Not If Affordability Data Adds Value But How to Add Real Value by Leveraging Affordability Data: Enhancing Predictive Capability of Credit Scoring Using Affordability Data

Abstract

The aim of this paper is to focus on the role of affordability of loans on credit risk at 3 different levels. The first being short term practical approach to optimally using affordability data to improve current credit scorecards by showing how Random Forests can be used to improve logistic regression. After this discussion we step back and look at how affordability fits into ensuring systemic risk and use the credit union industry as a base case model for using affordability measures to reduce systemic risk. Finally we turn to the question of whether credit risk is a source of sustainable competitive advantage and what measures would create greater welfare for the financial system as whole. It is the belief of the authors that proprietary black box credit systems and increasing complexity of unpredictable factors make it necessary to build an open credit systems viewpoint. Using this weltanschauung we conclude with a simple example of restructuring mortgage products using affordability and borrower well being as central to sustainable credit risk management. The article is comprised of survey and quantitative analysis showing how affordability has been used in the credit industry or not used. The article shows that credit decisions can be improved by leveraging affordability optimally and then also discusses policy implications of using affordability to ensure soundness and well being of consumers instead of simply determining whether the consumer will repay the loan successfully.