History-dependent fairness preferences - study 2
How to cite
Madland, Kjetil. 2022. "History-dependent fairness preferences - study 2." AEA RCT Registry. AEARCTR-0010180. https://doi.org/10.1257/rct.10180-1.2
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
This RCT builds on a previously registered RCT, ID: AEARCTR-0008997, which studies path-dependence in fairness decisions. The aim of the current experiment is to disentangle possible mechanisms underlying the effect observed in the previous experiment. In this experiment, there is no cost of redistribution. Instead, the is one luck group and one merit group. Spectators in the luck group make ten distributive decisions for pairs of workers, where the source of inequality is luck. The merit group is the same, except that the inequalities are due to differences in performance. In the final 11th decision, spectators in both groups face the same decision, which follows the pattern of the merit group. If I observe the same effect in this experiment as in the previous one, I can conclude that the effect is caused by a shift in what the spectator views as the fair outcome. If the effect is much smaller than in the previous experiment, the previously observed effect was most likely caused by a shift in the relative importance of fairness and efficiency.
Intervention
I conduct an incentivized online experiment. I recruit workers from MTurk to answer a Qualtrics survey, which includes a set of slider tasks. I then recruit impartial spectators from Prolific. These spectators decide how bonus earnings should be divided between the workers. The experiment is probabilistically incentivized. Ten per cent of the spectators are drawn to have one of their decisions implemented to decide the payment for a pair of workers.
Experimental design
I have impartial spectators make distributive choices for pairs of workers. Spectators are allocated to either a luck group or a merit group. Spectators in the luck group make ten distributive decisions for pairs of workers, where the source of inequality is luck. The merit group is the same, except that the inequalities are due to differences in performance. In the final 11th decision, spectators in both groups face the same decision, which follows the pattern of the merit group. If I observe the same effect in this experiment as in the previous one, I can conclude that the effect is caused by a shift in what the spectator views as the fair outcome. If the effect is much smaller than in the previous experiment, the previously observed effect was most likely caused by a shift in the relative importance of fairness and efficiency.
Design details
Cited by
No citation data available for this registration yet.