Universalism in Redistribution Preferences

How to cite

Sternberg, Henrike. 2025. "Universalism in Redistribution Preferences." AEA RCT Registry. AEARCTR-0017391. https://doi.org/10.1257/rct.17391-2.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Henrike Sternberg*

* Primary investigator

Registry ID
AEARCTR-0017391
DOI
https://doi.org/10.1257/rct.17391-2.0
First registered
Last updated
Status
On going
Trial dates
2025-11-25 – 2027-12-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/17391

Abstract

This study examines to what extent the concept of moral universalism extends to inequality aversion or acceptance, and thus, the preference to redistribute. I use experimental survey data from Germany to examine universalism in redistribution preferences in terms of its prevalence, its robustness towards the degree and source of inequality, and relevant individual-level correlates.

Experimental design

Respondents participate in a series of third-party spectator games. Acting as spectators, they will be presented with differential initial distributions of money between two hypothetical people. They then have to decide whether and to what degree they would like to reallocate these initial amounts.

In each decision scenario, one of the hypothetical people is an in-group member of some kind, while the other one is an out-group member. The group identity defining in- and out-group varies, such that each participant is presented with five different group identities. For each respondent, the initial inequality is in favor of an in-group member in half of the scenarios while, in the other half, the initial inequality is in favor of the out-group member. Thus, in total, each respondent is presented with ten decision scenarios (five different group identities, once with the initial inequality favoring the in-group and once favoring the out-group), which are presented in randomized order.

I randomly vary (between subjects) (i) the group identities (two different lists), (ii) the degree of initial inequality (low, medium, high), and (iii) the source of the differential amounts of money (winnings in a lottery vs. performance in a quiz). Each of the two lists of group identities is constructed in line with the different domains of universalism in altruism and trust following Enke et al. (2022).

Design details

Randomization method
Randomization is programmed and performed through the Qualtrics Survey Flow.
Randomization unit
Individual level
Clusters
approx. 4000 respondents
Observations
approx. 3200 (since a random 20% of the sample instead plays disinterested dictator games to measure universalism in Altruism)
Arms
approx. 1600 respondents per list of group identities (2 different lists of group identities) approx. 1067 respondents per degree of initial inequality (low, medium, high) approx. 1600 respondents per source of inequality (lottery/quiz)
Primary outcomes
Individuals' decisions in a series of spectator games where the group identities of the recipients vary

Cited by

No citation data available for this registration yet.

Keywords