Publications
Framework for Managing Machine Learning Models in Consumer Credit Underwriting
Report Summary
This report details key considerations and approaches for managing machine learning underwriting models. The report reflects discussions with banks that participated in a Technology Working Group convened under the Office of the Comptroller of the Currency’s Project REACh (Roundtable for Economic Access and Change).
The report summarizes potential benefits of using machine learning underwriting models and describes how industry practices are evolving, while recognizing that data science techniques are continuing to progress and that lenders vary in their strategies and technical implementation. The broader goal of the document is to provide a practical framework outlining core risk management principles, processes, and issues that are helpful to consider in adopting machine learning models.
Acknowledgments
This framework was developed through discussions with banks that participated in a Technology Working Group convened under the Office of the Comptroller of the Currency’s Project REACh (Roundtable for Economic Access and Change). It is intended to foster greater understanding among financial institutions, agency staff, and other stakeholders about how machine learning technologies are being used responsibly to facilitate innovation in credit underwriting. FinRegLab served as the co-chair of the working group and facilitator of the framework drafting process.
Related Publications
-
Advancing the Credit Ecosystem: Machine Learning & Cash Flow Data in Consumer Underwriting
This empirical white paper assesses the impacts on model predictiveness and credit access of both adopting machine learning techniques and incorporating electronic bank account information (often called cash flow data) in consumer underwriting models.
-
Explainability & Fairness in Machine Learning for Credit Underwriting: Policy Analysis
This policy analysis explores the regulatory and public policy implications of the increasing use of machine learning models and explainability and fairness techniques for credit underwriting in depth, particularly for model risk management, consumer disclosures, and fair lending compliance.
-
Machine Learning Explainability & Fairness: Insights from Consumer Lending
This empirical white paper assesses the capabilities and limitations of available model diagnostic tools in helping lenders manage machine learning underwriting models. It focuses on the tools’ production of information relevant to adverse action, fair lending, and model risk management requirements.