Competition Between Human and Artificial Intelligence in Digital Markets: An Experimental Analysis

How to cite

Schnurr, Daniel; Schauer, Andreas. 2022. "Competition Between Human and Artificial Intelligence in Digital Markets: An Experimental Analysis." AEA RCT Registry. AEARCTR-0010439. https://doi.org/10.1257/rct.10439-1.0

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

Daniel Schnurr* , Andreas Schauer

* Primary investigator

Registry ID
AEARCTR-0010439
DOI
https://doi.org/10.1257/rct.10439-1.0
First registered
Last updated
Status
In development
Trial dates
2022-11-23 – 2023-05-31
Countries
Private
JEL codes
L13, L41, C90
Registry record
http://www.socialscienceregistry.org/trials/10439

Abstract

In digital markets business decisions are increasingly taken by artificial intelligence (AI). Especially in e-commerce, a growing share of retailers uses AI-driven algorithmic pricing, whereas remaining vendors rely on manual price setting. However, policymakers have raised concerns about anti-competitive tacit collusion between humans and AI that could allow firms to soften competition. Therefore, we empirically investigate outcomes that arise when humans and AI repeatedly interact in digital markets. Based on an economic laboratory experiment in near real-time, we compare the degree of tacit collusion in duopoly markets across settings with different decision makers and settings with different degrees of algorithmic decision support for human decision makers. In a between-subjects treatment design we systematically vary (i) the decision makers in a market between humans only, algorithms only and mixed market settings where humans and algorithms compete; and (ii) whether human participants receive decision support from an AI-driven pricing algorithm. Altogether, our study sheds light on competition in digital markets where AI plays an increasingly important role and thus bears timely policy and managerial implications.

Intervention

We run five between-subjects treatments that can be derived from our two main treatment dimensions. Along the first treatment dimension, we vary the decision makers that represent firms in a price competition duopoly market. Along the second treatment dimension, we vary whether human participants receive decision support from an AI-driven pricing algorithm.

Experimental design

Experiments are run in an experimental laboratory at the School of Business, Economics and Information Systems at the University of Passau. Treatments are randomized at the session level. Participants will be recruited from the student subject pool of the University of Passau. Each subject participates in only one treatment (between-subject design). In all treatments, subjects are fully informed about the timeline of the experiment.

Design details

Randomization method
Randomization by computer in office
Randomization unit
Experimental sessions
Clusters
In all treatments the independent observation is at the market level, as competitors’ actions depend on each other and are correlated over periods. We schedule data collection aiming at 56 independent observations per treatment. In the human-human competition treatments an independent observation requires participation of two human subjects, whereas in the treatments with a mixed market setting an independent observation only requires participation of one human subject. Market outcomes for algorithm-algorithm competition treatments are derived from simulation outcomes and require no participation of human subjects.
Observations
We schedule data collection aiming at 56 observations per treatment (with 14 human participants per session). Thus, we aim for a total of 336 individual participants across the four treatments.
Arms
In each of the two treatments with a mixed market setting (HA) we plan with 56 participants, which totals 112 subjects. In each of the two treatments with a purely human market setting (HH) we plan with 112 participants, which totals 224 subjects.
Primary outcomes
The average degree of tacit collusion in a market.
Secondary outcomes
Average market price and profits of firms in a market.

Cited by

No citation data available for this registration yet.

Keywords