Auditing Fairness under Unobserved Confounding
This addresses fairness auditing in real-world settings like healthcare for policymakers and researchers, though it is incremental as it builds on existing causal fairness methods.
The paper tackles the problem of auditing fairness in decision-making systems when unobserved confounders exist, by showing that meaningful bounds on treatment rates for high-risk individuals can be computed without observing all risk factors, and demonstrates this with a real-world Paxlovid allocation study, provably identifying racial inequity not explainable by unobserved confounders.
Many definitions of fairness or inequity involve unobservable causal quantities that cannot be directly estimated without strong assumptions. For instance, it is particularly difficult to estimate notions of fairness that rely on hard-to-measure concepts such as risk (e.g., quantifying whether patients at the same risk level have equal probability of treatment, regardless of group membership). Such measurements of risk can be accurately obtained when no unobserved confounders have jointly influenced past decisions and outcomes. However, in the real world, this assumption rarely holds. In this paper, we show that, surprisingly, one can still compute meaningful bounds on treatment rates for high-risk individuals (i.e., conditional on their true, \textit{unobserved} negative outcome), even when entirely eliminating or relaxing the assumption that we observe all relevant risk factors used by decision makers. We use the fact that in many real-world settings (e.g., the release of a new treatment) we have data from prior to any allocation to derive unbiased estimates of risk. This result enables us to audit unfair outcomes of existing decision-making systems in a principled manner. We demonstrate the effectiveness of our framework with a real-world study of Paxlovid allocation, provably identifying that observed racial inequity cannot be explained by unobserved confounders of the same strength as important observed covariates.