Fairness Behind a Veil of Ignorance: A Welfare Analysis for Automated Decision Making

TL;DR

This paper introduces a welfare-based fairness measure for automated decision systems, grounded in economic social welfare and Rawlsian fairness, enabling trade-off analysis with accuracy and discrimination.

Contribution

It proposes a convex welfare-based fairness measure, integrating it into decision-making pipelines, and demonstrates its effectiveness in bounding individual inequality empirically.

Findings

Trade-offs between fairness, accuracy, and discrimination are characterized.

Lower bounds on welfare measures can bound outcome inequality.

The method is computationally feasible for practical use.

Abstract

We draw attention to an important, yet largely overlooked aspect of evaluating fairness for automated decision making systems---namely risk and welfare considerations. Our proposed family of measures corresponds to the long-established formulations of cardinal social welfare in economics, and is justified by the Rawlsian conception of fairness behind a veil of ignorance. The convex formulation of our welfare-based measures of fairness allows us to integrate them as a constraint into any convex loss minimization pipeline. Our empirical analysis reveals interesting trade-offs between our proposal and (a) prediction accuracy, (b) group discrimination, and (c) Dwork et al.'s notion of individual fairness. Furthermore and perhaps most importantly, our work provides both heuristic justification and empirical evidence suggesting that a lower-bound on our measures often leads to bounded inequality in algorithmic outcomes; hence presenting the first computationally feasible mechanism for bounding individual-level inequality.