The Fairness of Credit Scoring Models

46 Pages Posted: 18 Feb 2021 Last revised: 27 Apr 2021

See all articles by Christophe Hurlin

Christophe Hurlin

University of Orleans

Christophe Pérignon

HEC Paris - Finance Department

Sébastien Saurin

University of Orleans, Laboratoire d'économie d'Orléans, Students

Date Written: April 27, 2021

Abstract

In credit markets, screening algorithms discriminate between good-type and bad-type borrowers. This is their raison d’être. However, by doing so, they also often discriminate between individuals sharing a protected attribute (e.g. gender, age, race) and the rest of the population. In this paper, we show how to test (1) whether there exists a statistical significant difference in terms of rejection rates or interest rates, called lack of fairness, between protected and unprotected groups and (2) whether this difference is only due to credit worthiness. When condition (2) is not met, the screening algorithm does not comply with the fair-lending principle and can be qualified as illegal. Our framework provides guidance on how algorithmic fairness can be monitored by lenders, controlled by their regulators, and improved for the benefit of protected groups.

Keywords: Fairness; Credit scoring models; Discrimination; Machine Learning; Artificial Intelligence

JEL Classification: G21, G29, C10, C38, C55

Suggested Citation

Hurlin, Christophe and Pérignon, Christophe and Saurin, Sébastien, The Fairness of Credit Scoring Models (April 27, 2021). HEC Paris Research Paper No. FIN-2021-1411, Available at SSRN: https://ssrn.com/abstract=3785882 or http://dx.doi.org/10.2139/ssrn.3785882

Christophe Hurlin

University of Orleans ( email )

Université d'Orléans
Rue de Blois B.P. 6739 45
France

Christophe Pérignon (Contact Author)

HEC Paris - Finance Department ( email )

1 rue de la Liberation
Jouy-en-Josas Cedex, 78351
France

Sébastien Saurin

University of Orleans, Laboratoire d'économie d'Orléans, Students ( email )

Orléans cedex 2
France

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
197
Abstract Views
555
rank
185,337
PlumX Metrics