New Credit-Risk Model Integrates Debit Data to Better Predict Card Delinquency

A new behavioral credit-risk model developed by researchers from BI Norwegian Business School and NHH Norwegian School of Economics integrates credit and debit transactions to significantly improve prediction of credit card delinquency while offering clearer insight into the behavioral drivers behind repayment problems. The study, published in The Journal of Finance and Data Science, demonstrates that combining credit card data with customers' debit transactions substantially enhances the ability to predict which cardholders are at risk of missing payments.

First author Håvard Huse explains that credit data alone provides only a partial picture of a customer's financial situation. By integrating debit transactions, researchers gain insight into payday spending patterns, repayment behavior, and income flows—factors that strongly influence whether someone is at risk of missing credit card payments. The research team, which also includes Sven A. Haugland and Auke Hunneman, developed a hierarchical Bayesian behavioral model that consistently outperforms leading machine-learning algorithms including XGBoost, gradient boosting machines, neural networks, and stacked ensembles.

The study draws on detailed credit and debit transaction data from a large Norwegian bank, moving beyond traditional credit-risk models that rely heavily on monthly aggregates such as balance and credit limit. These conventional measures fail to reveal how customers actually manage their finances day-to-day. By capturing behavioral dynamics—including how repayment patterns evolve over time and how spending spikes after payday—the new model explains both why delinquency occurs and who is likely to default.

The model improves prediction accuracy at the individual level and identifies distinct behavioral segments with different "memory lengths," referring to the extent to which past financial states affect current repayment behavior. Co-author Auke Hunneman notes that customers in financial distress tend to be more influenced by earlier months' behavior, and the new model captures this dynamic far better than standard machine-learning tools. The research findings are detailed in the study available at https://doi.org/10.1016/j.jfds.2025.100166.

Beyond superior predictive performance, the team's approach offers greater interpretability than state-of-the-art algorithms. Hunneman emphasizes that banks need not only accurate predictions but also understanding of which behavioral patterns drive risk. The practical value of the model becomes evident when using a three-month prediction horizon, where early detection of at-risk cardholders could generate substantial cost savings by enabling timely intervention and reducing losses.

Co-author Sven A. Haugland notes that for financial institutions, this represents more than just an accuracy improvement—it provides a way to proactively help customers avoid serious financial problems through early intervention. The findings highlight an emerging shift in credit scoring from traditional static models toward richer behavioral analytics based on a complete picture of customer transactions. This approach could transform how financial institutions assess risk and support customers, potentially reducing delinquency rates and improving financial health outcomes across the banking industry.