Endogenous Inequality Aversion: Decision criteria for triage and other ethical tradeoffs

2026-01-29

AI Summary

This study addresses the ethical tension between utilitarianism and Rawlsian justice in resource-scarce contexts, such as medical emergencies. It proposes an endogenous inequality-averse social welfare function whose allocation criterion adaptively shifts with the overall welfare level: favoring utilitarianism to maximize total utility under low welfare conditions, and transitioning toward Rawlsian prioritization of the worst-off as welfare increases—or vice versa, depending on design specifications. By employing an axiomatic approach, the authors construct a self-referential social welfare function that formally yields a smooth, principled transition between these two ethical paradigms. This framework offers a unified, formal ethical foundation for resource allocation that remains applicable across both crisis and normal circumstances.

Technology Category

• Game Theory and Economic Paradigms: Fair Division
• Philosophy and Ethics of AI: Bias, Fairness & Equity
• Reasoning under Uncertainty: Decision/Utility Theory

Application Category

• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems
• Responsible Web: Algorithmic accountability and transparency on the web

Abstract

Medical ``Crisis Standards of Care''call for a utilitarian allocation of scarce resources in emergencies, while favoring the worst-off under normal conditions. Inspired by such triage rules, we introduce social welfare functions whose distributive tradeoffs depend on the prevailing level of aggregate welfare. These functions are inherently self-referential: they take the welfare level as an input, even though that level is itself determined by the function. In our formulation, inequality aversion varies with welfare and is therefore self-referential. We provide an axiomatic foundation for a family of social welfare functions that move from Rawlsian to utilitarian criteria as overall welfare falls, thereby formalizing triage guidelines. We also derive the converse case, in which the social objective shifts from Rawlsianism toward utilitarianism as welfare increases.

Problem

• inequality aversion
• social welfare function
• triage
• Rawlsian
• utilitarianism

Innovation

• endogenous inequality aversion
• self-referential social welfare
• triage ethics
• Rawlsian-to-utilitarian transition
• axiomatic foundation