![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
Abstract
Party system issue agendas are formed by the topics that individual parties decide to address, and these salience decisions are likely to be strategic. Two key strategies are commonly discussed in the literature: parties’ focus on (1) issues that they have ownership over and (2) issues that currently concern voters. Yet it is not known what explains the extent to which parties pursue each of these strategies. This paper argues that aspects of party organisation influence which salience strategy is pursued. Parties that have more resources will be able to ‘ride the wave’ of current concerns while parties with fewer resources are more likely to focus on their best issues. Furthermore, policy-seeking parties with strong activist influence will be less likely to ‘ride the wave’ and more likely to follow issue ownership strategies. An analysis of 105 election manifestos from 27 elections in 17 countries shows that aspects of party organisation are indeed strong and robust moderators of issue ownership strategies. Limited, albeit mixed, evidence is also found that party organisation affects the use of ‘riding the wave’ strategies. These results have important implications for our understanding of electoral campaigns, party competition and voter representation.
Acknowledgements
Research on this project was financed by the Austrian Science Fund (FWF) under the project ‘Austrian National Election Study’ (AUTNES, S10902-G11 and S10903-G11). We would like to thank our reviewers and the conference participants at the MPSA Annual Conference 2012, the EPSA General Conference 2012 and the CES Annual Conference 2012 for comments on earlier drafts of this paper. We would also like to thank Roman Senninger for research assistance. All errors remain our own.
Notes
1. In Walgrave et al.’s (Citation2012) approach, this is called ‘competence issue ownership’, so whether a party is seen as being able to deal with an issue effectively.
2. Note that this means that the term ‘ownership’ is in fact a little misleading as ownership need not be exclusive.
3. Issue ownership may also change over time: a long-term reputation does not equal an unchangeable reputation (Tresch et al. Citation2013).
4. The question wording has changed from CSES Module 2 (‘most important issue’) to Module 3 (‘the most important problem’). Jennings and Wlezien (Citation2011) show these changes do not affect the aggregate issue importance scores. Moreover, Bartle and Laycock (Citation2012) show that aggregate responses to ‘most important issue’ questions reflect the issue concerns of a typical voter. While individual responses most likely reflect the issue concerns of fellow citizens rather than personal preferences, averaging individual responses provides reasonable estimates for issue concerns in the electorate.
5. European integration is not considered for the countries outside Europe (i.e. Australia, Canada, New Zealand and the USA). We include it for non-EU countries in Europe (i.e. Norway and Switzerland) as the relationship with the EU is a matter of political debate in these countries.
6. For the CMP/MARPOR data we use here, existing research (Mikhaylov et al. Citation2012; Pennings Citation2011; Volkens et al. Citation2013) shows that some of 56 issue categories are not reliable and should not be used in isolation. We address the problem somewhat by aggregating the issue categories into 10 issue areas.
7. These documents, often carefully negotiated within parties, serve many purposes, not least to please party activists. Thus, they are more likely to focus on a party’s owned issues and less likely to incorporate current voter concerns than other forms of party communication such as press releases and media appearances.
8. The authors coded this list independently, with conflicting decisions reconciled through joint discussion. The full list matching CSES codes to the 10 issues is available from the authors.
9. Although we can only measure voter salience after the election took place, we believe that the post-election issue agenda should be a good proxy for the pre-election period on which manifestos were based. This is because the time span between the publication of a party’s manifesto and the election itself is usually short, often just a couple of weeks (Dolezal et al. Citation2012), and the public issue agenda is rather stable over a single election campaign (Page Citation1978: chapter 5; Clarke et al. Citation2009: chapter 3 and 152–6).
10. Party competence assessments for the most important issues are only included in Module 3 (but not in Module 2) of the CSES. Furthermore, even the available data is not suitable for our purposes because the number of respondents evaluating the parties’ issue competence is very small for the less salient issues.
11. Intuitively, one may expect parties’ resources and their policy focus to be related. Yet smaller parties with fewer resources are not necessarily activist-centred and policy-seeking. The difference is perhaps most apparent with regard to Green and small radical right parties (Burchell Citation2001; Mudde Citation2007).
12. We also excluded manifestos from the analysis if their salience coding was ‘estimated’ (i.e. not measured but interpolated from adjacent observations).
13. We also tested alternative model specifications that take characteristics of our data structure into account. The results for these models are very similar to the ones presented below (see supplemental data). We also assessed the robustness of our findings by excluding niche parties, accounting for the effect of government participation, and by excluding one issue at a time. Our substantive results do not change as a result of these re-specifications (see supplemental data).
14. Due to space constraints, we do not discuss these interaction effects of our control variables in greater detail. Marginal effect graphs are available upon request.
15. If issue importance increases by one standard deviation from 8 to 22 per cent, party issue emphasis increases by about 6 percentage points (e.g. from 10 to 10.6 per cent of the manifesto).
16. We thank an anonymous reviewer for pointing out this argument.