences of the Ukrainian War leading to the “backlash” that is at the core of this Section Thématique. Precise empirical assessment of these long term trends are rare. Longitudinal studies by Bozonnet and Halpern on France (2013) or Benedetta and Vincenzo on Italy (2019) express cyclic trends: a fluctuating – alternating ups and downs in a pseudo-sinusoidal form – but increasing level of concern. However, little has been done to compare trends in public opinion over space and between countries.
Literature and Research Question The first strand of literature that this project builds upon deals with environmental attitudes in the public, a multidisciplinary field of study bringing together several sciences: (social) psychology, sociology, political science and even political economy. Environmental attitudes can be defined “both as the intensity of positive or negative affect[s] about a particular environmental topic and as a hierarchical attitude system that connects and organizes more specific attitudes about a range of environmental topics” (Cruz et Manata 2020, 184). In other words, environmental attitudes are about positions on environmental issues as much as about the priority given to the environment. This paper’s research question is the following: How has public opinion regarding the environment changed over time, and what are the country-specific drivers of change? Recent works have shown that public opinion in Italy and Spain have moved in parallel over the last twenty years, following a non-linear but generally upward trend (Lorenzini, Monsch and Rosset 2021). This trajectory can be linked to the well-known thesis that developed economies have become more and more concerned about environmental preservation due to more pronounced post-materialist values (Inglehart 1995). This thesis that has been discussed and criticized multiple times ; first, by studies showing that developing country respondents are equally concerned about environmental preservation (Brechin and Bhandari 2011) ; second, by studies proving that the public in countries with high levels of national wealth seems less willing to pay for environmental protection than in countries with lower levels (Gelissen 2007). Indeed, national economic contexts (growth, unemployment) are often put forward to explain differences and variations in environmental support. While several studies tend to show that the level of environmental concerns is higher in a context of economic growth (Kenny 2020; Scruggs and Benegal 2012), other articles point to the lack of correlation over certain time periods (Bozonnet et Halpern 2013; Shwom et al. 2015). Amongst the other factors that might also explain cross-country and time variations, the influence of media coverage is fairly consensual (Brouard et al. 2013; Gamson and Modigliani 1989; Spence et al. 2010), even though it has been discussed in the French case (Bozonnet and Halpern, 2013). Finally, the structure of the party system and the nature of political debates in electoral campaigns might play a role, just like they play a role as regards law and order and immigration issues (Hatton 2017), but, to our knowledge, this hypothesis has not been tested regarding environmental issues. Variance in ideological positions on other issues also seems to play a great role in explaining variations in positions on the environment. In the United States, in a context of increasing political polarization of environmental issues, respondents’ political ideology is identified as the most important explanatory factor of the level of environmental concern by several studies (Dunlap, McCright and Yarosh 2016; McCright and Dunlap 2011; Shwom et al. 2015). These results need to be confirmed in the European case, where left-wing sensitivity seems to have a positive correlation with pro-environmental behavior in the UK (Fowler 2015) or in France (Gougou and Persico 2019).
Data and Methods Environmental concerns and public opinion about the environment (or climate) are hard to grasp in a comparative and longitudinal way. This is so for several reasons. First, questions arise about the fact that these concerns form a single or multidimensional concept or whether environmental concerns stem from a “higher-order worldview” (Bernstein 2020). Moreover, numerous measurement scales have been published since the 1970s – the most popular being the New Ecological Paradigm (Dunlap 2008; Dunlap and Liere 1978). Yet, many suffer from several criticisms, based on their lack of distinction between beliefs, worldviews, behaviors, and attitudes, their lack of standardization, including sample size, question order, and precise wording of questions. Finally, the lack of surveys asking the same questions over a long period of time and across countries makes it impossible to compare public opinion on the environment at a larger scale: data and questionnaire’s discontinuity lead to many problems in the comparative study of environmental attitudes (Prakash and Bernauer 2020). Cruz and Manata show, however, that survey-based statistical constructs are valid and reproducible to assess a latent environmental dimension (Cruz and Manata 2020), and Stimson (1999) developed the policy mood tool to precisely tackle shortcomings that scholars of environmental concerns have faced. The “policy mood” can be defined as an aggregate measure of representative surveys assessing the public’s latent preferences over any given policy issue (Page et Shapiro 1983; Stimson 1999 for seminal studies). The mood was created to cope with the absence of surveys repeated over time with the exact same question-items. It uses a dyadic ratio algorithm to tackle data and questions’ discontinuity (Atkinson et al. 2021). Stimson and colleagues first used a policy mood to assess the position of the American public on the Conservative vs. Liberal conflict dimension (Erikson, MacKuen and Stimson 2002). This work was then reproduced in France to measure public opinion on the cultural and economical dimension of competition (Stimson, Tiberj and Thiébaut 2010) or the opinion mood vis-à-vis nuclear electricity (Brouard et al. 2013). A comparative opinion mood on European integration was also recently computed (Guinaudeau and Schnatterer 2019). The paper will draw from several comparative surveys of public opinion covering 12 West European countries: Eurobarometers, 96 waves (1973-2022); European Value Survey, 5 waves (1981-2017); International Social Science Programme, 7 waves (2000-2020); European Social Survey, 10 waves (2002-2020), International Observatory on Climate and Opinion: 3 waves (2020-2022). The project will also make a specific focus on France to 1) assess the quality of the comparative data with a higher number of measure points and 2) be more specific about the French case. The longitudinal survey data that will be used in the French case are the following: CREDOC, 32 waves (1979 - 2020); INSEE: 25 waves (1997 - 2019); DREES: 8 waves (2000 - 2019); French electoral study: 4 waves (2007 - 2022); EDF-Baromètre développement durable : 5 waves (2007-2012); ELIPSS: 24 waves (2013 - 2020).
Plenary — Keynote & Roundtable
Keynote lecture
Roundtable in honor of Bryan Jones
Friday, 19 June
6.1 · Polarization, Inequality & Democratic Challenges II
Polarization as an Agenda Cue: Issue Salience, Media Framing, and Electoral Choice in the 2025 Dutch Election
Abstract
Polarization has become one of the central concerns in contemporary democracies, both in public debate and in academic research. Citizens frequently express worries that politics is becoming increasingly polarized, yet it remains unclear whether and how this perceived polarization actually shapes political priorities and electoral behavior. Building on agenda-setting and priming theories, this paper examines whether issues that are presented as polarized in (social) media coverage become more salient to voters and exert a stronger influence on their vote choice.
We argue that polarization operates as a powerful agenda cue: when media portray political issues as sites of conflict between opposing camps, these issues are not only more visible but also cognitively and affectively charged, making them more likely to be used as criteria in political judgment. In this way, polarized framing may amplify the agenda-setting effects of issue coverage and strengthen the priming of specific considerations in electoral decision-making.
Empirically, we draw on a five-wave panel survey conducted in the run-up to the 2025 Dutch parliamentary elections N=3,000 in wave 1 and 2,461 in wave 5). The survey includes repeated measures of vote intentions and considerations, closed and open-ended most important problem (MIP) questions, perceptions of polarization, and detailed indicators of individual media exposure. Additionally, a substantial amount of respondents provided insight in their political (social) media exposure through data donations. These data are combined with a large-scale content analysis of news and social media content to which respondents were exposed, using large language models to detect issue attention and the degree to which issues are framed as polarized. Through linkage analysis, we connect individual-level media diets to the polarization and salience of the issues that citizens encounter.
We test whether exposure to polarized issue coverage increases (a) the likelihood that these issues are named as most important and (b) their weight in explaining vote choice over time. By integrating agenda-setting, priming, and polarization research in a longitudinal, exposure-based design, this study provides new insight into how polarization in media agendas reshapes the structure of political competition and citizens’ electoral decision-making.
Who gets the say in the coalition agreement? Local programs and municipal coalition agreements in the Netherlands
Abstract
Under what conditions do parties negotiating to form a coalition government get an agreement close to their preferences? Coalition agreements form a key feature of coalition governments. Most research into coalition agreements focuses on the national level. We apply existing insights to the Dutch local context. We test party size and issue salience as explanatory factors for policy payoffs. We examine three outcomes: the distribution of priorities in the coalition agreement, its overall left-right position and its position on twenty-one issue-specific dimensions. We look at more than 3,000 parties negotiating in nearly 1,000 Dutch municipality*election dyads between 2014 and 2022. State-of-the-art quantitative text analysis methods, based on LLMs, show that party size is correlated with getting the agreement closer to a party’s desired general position and priorities. Moreover, parties who prioritise an issue get a policy outcome closer to their desired position on that issue.
A tale of two neighbors: Assessing threat perception through in political narratives
Abstract
While existing research tends to rely on objective metrics, such as geographic proximity or spending of adversaries, to assess threats—a key determinant of military spending—these approaches often overlook the subjective perceptions that shape policy decisions. While Threat Perception has been acknowledged as a key factor shaping spending behavior, previous work has not fully considered how narratives, as collective expressions of significant events, contribute to the said perception of the “Threat”. This paper introduces a framework that incorporates narrative analysis for the modelling of threat perception. The study employs desk research to review and synthesize existing theories, followed by a discussion of potential methodologies for incorporating narrative data into the modelling of threat perception by leveraging recent statistical techniques from the Natural Language Processing (NLP) and Large Language Model (LLM) subfields in computer science to quantify these narratives through a case study between Belgium and the Netherlands.
Policy implications of the Green Backlash for the EU Green Deal: comparative analysis of agriculture, consumer protection and transport policies
Abstract
After gaining unprecedented momentum towards the end of the 2010s, the green transition is now facing a growing polarization and contestation across different regions. This phenomenon has been conceptualized as a “green backlash” (Bosetti et al., 2024 ; Patterson et al., 2025). Despite its timely relevance, our literature review (Mandelli & Halpern, forth.) shows that the notion is useful for describing a process of conflictual politicization directed against a green policy. Yet the concept remains underdeveloped, and much more theoretical and empirical work is still needed to define its boundaries. This is particularly the case when it comes to analysing the policy implications of Green backlash, that is how this socio-political phenomenon effectively shapes (enables, constrains) green policy developments, resulting in their delaying, softening or dismantling.
In this paper, we propose focusing on the European Green Deal Strategy, which was launched by the European Commission in 2018 to achieve climate neutrality by 2050. This ambitious agenda built on the global momentum around the climate emergency and consisted of both cross-cutting goals as well as a wide range of issue-specific policy initiatives (Eckert, 2021 ; Hohmeyer et al., 2022). Five years later, in a radically different geopolitical context, there is clear evidence of such green backlash shaping the delivery and further development of the EU green deal agenda. Yet EU parliamentary pushback only seems to account for some of the developments underway, together with changes in the international political outlook, interest groups mobilizations and levels of consensus within the EU Commission.
Drawing on the literature on comparative policy analysis (Fontaine & Peters, 2024) and public policy research public policy change (Hall, 1993 ; Howlett & Cashore, 2009; Mahoney & Thelen, 2009), we argue that the green backlash does not emerge uniformly, nor does it produce the same policy outcomes everywhere. Rather, we assume that some variations are to be expected across policy issues, depending on levels of (de-)politicization (Pierson, 2009 ; Fawcett et al., 2017) and pre-existing policy dynamics. More specifically, we propose to re-analyse datasets from our own previous research on the deployment of the EU Green Deal in three different policy areas (common agriculture, consumer protection and transport). In each case, these qualitative datasets consist of a document analysis of major policy strategies between 1990 and 2025 as well as some 8-10 interviews with EU policymakers, non-governmental organizations, business groups representatives and experts. This dataset allows to 1) account for the emergence of backlash in each sector (when, who and why), 2) examine its implications on within policy dynamics (problem framing, choice of policy instruments, logics of collective action) and in terms of policy results (the direct result of the decision-making process, i.e., the adoption / rejection of specific programmes, measures, laws or regulations).
In all three cases, policy developments were impacted by the Green backlash but this impact is far from being homogenous. Preliminary findings from the comparative analysis suggest that variations depend on drivers and agents of politicization on the one hand, and the stage of public policy developments on the other hand. Following large-scale farmers’ protests across the EU in early 2024, the EU significantly scaled back its climate ambitions in the common agricultural policy. In the case of consumer protection, a first Directive to empower consumers for the Green transition was adopted with the support of the industry in February 2024 but the EU Commission announced the withdrawal of the proposed Green claims Directive a few months later. In the case of transport, the EU Commission has faced growing pressure from member states and the automobile industry to delay or suspend the ban on fossil fuels vehicles by 2035. Drawing on these conclusions, we hope to develop a comparative analysis framework within and cross regional.
6.2 · Parties, Elections & Representation IV
What Matters Where? Urban–Rural Differences in Policy Priorities
Abstract
A growing number of studies examine how citizens in rural and urban areas differ in their preferences and attitudes on several political topics. For instance, Huijsmans et al. (2021) find that urban residents have become more cosmopolitan over the last decades, whereas rural residents have remained the same. However, we know little about whether urban and rural populations differ in the policy problems they view as most important, with current studies circumscribed to the US case (Yildirim and Solvig, 2025). This paper offers a threefold contribution to this nascent literature. First, we explore the rural-urban divide regarding the most important problem in the European context. Specifically, we use more than 300 quarterly surveys containing the most important problem question from the Spanish Centro de Investigaciones Sociológicas between 1996 and 2025, which have been merged and CAP-coded. Second, unlike previous studies that focus on a dichotomous classification of urban and rural areas, we examine a more fine-grained classification that considers local communities of different sizes. Third, we explore whether the previous results are driven by compositional effects arising from the different occupational composition of rural and urban areas in Spain, with the latter more likely to be occupied by sociocultural professionals with distinct policy preferences.
The dynamics of political attention in regime transitions: punctuations and agenda diversity in Colombia and Chile (1958–2025).
Abstract
Drawing on Punctuated Equilibrium Theory (PET)—which posits that policy agendas are characterised by prolonged stability interspersed with abrupt, rare discontinuities—this paper extends the framework to the Global South. While PET evidence is largely concentrated in the Global North, knowledge of these patterns in middle-income Latin American countries remains insufficient. This study analyses a 67-year period (1958–2025) in Chile and Colombia to characterise policy change dynamics and the diversity of agendas across regime transitions, seeking to understand how their particular institutional contexts affect policy stability. The analysis employs a statistical approach utilizing primary indicators coded according to the Comparative Agendas Project (CAP): Enacted Laws (1958–2025), representing the decision-making agenda, and Presidential Programmatic Agendas (Speeches and Messages, 1990–2025), representing the programmatic agenda. Volatility is measured using L-Kurtosis, confirming the General Punctuation Hypothesis (H1) across all analyzed streams (LK > 0.123). Supporting the Informational Advantage Hypothesis (H4), punctuation intensity was consistently higher in less democratic/centralized periods, such as Chile’s dictatorship (LK = 0.277) compared to the democratic era (LK = 0.261), and Colombia’s pre-1990 period (LK = 0.200) versus the post-1990 constitutional era (LK = 0.192). Paradoxically, the findings challenge the Progressive Friction Hypothesis (H3), revealing an inverted pattern where symbolic executive agendas exhibit significantly higher volatility (Chile Speeches LK = 0.53; Colombia Messages LK = 0.42) than legislative outputs. Furthermore, the Post-transition Diversity Hypothesis (H5) is supported, as Shannon H entropy scores increased in both nations following democratisation (Chile: 2.46 to 2.68; Colombia: 2.41 to 2.61). These results demonstrate how institutional opening and improved informational feedback in Latin American transitions facilitate a broader policy scope while moderating extreme punctuations.
How (statistical) polling information and (episodic) emails from individual constituents jointly influence politicians’ personal opinions, public opinion perceptions, and policy intentions
Abstract
In this paper, we investigate how elected representatives deal with different types of public opinion information. More specifically, we focus on how emails from individual constituents skew representatives’ interpretation of representative polling data. What happens when an opinion poll indicates a majority of people supports a policy proposal, while an individual citizen explicitly opposes the proposal in a personal, emotional message to the representative—how do politicians navigate this tension? To what extent do these signals influence their own opinions? How do they contribute to the politicians’ perception of public opinion more broadly? And how do these signals influence politicians’ policy intentions? To study these questions, we rely on one survey experiment with national and regional politicians in fourteen countries (Study 1) and one with local politicians in Flanders (Study 2) (N = +/- 1000 per study). We present all politicians with polling information about a concrete policy proposal (control group + treatment groups). Some politicians also receive an email from an individual constituent (treatment groups only). The results show that ‘episodic information’ from the individual e-mail influences politicians’ personal opinion on the matter, skews their estimates of public opinion in general, and affects their policy intentions.
Who politicizes the ’left behind’? Party competition and the return of the center-periphery cleavage in European politics
Abstract
Schools close, house prices stall, and the local pub closes. Across Europe, rural areas a falling behind metropolitan areas, and this is transforming the political landscape and entering the center of political debate, party competition and policy reforms. A burgeoning literature overwhelmingly shows that the losers in the periphery turn to populist, far-right, anti-elite parties to get sympathy and voice their frustration with the status quo. In comparison, scholars of party competition, parliaments, and political agenda-setting remain silent on the periphery’s way into political conflict. We provide a first systematic study of the inside of the politicization of the ‘left behind’. Who politicizes the geographical outskirts, and how? Our endeavor relies on the tiny and therefore hard case of Denmark where we have identified geographical references – cities – and classified the issue content in each question to the minister in a 20-year period since 2003 and matched these granular data with detailed biographical information on the politicians and the sociodemographic ‘misery’ of the mentioned city. Our study has important implications for our understanding of modern European politics and the major lines of political contestation.
6.3 · Budgeting & Public Spending II
Outsourced Surveillance: How Congress Subsidizes the Costs of Policy Oversight
Abstract
The canonical model of oversight argues that police-patrol forms of oversight should be rare because it requires committees to invest substantial time and resources into discovering whether agencies are complying with legislative intent. Yet, congressional committees regularly engage in policy oriented oversight. We argue that committees expand their capacity to perform oversight by subsidizing investigation costs through external actors. Specifically, when committees suspect an agency has violated legislative intent, they request reports from external auditors to conduct investigations on their behalf, then use these findings to determine whether formal oversight hearings are warranted. To test this theory, we examine the role of the Government Accountability Office (GAO), which is an independent and nonpartisan investigative agency that works on behalf of Congress. Analyzing almost 60,000 recommendations across 51,000 CAP-coded GAO reports published over the past 100 years alongside testimony from federal agencies at committee oversight hearings, we show that agencies targeted by GAO recommendations were significantly more likely to be the subject of committee oversight hearings in the following year. Additional results examining the origin of GAO report requests confirm the strategic mechanism, showing that reports are primarily requested by members of the congressional committee with oversight authority over each agency. These findings help explain the prevalence of committee oversight in Congress.
“You Can’t Always Get What You Want”. Budget amendments and the balance of power in Europe and Brazil
Abstract
The budget process is a fundamental mechanism for the quality of democratic policy-making and a litmus test for assessing the balance of power between the executive and the legislature. Recent scholarship highlights the decline of legislatures in Europe, now driven by external factors such as supranational governance, technocratic influence, policy complexity, and crises, alongside traditional executive dominance. In other regions of the world, such as Brazil, however, parliaments have increasingly asserted their power, representing a break with the local historical pattern of executive dominance. Budget amendments constitute a fundamental analytical tool for investigating these dynamics as their proposal is influenced by institutional settings, procedural rules, and also political factors. Yet, despite their relevance, they have received limited scholarly attention, even among experts of budgetary policy. Most existing work originates in legal scholarship and focuses on parliamentary rules of procedure and the institutional framework. Against this background, this preliminary study aims to compare several countries in Europe and beyond in order to: (1) build a comprehensive dataset on budget amendments as a common basis for comparative analysis; (2) analyze trends in budget amendment proposals over time within each country, examine how government and opposition parties use amendments, and compare these patterns cross-nationally; (3) assess the impact of institutional and political variables on the volume of budget amendment proposals. We focus on the following cases: Brazil, France, Italy, and Spain, which provide substantial variation across several dimensions (e.g. geographical, political, institutional, and economic). As policy scholars, we consider it essential to shed light on the complex dynamics of the budgetary amendment process by comparing it across countries and over time. In a context marked by growing distrust in parliaments, shifts in the balance of power between the executive and legislative branches, and increasing difficulties faced by governments in budget design and approval, such an analysis may help scholars and policy-makers develop prescriptions for more rational, less erratic, and ultimately more effective national budgetary policies.
Issue Ownership in the Budget Arena: How Opposition Parties Allocate Amendment Effort
Abstract
Budget amendments are proposed every year, yet we know surprisingly little about how opposition parties use them strategically. I argue that opposition parties do not spread their proposals evenly across policy areas. Instead, they concentrate on the issues they own — those where voters already trust them and associate them with distinctive priorities. However, this concentration is not fixed: parties propose more amendments on issues they own when public concern about those issues rises. I test these arguments using an original dataset of budget amendments from France, Sweden, and the Netherlands, translated into English and coded according to the Comparative Agenda Project coding scheme. The study contributes to research on issue ownership by showing that ownership dynamics influence how parties allocate amendment effort in the parliamentary budget process. It also speaks to the legislative behavior literature by arguing that electoral incentives drive opposition amendment strategies.
Exploring the Mountains of Money
Abstract
What characterises Norwegian budget policy over the past 25 years? Public expenditure in the Norwegian Fiscal Budget is a recurring topic of debate in public discourse, and one that politicians, bureaucrats, and academics alike express concern about. Despite its continuous relevance and increasing public attention, there are still few studies that systematically examine long-term developments in budget policy in Norway. Using an extensive dataset comprising budget figures from 1999 to 2024, this article demonstrates (1) how the structure of the Fiscal Budget has changed over time, and (2) what characterises the distribution of changes in appropriations within the Fiscal Budget. I show that the Fiscal Budget, as a policy agenda, has become smaller and more concentrated around fewer policy areas. Furthermore, I show that Norwegian budget policy—much like other countries—is characterised by extended periods of stability interrupted by disjoint and abrupt change.
Agenda Alignment, Executive–Legislative Rifts, and Change in Public Spending in Argentina
Abstract
We present the first results from the Argentina node of the Comparative Agendas Project, combining coded data at the national level on executed expenditures, budgets, legislation and presidential decrees between 1995 and 2025. We examine the degree of alignment and mismatch between executive and legislative agendas and assess whether these rifts are associated with greater year-to-year change in actual expenditures, rather than in approved budgets. By focusing on execution rather than budget approval, the project speaks directly to debates on executive autonomy in Latin America, assessing whether presidents can translate agenda control into substantive reallocations of public spending even in the presence of legislative divergence.
Closing Plenary
New Country Project: Agenda-setting in Argentina
Concluding remarks and perspectives for CAP