Meritocracy and Persistent Luck

How to cite

Drucker, L. Flóra. 2021. "Meritocracy and Persistent Luck." AEA RCT Registry. AEARCTR-0006709. https://doi.org/10.1257/rct.6709-1.0

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

L. Flóra Drucker*

* Primary investigator

Registry ID
AEARCTR-0006709
DOI
https://doi.org/10.1257/rct.6709-1.0
First registered
Last updated
Status
In development
Trial dates
2021-12-01 – 2021-12-10
Countries
Private
JEL codes
C91, D63, D91
Registry record
http://www.socialscienceregistry.org/trials/6709

Abstract

In a meritocratic society inequality in economic outcomes that reflects differences in individual achievements is considered fair. However, a particular achievement is not equally difficult for everyone. I study two types of differences in difficulties: external circumstances – difficulty of the task an individual faces – and individual ability in the task. In an online experiment, I test how people redistribute income within pairs that was earned by merit with information about how difficult the task was – either in terms of external difficulty or ability in the task – for the members of the pair.

Experimental design

The experiment has two parts. In the first part, participants earn income by production, and in the second part, they make redistribution decisions over the income earned in pairs. Participants are first assigned into the task length or the ability treatment. In the task length treatment the differences in the difficulties in production come from an exogenously assigned task length, while in the ability treatment participants work on similarly difficult tasks, but they may not be equally good at them. In both treatments participants first have to do 10 tasks, as fast as they can, to measure how fast they can solve the tasks. In the task length treatment this serves as information about the relative difficulties of the different task lengths, and in the ability treatment about the relative abilities of the participants in the task. Then they perform the task for 15 minutes. Their income from the first part will be the number of tasks they solved correctly during the 15 minutes.

The second part is the redistribution part, where participants are randomly assigned into pairs, and decide on how to redistribute the first-part income within their pair as stakeholders or for another pair, as spectators. In both the task lenght and the ability treatments for the decisions participants learn the production levels and the task length or the ability of the two members of the pair.

Design details

Randomization method
by a computer
Randomization unit
Task length or ability treatment: individual level Task length within the task length treatment: individual level Spectator or stakeholder role in the redistribution stage: pair level.
Clusters
2 treatments: task length and ability 3 task lengths within the task length treatment 2 roles (spectator and stakeholder) in the redistribution decisions within both the task length and the ability treatment.
Observations
10 decisions per person: 10*500 = 5000 decisions
Arms
250 participants in the task length treatment and 250 participants in the ability treatment
Primary outcomes
share of tokens from the joint number of tokens of a pair allocated to the initially lower-income participant in the pair

Cited by

No citation data available for this registration yet.

Keywords