The Party-Political Integration of Members with Migrant Backgrounds

ABSTRACT

Members with a migrant background (MMB) are underrepresented in political parties. So far, most research on this issue only scrutinises the representation of elite-level politicians with migrant backgrounds. To enhance this literature, I survey approaches toward the party-political integration of MMB at the rank-and-file level. The paper is centred around the following question; are local party leaders committed to the party-political integration of MMB or are they biased against them? In two original studies in Germany, including a field experiment and a conjoint experiment, I show that local party leaders are equally willing to engage in public discourse about the recruitment of MMB and general member recruitment and that many of them declare a commitment toward increasing the share of MMB. However, this declared commitment is in large part driven by social-desirability bias: when asked to declare preferences based on multiple attributes of potential members, the commitment to increase the share of MMB disappears for most parties. Much more than their migration background, the age and gender of potential members dictate preferences: young people and women are most often mentioned as preferred new members.

Acknowledgements

I thank Christina Zuber and Gabriele Spilker for their advice and help at various stages of this research project. I thank Javier Martínez-Cantó, Stephanie Hofmann, and the participants of the EUI ‘Reviewing and Publishing’ workshop for valuable feedback. I thank the editors and reviewers for requesting sensible revisions that improved an earlier version of this manuscript. Last but not least, I thank Richanet Grace Bolina for her support.

Disclosure Statement

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

Supplemental Data and Research Material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/09644008.2024.2321441. Replication Files are also available at https://github.com/fresej.

Notes

1 Throughout the manuscript, I refer to political party members who are first- or second-generation migrants as MMB. Likewise, in my communication with the German local party leaders, I use the term Migrationshintergrund, which translates to migrant background and includes first- and second-generation migrants. I acknowledge that this term encompasses many different groups that might be discriminated against to different degrees based on their ethnicity, religion, or other factors. However, I focus on the broad migration background anyway because this is an umbrella category that is well-recognized in German politics and society. To account for the heterogeneity of origins, the empirical analyses include multiple countries of origin that are randomly assigned to vignettes.

2 Even with quotas in place, equal representation is difficult if a minority group is underrepresented at the member level, as has been shown in the case of gender quotas (see, e.g., Davidson-Schmich 2006; Davidson-Schmich 2016).

3 The overall number of MMB parliamentarians in state parliaments increased from 1,4% (26 parliamentarians) in 2005 to 4,5% (83 parliamentarians) in 2015.

4 After the 2017 federal election, 58 out of 709 parliamentarians (Bundestagsabgeordnete) had a migrant background. After the 2021 election, 83 out of 735 parliamentarians had a migrant background.

5 He was appointed as federal minister of food and agriculture in an SPD/Green/FDP government following the 2021 federal election.

6 Surprisingly, this holds true for members of parliament from both left and right parties.

7 Full Names: Social Democratic Party, Christian Democratic Union/Christian Social Union, The Greens, Free Democratic Party, The Left, Alternative for Germany.

8 By local chapter, I am referencing what is generally called a Kreisverband in Germany. This name makes sense because one Kreisverband usually covers the area of one Landkreis/district. However, for a multitude of reasons, several parties have different names for their local chapters in some states. In the three city states, Berlin, Bremen, and Hamburg, as well as Niedersachsen, many local chapters are called Bezirksverband instead, but these mostly cover the same areas that the Kreisverbände of other parties cover. Some chapters in cities that don’t belong to a district (Kreisfreie Städte) are called Stadtverband by some of the parties in Sachsen and Sachsen-Anhalt. Again though, these cover the same areas as the Kreisverbände of other parties. Finally, in Sachsen, Bayern, Brandenburg, Hessen, Rheinland-Pfalz, Niedersachsen, and Nordrhein-Westfalen, some local chapters are called Unterbezirk, which again only constitutes a difference in name, not in function or scope. In sum, while not all local chapters included in the case selection have the same nomenclature, they all fulfill the same function on the political map.

9 The precise question wording for the entire survey is included in the Appendix.

10 E.g., head of the chapter (Kreisvorstand/Kreisvorsitzender), or speaker of the chapter (Kreissprecher).

11 After three days, both groups received reminder e-mails that again only differed in the mention of the treatment/control topic. No new information was added to these e-mails, to ensure that late respondents would not have a different perception of the survey compared to early respondents.

12 Next to pre-specified groups, participants were also able to name additional groups of their own. These open field mentions are not included in the plot because their frequencies are not comparable due to availability bias: respondents are naturally more inclined to go for examples already presented to them.

13 In the case of ethnic minorities, it seems likely that respondents saw an overlap with MMB and thus chose to mention only one of the two groups, with migrant background being a more commonly used term in Germany than ethnic minority.

14 The pre-analysis plan lists a third hypothesis which is not tested in this manuscript as it was conditional on a certain outcome of the second hypothesis which did not transpire.

15 With a set order of attributes, respondents might be inclined to pay more attention to the first attribute or the last attribute rather than treating all attributes with the same amount of care. This would go contrary to their instructions of taking all five attributes into consideration.

16 The approach of measuring issue ownership through party manifestos has also been used by e.g., Busemeyer, Franzmann, and Garritzmann 2013. Other papers measure issue ownership through the perceptions of voters, but this seems inappropriate here because the parties’ self-perceptions of issue ownership are more relevant to the hypothesis generation.

17 The exact immigration/multiculturalism positions as reported in the Chapel Hill Expert Survey, as well as the salience of immigration/multiculturalism in the party manifestos as reported in the Comparative Manifestos Data, are included in the Appendix.

Additional information

Notes on contributors

Joris Frese

Joris Frese is a doctoral researcher in political and social sciences at the European University Institute in Italy. In his research, he is interested in all things related to political behaviour and political communication.