Fairness Across the World

How to cite

Tungodden, Bertil; Cappelen, Alexander; Almås, Ingvild; Sørensen, Erik. 2019. "Fairness Across the World." AEA RCT Registry. AEARCTR-0003980. https://doi.org/10.1257/rct.3980-1.0

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

Bertil Tungodden* , Alexander Cappelen , Ingvild Almås , Erik Sørensen

* Primary investigator

Registry ID
AEARCTR-0003980
DOI
https://doi.org/10.1257/rct.3980-1.0
First registered
Last updated
Status
On going
Trial dates
2017-04-10 – 2020-03-31
Countries
Private
JEL codes
D63, C93, I30, J33, J61, O57, Z12
Registry record
http://www.socialscienceregistry.org/trials/3980

Abstract

This is a pre-analysis plan for a unique study of the role of fairness and efficiency considerations in justifying inequalities across the world.1 The classical approach in economics has been to focus on the trade-off between equality and efficiency, where the basic idea is that there is a cost of redistribution and inequality acceptance therefore depends on the relative importance assigned to efficiency versus equality. A growing literature in political economy and behavioral economics, however, has shown that not only efficiency considerations seem to matter for inequality acceptance; fairness considerations may also be of great importance. For example, inequalities reflecting differences in effort appear to be more acceptable than inequalities reflecting differences in luck (Konow, 2000; Cappelen, Drange Hole, Sørensen, and Tungodden, 2007; Almaas, Cappelen, Sørensen, and Tungodden, 2010).

This project provides a novel data set on the nature of inequality acceptance. It provides unique descriptive data on inequality acceptance across the world; comprehensive data on people’s beliefs about the sources of inequality and cost of redistribution; novel data on how unfair people perceive the income distribution to be and their attitude to redistribution; causal evidence on the importance of the source of inequality and efficiency considerations for inequality acceptance; rich individual data that will support a number of studies on how inequality acceptance is associated with socioeconomic status, religion, gender, age, and other relevant background variables.

Intervention

International representative survey in which participants implemented an actual distribution of bonus payments for workers on an international labor market platform. The intervention randomized the context of pre-distribution earnings, which were caused by luck or merit, and whether redistribution was costless.

Experimental design

We use two types of participants in this study, spectators and workers, where our primary interest is in the distributive choices made by the spectators. The workers are recruited through an online international market platform, to conduct some work for the research team, where they are told that, except for a small participation fee, the payment for the work will be determined by a third party. The spectators are told that, in contrast to traditional survey questions that are about hypothetical situations, in this study they make a choice that has consequences in a real life situation. They are then presented with a real life situation involving two of the workers and are told that it is their task to determine the final payment for these two workers. The spectators are informed that different earnings have been assigned to the workers (not paid out) as well as the basis for the allocation. The decision that the spectators have to make is how much to redistribute between the two workers. Participants were randomized into different spectator treatments (between-individual design).

There are three main treatments, each informing us about an interesting outcome. We are interested in the treatment differences, as outlined in the enclosed PAP, but there is no treatment that is only meant as a "control", and hence no treatment is labeled as such.

Design details

Randomization method
Randomization by computer
Randomization unit
individual
Clusters
60,000 individuals
Observations
60,000 individuals
Arms
20,000 individuals
Primary outcomes
fairness preferences, beliefs about the source of inequality, beliefs about the cost of redistribution, attitudes to income inequality and redistribution

Cited by

No citation data available for this registration yet.

Keywords