POPPRO. The Aarhus University Research Foundation (AUFF) has generously allowed me to initiate a project on how political parties prioritize societal problems such as rising unemployment or crime (Grant no. 33693). The project runs from September 2020 to August 2023, and the 2,000,000 DKR (£275,000) offers me the opportunity to study parties' problem responsiveness together with Roman Senninger, who the project funds.
Publications. |
|
- Kristensen, Thomas, Mortensen, Peter, Green-Pedersen, Christoffer, and Henrik Bech Seeberg (2022). ‘The policy agenda effects of problem indicators. A comparative study in seven countries’, Journal of Public Policy.
- Kristensen, Thomas, Green-Pedersen, Christoffer, Mortensen, Peter, and Henrik Bech Seeberg (2022). 'Avoiding or Engaging Problems? Problem Indicators, Party Preferences and Issue Competition', Journal of European Public Policy.
- Seeberg, Henrik Bech (2021). ‘Avoidance and Engagement: Do Societal Problems Fuel Political Parties’ Issue Overlap?', Party Politics. download LINK
- Borghetto, Enrico, Bevan, Shaun, and Henrik Bech Seeberg (2023). 'Do Parties Respond to Problems? A Comparative Study of Parliamentary Questions Across Multiple Countries', Journal of European Public Policy.
Work in progress
- Senninger, Roman and Henrik Bech Seeberg and (2022). 'What Makes Politicians Attend to Societal Problems? Evidence from a Field Experiment and Interviews with Party Candidates' (in review). LINK
- Tromborg, Mathias, Senninger, Roman and Henrik Bech Seeberg (2023). 'Who seeks what information? Evidence from an elite field experiment on politicians' (in progress)
Project description.
There are a million problems in society in need of a political solution. Far more problems than any political party can possible attend to. The key question for the central actors in representative democracy, i.e. the political parties, is therefore which problems should get attention. Although some problems may appear more severe than others – poverty is a larger problem than dangerous dogs – there is no objective answer as to which problems politicians should attend to. The allocation of scarce attention to problems is an inherent political process (Baumgartner & Jones 2005).
Although the allocation of parties’ attention to problems is a crucial, initial phase of decision-making and problem solving – parties do not solve the problems unless they attend to them in the first place – there is a surprising absence of research. Existing party research provides scant knowledge because it is primarily interested in how parties respond to each other or to voters (Adams 2012; Meguid 2008). There are no systematic, large-scale analysis of party attention to problems. The most ambitious attempt to understand how problems enter politics is Baumgartner and Jones’ (2005) ‘disproportional information processing model’, which concerns the oversupply of problems in society. However, this research focuses on human cognition that has little to say about the political logic in party problem attention. Hence, it is crucial to develop a theoretical model to understand parties’ strategies in their problem attention. This project poses the research question: To what extent and under which conditions do political parties attend to problems?
By revealing the party incentives to address problems, and thus explaining varying reactions across problems, the project adds to the ongoing public debate about variation in political reactions to major societal problems such as climate changes (Biesbroek et al. 2010), terrorism (Epifanio 2011), or financial crises (Bermeo & Pontusson 2012).
To answer the research question, we develop a new theoretical model that unfolds the political logic in the parties’ problem attention. This logic derives from the central tenet of representative democracy; the accountability mechanism. The voters use elections to hold the parties to account for their performance in tackling problems (Key 1966). Hence, elections incentivize vote- and office-seeking parties to attend to problems to avoid vote lose. We argue that the electoral incentives not only ensure baseline attention to problems but also bring particular attention to some problems. The allocation of attention depends on factors that make a problem electorally important. Thus, party attention to problems is not only a mechanism common to all modern democracies, but also a mechanism that varies across issues, parties, and party systems, and during the electoral term. Our theoretical model allows understanding this multilevel variation. We expect that a party attends more to problems to which a rival party already attends (variation at party-level; Green-Pedersen & Mortensen 2010; Meyer & Wagner 2016), especially in two-party systems such as the British (party-system-level; Hobolt et al. 2013), if the rival party has issue ownership or if a focusing events take place (issue-level; Birkland 1998; Petrocik 1996), and closer to elections (electoral term variation).
The theoretical model is tested on 40,000 press releases from political parties in Denmark and Britain, 2004-2019. We study two very different political systems across seven issue areas and several electoral terms to test all of the model implications. Denmark and Britain harbour very different party systems (multiparty vs. two-party), and this most different system design allows to generalize the results to other Western countries. We operationalize problems as statistical indicators from national bureaus since these have become increasingly available in the digital age and central to public and political debate (Kelley & Simmons 2015). To avoid reverse causality, we use exogenous indicators such as the rates of unemployment, violence, and immigrants. To match the frequently changing indicators, we collect press releases at a weekly interval. This is a major improvement to existing, often annual data sources. The press releases are coded into issue categories (Baumgartner et al. 2019) using a human-supervised machine algorithm (Loftis & Mortensen 2018). To test the model, we use cross-section, cross-time statistical estimation.
I will lead a research team at the Department of Political Science at Aarhus University that includes Roman Senninger (Associate Professor at Department of Political Science at Aarhus University), research assistants, and a four-person international advisory board.
Although the allocation of parties’ attention to problems is a crucial, initial phase of decision-making and problem solving – parties do not solve the problems unless they attend to them in the first place – there is a surprising absence of research. Existing party research provides scant knowledge because it is primarily interested in how parties respond to each other or to voters (Adams 2012; Meguid 2008). There are no systematic, large-scale analysis of party attention to problems. The most ambitious attempt to understand how problems enter politics is Baumgartner and Jones’ (2005) ‘disproportional information processing model’, which concerns the oversupply of problems in society. However, this research focuses on human cognition that has little to say about the political logic in party problem attention. Hence, it is crucial to develop a theoretical model to understand parties’ strategies in their problem attention. This project poses the research question: To what extent and under which conditions do political parties attend to problems?
By revealing the party incentives to address problems, and thus explaining varying reactions across problems, the project adds to the ongoing public debate about variation in political reactions to major societal problems such as climate changes (Biesbroek et al. 2010), terrorism (Epifanio 2011), or financial crises (Bermeo & Pontusson 2012).
To answer the research question, we develop a new theoretical model that unfolds the political logic in the parties’ problem attention. This logic derives from the central tenet of representative democracy; the accountability mechanism. The voters use elections to hold the parties to account for their performance in tackling problems (Key 1966). Hence, elections incentivize vote- and office-seeking parties to attend to problems to avoid vote lose. We argue that the electoral incentives not only ensure baseline attention to problems but also bring particular attention to some problems. The allocation of attention depends on factors that make a problem electorally important. Thus, party attention to problems is not only a mechanism common to all modern democracies, but also a mechanism that varies across issues, parties, and party systems, and during the electoral term. Our theoretical model allows understanding this multilevel variation. We expect that a party attends more to problems to which a rival party already attends (variation at party-level; Green-Pedersen & Mortensen 2010; Meyer & Wagner 2016), especially in two-party systems such as the British (party-system-level; Hobolt et al. 2013), if the rival party has issue ownership or if a focusing events take place (issue-level; Birkland 1998; Petrocik 1996), and closer to elections (electoral term variation).
The theoretical model is tested on 40,000 press releases from political parties in Denmark and Britain, 2004-2019. We study two very different political systems across seven issue areas and several electoral terms to test all of the model implications. Denmark and Britain harbour very different party systems (multiparty vs. two-party), and this most different system design allows to generalize the results to other Western countries. We operationalize problems as statistical indicators from national bureaus since these have become increasingly available in the digital age and central to public and political debate (Kelley & Simmons 2015). To avoid reverse causality, we use exogenous indicators such as the rates of unemployment, violence, and immigrants. To match the frequently changing indicators, we collect press releases at a weekly interval. This is a major improvement to existing, often annual data sources. The press releases are coded into issue categories (Baumgartner et al. 2019) using a human-supervised machine algorithm (Loftis & Mortensen 2018). To test the model, we use cross-section, cross-time statistical estimation.
I will lead a research team at the Department of Political Science at Aarhus University that includes Roman Senninger (Associate Professor at Department of Political Science at Aarhus University), research assistants, and a four-person international advisory board.
Organization. The project is structured through three work packages (WPs) on the party level variation in party attention to problems (WP1), the issue level variation (WP2), and the variation during the electoral term (WP3). The WPs are mostly a device to manage the large-scale analysis of the common source of data, and the WPs will run in parallel as soon as data are collected and coded. I will be responsible for developing and executing the WPs together with the postdoc that will be recruited to form the research group with me. To allow the postdoc to gain project leader experience he/she will develop one of the WPs and lead the research and publications in this WP, and I will mainly focus on the other WPs. Each WP is a crucial part of the project and this will position the postdoc at the core of the team. Across the three WPs, the postdoc oversees together with me our research assistants’ collection of the press releases and the content coding of the press releases.
