Counterfactual Fairness in Mortgage Lending via Matching and Randomization

TL;DR

This paper addresses racial unfairness in mortgage lending by applying counterfactual fairness with a novel causal graph and matching approach, highlighting limitations of data balancing alone in achieving fairness.

Contribution

It introduces a new causal graph for HMDA data and employs a matching-based method to isolate race, advancing counterfactual fairness in mortgage lending models.

Findings

Unfairness exists in mortgage approval and interest rates between racial groups.

Matching balances data but does not ensure counterfactual fairness.

Counterfactual fairness requires more than data balancing.

Abstract

Unfairness in mortgage lending has created generational inequality among racial and ethnic groups in the US. Many studies address this problem, but most existing work focuses on correlation-based techniques. In our work, we use the framework of counterfactual fairness to train fair machine learning models. We propose a new causal graph for the variables available in the Home Mortgage Disclosure Act (HMDA) data. We use a matching-based approach instead of the latent variable modeling approach, because the former approach does not rely on any modeling assumptions. Furthermore, matching provides us with counterfactual pairs in which the race variable is isolated. We first demonstrate the unfairness in mortgage approval and interest rates between African-American and non-Hispanic White sub-populations. Then, we show that having balanced data using matching does not guarantee perfect counterfactual fairness of the machine learning models.