Skill-Biased Inequality and Redistributive Preferences

Last registered on February 09, 2026

View Trial History

Pre-Trial

Trial Information

General Information

Title

Skill-Biased Inequality and Redistributive Preferences

RCT ID

AEARCTR-0011869

Initial registration date

August 09, 2023

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

August 10, 2023, 1:41 PM EDT

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

Last updated

February 09, 2026, 3:16 AM EST

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Country

United States of America

Region

Primary Investigator

Name

Jeffrey Yusof

Affiliation

Department of Economics, University of Zurich

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Simona Sartor

PI Affiliation

Department of Economics, University of Zurich

Contact Investigator

Additional Trial Information

Status

Completed

Start date

2023-08-09

End date

2024-08-31

Keywords

Behavior

Additional Keywords

inequality, redistribution, fairness

JEL code(s)

Secondary IDs

Prior work

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

Abstract

The rewards for different skills and the resulting skill-biased inequality are often determined by exogenous market mechanisms over which individuals cannot exert control. We refer to this driver of income inequality as market luck. From the perspective of economic efficiency, skill-biased inequality might appear justified as higher rewards reflect higher productivity. However, according to the principles of meritocracy, inequalities are only justified if they are due to differences in individual effort and performance but not due to factors outside of individuals' control. In this paper, we design an experiment to study this trade-off in fairness views and improve the understanding of individuals' preferences for redistribution by asking the following research question: Are inequalities arising from market luck perceived as fair? In our experiment, we design a setting where skill-biased inequality between workers arises because exogenous shocks to market demand make certain skills more valuable. We hypothesize that there are fundamental features of market-driven inequalities that increase individuals' inequality acceptance, even though they are fully aware that the market-driven inequalities result from exogenous and random factors.

External Link(s)

Registration Citation

Citation

Sartor, Simona and Jeffrey Yusof. 2026. "Skill-Biased Inequality and Redistributive Preferences." AEA RCT Registry. February 09. https://doi.org/10.1257/rct.11869-3.1

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

Intervention (Hidden)

Intervention Start Date

2023-08-09

Intervention End Date

2024-08-31

Primary Outcomes

Primary Outcomes (end points)

The redistributive behavior of third-party spectators. (More details in the pre-analysis plan.)

Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)

Secondary Outcomes (explanation)

Experimental Design

We separately recruit three types of subjects; workers, producers, and spectators. In all of our treatments, workers are randomly assigned to one of two tasks, i.e. one of two skills, and then provide effort. In every treatment, we pair two workers with different skills where only one worker earns an additional income for her effort (6 USD). What we vary is the source of inequality between treatments. As our primary outcome, we measure the redistributive behavior of separately recruited third-party spectators. Each spectator is randomly assigned to one treatment and can decide how much income they want to redistribute between two workers who earned unequal incomes.

In our main treatment, the market luck treatment, the worker pair is matched with a producer who demands a specific type of
skill. Matching a pair of differently skilled workers randomly with a producer determines whether a worker can meet a producer’s demand and sell her labor. The fact that only one worker can sell her labor and earn an additional income introduces income inequality between the two workers.

In our control luck treatment, a coin flip decides which worker earns the high income. In the benchmark treatment, both workers work on the task for the same amount of time and the worker with the higher productivity earns the high income. In both treatments, control luck and benchmark, there are no producers. We design additional treatments to investigate the mechanisms and robustness of our main treatment effect.

More details are provided in the pre-analysis plan.

Experimental Design Details

Randomization Method

Individuals will be randomised by a computer.

Randomization Unit

Individual randomization

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

About 1600 spectators, 320 workers, and 110 producers.

We will implement data collection in 2 waves:

Wave 1: Control luck, market luck, and benchmark treatment, each with 200 spectators per treatment (and 120 workers and ca. 40 producers).

Wave 2: Conditional on identifying an effect for the market luck treatment, we will collect data for the mechanism and robustness treatment as well as additional observations for the control and the market luck treatment in wave 2. We will use results from wave 1 for power calculations to determine sample sizes in wave 2. (More details in the pre-analysis plan)

Sample size: planned number of observations

About 1600 spectators, 320 workers, and 110 producers. We will implement data collection in 2 waves: Wave 1: Control luck, market luck, and benchmark treatment, each with 200 spectators per treatment (and 120 workers and ca. 40 producers). Wave 2: Conditional on identifying an effect for the market luck treatment, we will collect data for the mechanism and robustness treatment as well as additional observations for the control and the market luck treatment in wave 2. We will use results from wave 1 for power calculations to determine sample sizes in wave 2. (More details in the pre-analysis plan)

Sample size (or number of clusters) by treatment arms

About 350 spectators in each treatment across both waves of data collection.

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

Supporting Documents and Materials

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

IRB