Eliciting preferences over distributive justice principles

Last registered on October 07, 2024

View Trial History

Pre-Trial

Trial Information

General Information

Title

Eliciting preferences over distributive justice principles

RCT ID

AEARCTR-0014413

Initial registration date

September 25, 2024

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published

October 07, 2024, 7:00 PM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Country

United Kingdom of Great Britain and Northern Ireland

Region

Primary Investigator

Name

Marco Lambrecht

Affiliation

University of Stavanger

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Ola Andersson

PI Affiliation

Uppsala University

Contact Investigator

PI Name

Topi Miettinen

PI Affiliation

Hanken School of Economics & Helsinki GSE

Contact Investigator

Additional Trial Information

Status

In development

Start date

2024-09-26

End date

2024-10-11

Keywords

Other

Additional Keywords

Methodology, Distribution

JEL code(s)

C83, D31, D63

Secondary IDs

Prior work

This trial does not extend or rely on any prior RCTs.

Abstract

Distributive justice principles are commonly used in societal debates to support policy positions and argue for their legitimacy. Prominent formal incarnations of such principles are the Rawlsian, Utilitarian, Efficient transfer (e.g. Pigou-Dalton), and Leaky-bucket transfer (e.g. Hammond) principles, but could also be formulated by the minimization of the Gini coefficient, for instance. Individual preferences over such principles may arguably depend on individuals’ understanding of their application. In an incentivized task, we observe heterogeneity in individuals’ abilities to apply distributive justice principles to specific income distribution scenarios. We systematically test whether exposure to applying these principles, with and without feedback, changes preferences over principles. We compare responses in a student sample to a sample representative of the general population in the UK. Our findings may have important implications for the elicitation of preferences over distributive justice principles.

External Link(s)

Registration Citation

Citation

Andersson, Ola, Marco Lambrecht and Topi Miettinen. 2024. "Eliciting preferences over distributive justice principles." AEA RCT Registry. October 07. https://doi.org/10.1257/rct.14413-1.0

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

The experimental design includes three treatments:
1) Baseline: subjects apply a dummy principle to pairs of income distributions
2) Application: subjects apply distributive justice principles to pairs of income distributions
3) Feedback: subjects apply distributive justice principles to pairs of income distributions and receive feedback whether their answers are correct

Intervention (Hidden)

We use the same set of pairs of income distributions in all three treatments. In treatments Application and Feedback, our subjects apply five principles to two pairs each. In treatment Baseline, they apply the same dummy principle to all ten pairs. The order of pairs is randomized. In all three treatments, subjects go through their decisions once again after responding to all ten pairs. In treatment Baseline and Application, they are reminded of their answers. In treatment Feedback, they additionally learn whether their submitted answer is correct. In all three treatments, subjects then submit a ranking over five distributive justice principles (the ones which they already applied in treatments Application and Feedback). Finally, all subjects go through a short survey which elicits controls and secondary outcome variables.

Intervention Start Date

2024-09-26

Intervention End Date

2024-10-11

Primary Outcomes

Primary Outcomes (end points)

Preferences over five distributive justice principles

Primary Outcomes (explanation)

Subjects see all five principles simultaneously and submit a ranking from 1 (most preferred) to 5 (least preferred) for each principle.

Secondary Outcomes

Secondary Outcomes (end points)

Policy preferences

Secondary Outcomes (explanation)

Using a continuous scale from "Strongly Disagree" to "Strongly Agree", we elicit preferences for
1) funding education vs benefits for the poorest in society
2) inheritance tax vs progressive income tax
3) equalizing incomes
4) inheritance tax
5) capital gains tax
6) unconditional basic income

Experimental Design

Stages of the experiment:
- application task
- ranking
- survey questions
- summary

The experimental design includes three treatments:
1) Baseline: subjects apply a dummy principle to pairs of income distributions
2) Application: subjects apply distributive justice principles to pairs of income distributions
3) Feedback: subjects apply distributive justice principles to pairs of income distributions and receive feedback whether their answers are correct

We collect responses online from two different samples:
A) Student sample (UK students, via Prolific)
B) Representative sample (UK population, via Prolific)

Experimental Design Details

Randomization Method

Recruitment via Prolific; Randomization to treatment in oTree

Randomization Unit

individual

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

300 individuals in the representative sample
300 individuals in the studen sample

Sample size: planned number of observations

600 individuals

Sample size (or number of clusters) by treatment arms

100 Baseline representative, 100 Application representative, 100 Feedback representative
100 Baseline student, 100 Application student, 100 Feedback student

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Power calculation using G*Power based on oneway ANOVA (parametric alternative to estimate power for Kruskal-Wallis-Test). Given a 5% error probability (alpha), n=300, and 3 groups, the mdes at 80% power is 0.18. Given a 5% error probability (alpha), n=300, and 3 groups, the mdes at 95% power is 0.23.

Supporting Documents and Materials

IRB