Paper Type
Complete
Abstract
Algorithmic evaluation is increasingly prevalent in online labor markets, where automated systems now filter and rank applicants. Unlike traditional employment, platform work requires applicants to declare a minimum acceptable wage, turning evaluation into self-valuation. This creates a feedback loop: the system that judges workers can also shape their self-worth. We argue that algorithmic evaluation introduces an impersonal condition that elicits counterfactual fairness perceptions (CFPs), beliefs that one would have been treated more fairly under an alternative evaluator (human vs. algorithm). We conducted an online experiment where applicants were randomly assigned to human or algorithmic evaluation. We found that women set lower minimum wages than men overall, and this gap widened significantly under algorithmic evaluation. This divergence was mediated by CFPs. Women reported lower CFPs and reduced bids, whereas men reported higher CFPs and maintained or increased them, revealing how algorithmic evaluation can unintentionally amplify wage inequality and reshape fairness reasoning.
Paper Number
1399
Recommended Citation
Lakhiwal, Akshat; Siamionava, Katsiaryna; and Liu, Che-Wei, "Devalued by Design? How Counterfactual Fairness Perceptions Shape Wage Expectations in Algorithmic Hiring" (2026). AMCIS 2026 Proceedings. 3.
https://aisel.aisnet.org/amcis2026/ai_aiaa/ai_aiaa/3
Devalued by Design? How Counterfactual Fairness Perceptions Shape Wage Expectations in Algorithmic Hiring
Algorithmic evaluation is increasingly prevalent in online labor markets, where automated systems now filter and rank applicants. Unlike traditional employment, platform work requires applicants to declare a minimum acceptable wage, turning evaluation into self-valuation. This creates a feedback loop: the system that judges workers can also shape their self-worth. We argue that algorithmic evaluation introduces an impersonal condition that elicits counterfactual fairness perceptions (CFPs), beliefs that one would have been treated more fairly under an alternative evaluator (human vs. algorithm). We conducted an online experiment where applicants were randomly assigned to human or algorithmic evaluation. We found that women set lower minimum wages than men overall, and this gap widened significantly under algorithmic evaluation. This divergence was mediated by CFPs. Women reported lower CFPs and reduced bids, whereas men reported higher CFPs and maintained or increased them, revealing how algorithmic evaluation can unintentionally amplify wage inequality and reshape fairness reasoning.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
SIG AIAA