Corruption experience does not make voters harsher towards the corrupt incumbents – evidence from a survey experiment in Poland

ABSTRACT

This study investigates the linkage between the experience of petty corruption and a more harsh stance on corruption scandals. A randomised vignette experiment in Poland tests how people tend to condone corruption. There are substantial mitigating effects of the arguments based on the implicit exchange and lack of credible information and no effect for the argument based on the lack of clean alternatives. Yet, people directly exposed to petty corruption were not less sensitive to any of the justifications. Thus, the linkage between the experience of corruption and more punitive attitudes towards corrupt officials may not work as expected.

KEYWORDS:

• Corruption scandals
• pocketbook corruption voting
• local corruption
• local elections
• mayors

Disclosure statement

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

Data availability statement

The data are available on request from the author. The data are not publicly available as their containing information (geocoding + respondents’ characteristics) could compromise the privacy of research participants.

Ethics approval statement

This study was approved by the proper Ethical Committee (Faculty of Geography and Regional Studies, University of Warsaw).

Notes

1 The survey completion time varied, with an average time of 17.9 min, and median time of 10.5 min. Once I excluded 5% observations with the shortest response times (below 4.63 min) and 5% observations with the longest response times (above 35.43 min), the analyses yielded similar results.

2 Additionally, as in Polish language nouns, verbs, and adjectives have grammatical gender, the study randomly varied the gender of the mayor in the displayed story. This manipulation could serve as an additional, embedded experiment testing the effect of mayor”s gender on voters” reactions. Yet, for the clarity of argument, it is not reported here and the estimates are averaged over mayor”s gender.

3 As the “don’t know” responses are difficult to interpret, I ran the analyses after excluding “don’t know” answers and (b) assuming that all “don’t know” answers are escape responses of respondents who in fact experienced corruption. In both variants, the results were very similar to the baseline formulation. The additional models are presented in the Appendix (Table A4, Table A5).

4 Pearson correlation coefficients between the residuals of the linear regression model of the reported corruption experience (binary) and a scale measuring local corruption perception on a set of basic socio-demographic covariates (sex, age, education level, living in urban locality). A 0–10 scale measuring perceptions used two survey items referring to perceptions: (1) In local politics, it is difficult to get things done without bribes; (2) Taking into account the situation in my locality, the corruption is widespread.

Additional information

Funding

This work was supported by the University of Warsaw (IDUB Excellence Initiative – Research University program) and the National Agency for Academic Exchange (Bekker Fellowship program, grant number 2020/1/00454).

Notes on contributors

Adam Gendźwiłł

Adam Gendźwiłł is an Associate Professor and Head of the Center for Electoral Studies in the Department of Sociology, University of Warsaw. His research focuses on local governments, political representation, and electoral behaviour. He recently co-edited The Routledge Handbook of Local Elections and Voting in Europe (2022). His articles appeared, among others, in British Journal of Political Studies, Party Politics, and Politics & Gender.