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 transaction data to substantially improve prediction of credit card delinquency. Published in The Journal of Finance and Data Science, the study demonstrates that this approach outperforms leading machine-learning algorithms while providing clearer insight into the behavioral drivers behind repayment problems.

First author Håvard Huse explains that credit data alone provides only a partial view of a customer's financial situation. "By integrating debit transactions, we gain insight into payday spending, repayment behavior, and income patterns—factors that strongly influence whether someone is at risk of missing payments," Huse states. The research team, which includes Sven A. Haugland and Auke Hunneman, developed a hierarchical Bayesian behavioral model that consistently outperforms algorithms such as XGBoost, GBM, neural networks, and stacked ensembles.

The study draws on detailed credit and debit transaction data from a large Norwegian bank. Traditional credit-risk models rely heavily on monthly aggregates like balance and credit limit, but these measures fail to reveal how customers manage their finances day-to-day. "By capturing behavioral dynamics—such as 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," Huse shares.

The model improves prediction accuracy at the individual level and identifies distinct behavioral segments with different "memory lengths"—the extent to which past financial states affect current repayment behavior. "Customers in financial distress tend to be more influenced by earlier months' behavior, and our model captures this dynamic far better than standard machine-learning tools," notes co-author Auke Hunneman. The approach not only performs better than state-of-the-art algorithms but is also more interpretable, addressing banks' need to understand which behavioral patterns drive risk.

Using a three-month prediction horizon, early detection of at-risk cardholders could generate substantial cost savings by enabling timely intervention and reducing losses. "For banks, this is more than an accuracy improvement—it is a way to proactively help customers avoid serious financial problems," says co-author Sven A. Haugland. The findings highlight an emerging shift in credit scoring from traditional static models toward richer behavioral analytics based on a full picture of customer transactions. The study is available at https://doi.org/10.1016/j.jfds.2025.100166.