‘Parties in the Shadows’ –Do Small and Marginal Parties Cater to a Niche?

Abstract

Small and marginal parties are seldom the focus of studies in political science and are usually excluded from analyses altogether. However, an increase in the number of these ‘other’ parties can be observed during federal elections, though their chances of gaining seats are almost non-existent. Moreover, small and marginal parties can have an impact on the party system by diffusing topics and posing a potential electoral threat to established parties, with whom they face the challenging task of competing. One way of increasing their chances of getting votes is by gaining issue ownership over a niche topic. In this study, we use a quantitative content analysis and a dictionary approach to investigate 16 small and marginal German parties in order to determine whether they aim to cater to a niche or act as ‘failed’ mainstream (topic) parties. Our analysis reveals that small and marginal parties indeed have incentives to cater to a niche in order to gain electoral advantages, though the same strategy also contributes to their small and marginal existence.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the author(s).

Notes

1 There are many possible methods and datasets for analysing manifestos. One of the most prominent datasets comes from the Manifesto Project (Volkens, Lehmann, and Matthieß (2016)). A multitude of studies, including those in the field of niche parties, make use of the impressive data pool (for example Bischof (2015); Meguid (2005); Meyer and Miller (2015)). Aside from some criticism regarding noise in the data (Hansen (2008); Mikhaylov, Laver, and Benoit (2008)), the Manifesto Project also only includes parties in Parliament or on the verge of entering Parliament and is therefore discounted in our research because most of the parties we cover are naturally not in parliament.

Another option in analysing manifestos is automated corpus analysis, provided by Wordscores (Laver, Benoit, and Garry (2003)) or Wordfish (Slapin and Proksch (2008)). These two options are tempting but ultimately come with significant caveats. The Wordscores method relies on reference texts, which are needed for a priori position estimations. In the present study, we are instead concerned with the salience of certain topics within the manifestos rather than with the positioning of the parties. As an unsupervised method of text scaling, Wordfish suffers from the same problem in that it also only retrieves positional values instead of saliency, which is the focus of this work.

2 The consistency between the two coders was reasonably high, with a Cronbach’s alpha of 0.74. Words with different codes between the coders were excluded. By doing so we have established a dictionary which avoids the problem that words potentially influence another category. Otherwise, the words would drive up the score for improperly assigned categories. If, for example, the word "unequal" impacts the "social" as well as the "democracy and participation" category, this would distort the party's focus. Furthermore, an inspection of the keywords in context showed that the words have been appropriately categorised in our dictionary.

3 One reason of using the Meyer and Miller (2015) approach in our study was its openness to emerging niches which is why we have rather broad policy fields.

4 It should be noted that due to our research focus, we use a rather broad definition of the party system that encompasses all parties that compete in elections, which is in contrast to other definitions that focus on the ‘relevant parties’ in the system.

5 We used the official manifesto of Die PARTEI; however, a ‘second’ manifesto also exists. On its website, the party provides an ‘individual manifesto’, in which its manifesto is altered to match users’ preferences.

6 Of course, differences in nicheness can still be seen for parliamentary parties, as is claimed by the niche party literature. For example, the AfD is considerably more niche than the very mainstream CDU or the SPD, a finding that is in line with the literature on radical right niche parties. See e.g. Abou-Chadi (2016), Meyer and Miller (2015), or Wagner and Meyer (2016). However, the mean nicheness score of the AfD is below that of Die LINKE. The possibility that economy-driven parties are niche is precisely why Meyer and Miller (2015) argue for the inclusion of the rival parties’ focus instead of using party family affiliations when determining parties’ nicheness.

7 Nevertheless, it is still possible that these parties do not act fully strategically and are more constrained by policy-seeking party members. However, this does not alter the conclusions of this study that being more niche is an electorally beneficial factor.

Additional information

Notes on contributors

Jan-Eric Bartels

Jan-Eric Bartels is a research associate at the Chair for German Politics, based in the Department of Political Sciences, University of Göttingen. His interests include small and marginal parties, party competition and quantitative content analysis.

Max Leonard Remke

Max Leonard Remke is a PhD student at the Chair of Quantitative Methods and Statistics based in the Center of Methods in Social Sciences, University of Göttingen. His interests include party competition and political communication in the context of historical discourses.