How Costly is Noise? Data and Disparities in Consumer Credit
This addresses inefficiencies and inequalities in consumer credit, particularly for disadvantaged groups, with incremental improvements in quantifying and mitigating disparities.
The paper tackles the problem of information disparities in US credit markets, showing that credit scores are noisier indicators of default risk for historically under-served groups, and finds that equalizing score precision can reduce approval rate disparities and credit misallocation by about half.
We show that lenders face more uncertainty when assessing default risk of historically under-served groups in US credit markets and that this information disparity is a quantitatively important driver of inefficient and unequal credit market outcomes. We first document that widely used credit scores are statistically noisier indicators of default risk for historically under-served groups. This noise emerges primarily through the explanatory power of the underlying credit report data (e.g., thin credit files), not through issues with model fit (e.g., the inability to include protected class in the scoring model). Estimating a structural model of lending with heterogeneity in information, we quantify the gains from addressing these information disparities for the US mortgage market. We find that equalizing the precision of credit scores can reduce disparities in approval rates and in credit misallocation for disadvantaged groups by approximately half.