Skip to main content

The Populist Marketplace: Unpacking the Role of “Thin” and “Thick” Ideology


A growing body of work adopts a “thin” ideology conception of populism, which attributes populist parties’ electoral success to anti-elite and people-centric appeals that resonate with voters holding populist attitudes. A second tradition, however, has attributed the success of populist parties to particular “thick” or “host” ideologies, such as anti-immigration, anti-globalization, or pro-redistribution positions. This creates a need to unpack which exact components of thin and/or thick populist ideology attract voters to these parties. We address this question by leveraging conjoint survey experiments that allow us to causally identify the effects of several thin and thick populist attributes on vote choice. Examining the case of Germany, results from experiments embedded in two high-quality panel surveys demonstrate that populist anti-immigration and pro-redistribution positions as well as people-centric political priorities are the most vote-maximizing components of populist ideology. In contrast, anti-elite priorities as well as Eurosceptic and anti-globalization positions do not boost support, not even among voters with strong populist attitudes. Our findings also call into question conventional wisdom about the interplay between supply and demand in the electoral marketplace. Surprisingly, populist voters, in general, are not significantly more attracted to candidates who advocate populist priorities than non-populist voters.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2


  1. All data and scripts required to replicate the analyses in the manuscript and the Online Appendix are deposited in the Political Behavior Dataverse and available at

  2. This design has been recommended by Hainmueller et al. (2015), who show that estimates based on it are slightly closer to behavioral benchmarks than those based on single-profile or vignette text designs. Note that Hainmueller et al. (2015) suggest a paired conjoint design without forced choice response options. However, their experiments on voting for the naturalization of immigrants in Switzerland represent a case in which voters can in actuality vote to naturalize both immigrants. In contrast, in our case, voters can only vote for one candidate in reality, which is why we use a forced choice design.

  3. Researchers could also add less prominent components of thin or thick populist ideology (e.g., populist positions on punitive justice).

  4. Hence, even in its ex post transformation, our design has a number of constraints on attribute level combinations (e.g., combinations of fewer than two, or three or more issues being a priority are excluded). As a result, the marginal means of “Priority” versus “No priority” are hardly comparable, since they marginalize over different subsets of the possible levels for the other issues (e.g., with “No priority” all cells with combinations of two other issues as “Priority” are possible but they are excluded under “Priority”, since only one other issue can be a priority then). Below we therefore abstain from reporting marginal means for “No priority”. However, the marginal means for the “Priority” levels can be compared, as they marginalize over symmetrical sets of levels for the other variables.

  5. Note that the cregg package uses the survey package to calculate MMs as survey-weighted means with appropriate standard errors.

  6. In Sect. 8 of the Online Appendix, we show that our results are robust to using an alternative definition and operationalization of the subgroups.

  7. In particular, it demonstrates that our results are largely independent of trends in party support. In March 2017, when we fielded the design the first time, opinion polls by infratest dimap showed about 31-32% vote share for the SPD if a federal election were held that Sunday (a high level of SPD support that had not occurred for over five years). By May/June 2018, at the time of fieldwork for the replication, the SPD’s projected vote share had fallen to 17-19%. This drop in the SPD’s vote share was the most significant drop in vote share for a party in Germany in at least two decades. Our experimental estimates are entirely unaffected by this change in partisan context.

  8. We chose this particular component of thin populism, since it can most credibly be presented as a positional issue. For instance, we think that candidates who would advocate positions for political corruption or against defending citizens’ interests would be perceived as hardly credible by respondents.


Download references


The authors thank Robert Vehrkamp from the Bertelsmann Stiftung for the excellent collaboration on this project. The data collection for this project was financed by the foundation. We are also indebted to Yaoyao Dai, Sergi Pardos-Prado, Andrej Zaslove, Sven-Oliver Proksch, and Bruno Castanho Silva for helpful comments on earlier versions of the manuscript. Previous versions of this work were presented at the 2018 Midwest Political Science Association Annual Conference, the 2018 European Political Science Association Annual Conference, the 2019 International Society of Political Psychology Annual Conference, and the 2018 “Populist Attitudes in Comparative Perspective” research workshop at the University of Bamberg. All remaining errors are our own.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Fabian G. Neuner.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary file1 (PDF 529 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Neuner, F.G., Wratil, C. The Populist Marketplace: Unpacking the Role of “Thin” and “Thick” Ideology. Polit Behav 44, 551–574 (2022).

Download citation

  • Published:

  • Issue Date:

  • DOI:


  • Populist voting
  • Populist attitudes
  • Extremist parties
  • Conjoint experiment
  • Germany