Topic Coverage of Coalition Agreements in Multi-Level Settings: The Case of Germany

Abstract

Scholars mostly concentrate on those agreements that are written at the national level when analysing coalition agreements. However, there are also a considerable amount of coalition agreements that are written at sub-national levels. This paper analyses the commonalities and differences in coalition agreements in the German multi-level system at the national, regional and local level. From a legal jurisdiction perspective, one would expect that there are major differences across political levels in the topics covered in the agreements. From a multi-level governance perspective, however, one would additionally expect that government parties also devote their attention to policy domains that lie outside their realm of legislative decision-making. We take Germany as a prime example of a political system characterised by joint decision-making within cooperative federalism. Combining data from the Political Documents Archive (www.polidoc.net) with newly gathered data from the Local Manifesto Project (LMP; www.localpartymanifestos.de), we analyse nearly 200 coalition agreements at the national, regional and local level in Germany by applying quantitative text analysis techniques. The empirical results show that governing parties mostly discuss the policy areas they are legally responsible for. However, particularly local and regional governing parties also address issues that are primarily part of federal jurisdiction.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.

Notes

1 In contrast to other authors (see e.g. Klüver and Bäck, forthcoming) we are not interested in the outcome of coalition negotiations (i.e. the influence of coalition parties on the content of the coalition agreement) but rather in the distribution of attention towards certain topics in the coalition agreement.

2 Coalition governments in parliamentary democracies are prone to less transparent politics because it is likely that negotiations within the coalition or between government parties are more private and internal than in single-party governments (Strom 2000, 281), and thus cannot enter voters’ considerations on election day (León 2018, 707). Coalition agreements are one way to counter this lack of transparency since government parties publicly commit themselves to jointly agreed policy proposals.

3 In some countries, for example in the German multi-level governance system (see below), the local level does not have a say in policy-making at the national and state level; yet, the local level often has large execution competences, i.e. in some policy areas it is local political actors that decide to what extent and in which way they execute national and state laws.

4 Additionally, the local level in many European democracies is characterised by an interplay between local council majorities and directly elected mayors. Since mayors are the most prominent figure in local politics, thus resembling presidents in (semi-)presidential systems (cf. Hobolt, Tilley, and Banducci 2013; Oktay 2018; Valdini and Lewis-Beck 2018), ‘clarity of responsibility’ and pledge fulfilment might be more dependent on the performance of the mayor and less on local government parties’ performance and promised policies in their coalition agreements.

5 Note that this last point will change with the reorganisation of federal-state financial arrangements from 2020 on, when the federal level will be allowed to help municipalities financially, at least to some extent (Renzsch 2017, 769).

6 Local governments ‘fulfil tasks either by own competences of self-government—some of them being obligatory and others optional—or by competences delegated by the Land government’ (Benz and Zimmer 2011, 160–161).

7 Note that the institutional setting at the German local level with a directly elected mayor and a local council is neither a pure presidential nor a pure parliamentarian system (Bäck 2005; Egner 2015) but rather a ‘mixed democracy’ (Gross and Debus 2018a) or ‘institutional hybrid’ (Egner 2015) between a ‘quasi-presidential’ (Benz and Zimmer 2011; Egner and Stoiber 2008) or a ‘semi-presidential’ (Debus and Gross 2016; Bäck 2005) government system. This does not change, however, the equivalence of local coalition agreements to coalition agreements at the federal and state level because local government parties can still shape policy-making and portfolio allocation at the local level, even if the directly elected mayor is from an opposing party.

8 For a detailed justification of why cities in Germany with at least 100,000 inhabitants are ‘parliaments in disguise’ and why parties frequently form coalitions see Egner (2015) and Gross (2018). Note that coalition agreements at the local level were not always publicly available, thus we might be slightly underestimating the number of coalition agreements at the local level.

9 Note that Pappi and Seher (2009) explicitly are interested in party manifestos’ ‘coalition-relevant content’ because of the ‘policy domain-specific policy signals’ to ‘potential coalition partners’ (Pappi and Seher 2009, 408), thus coming close to our intention in this article to study policy-relevant topics in coalition agreements.

10 The results were obtained after we cleaned and set up our text corpus with the ‘tm’ R-package (Feinerer and Hornik 2018) and ran the dictionary analysis with the ‘quanteda’ R-package (Benoit et al. 2018).

Additional information

Notes on contributors

Martin Gross

Martin Gross is an assistant professor at the Geschwister-Scholl-Institute of Political Science at the Ludwig-Maximilians-University Munich, Germany. He works on local politics, political institutions, party competition, coalition politics in multi-level systems, and EU cohesion policy.

Svenja Krauss

Svenja Krauss is a lecturer at the University of Essex. Her research interests include coalition governments, political parties, political behaviour and quantitative methods.