Structural Interventions and the Dynamics of Inequality

Abstract

Recent conversations in the algorithmic fairness literature have raised several concerns with standard conceptions of fairness. First, constraining predictive algorithms to satisfy fairness benchmarks may sometimes lead to non-optimal outcomes for disadvantaged groups. Second, technical interventions are often ineffective by themselves, especially when divorced from an understanding of structural processes that generate social inequality. Inspired by both these critiques, we construct a common decision-making model, using mortgage loans as a running example. We show that under some conditions, any choice of decision threshold will inevitably perpetuate existing disparities in financial stability unless one deviates from the Pareto optimal policy. This confirms the intuition that technical interventions, such as fairness constraints, often do not sufficiently address persistent underlying inequities. Then, we model the effects of three different types of interventions: (1) policy changes in the algorithm’s decision threshold, and external changes to parameters that govern the downstream effects of late payment for (2) the whole population or (3) disadvantaged subgroups. We show how different interventions are recommended depending on the difficulty of enacting structural change upon external parameters and depending on the policymaker’s preferences for equity or efficiency. Counterintuitively, we demonstrate that preferences for efficiency over equity may sometimes lead to recommendations for interventions that target the under-resourced group alone. Finally, we simulate the effects of interventions on a dataset that combines HMDA and Fannie Mae loan data. This research highlights the ways that structural inequality can be perpetuated by seemingly unbiased decision mechanisms, and it shows that in many situations, technical solutions must be paired with external, context-aware interventions to enact social change.