Non-maturity deposit risk under interest rate stress: a behavioral modeling framework

This paper develops a behavioral framework for modeling non-maturity deposit risk under interest rate stress within an interest rate risk in the banking book (IRRBB) context. The analysis focuses on the probability of withdrawal events, estimated through a multivariate logistic regression model with autocorrelated data. A key contribution of the paper lies in the use of financial market variables that are directly consistent with the supervisory interest rate shock scenarios prescribed by the Basel Committee on Banking Supervision. By relying exclusively on these “shockable†variables, the model ensures internal consistency between behavioral estimates and regulatory stress testing frameworks, avoiding the misalignment that may arise when macroeconomic variables are held constant under interest rate shocks. Empirical results show that short-term interest rates represent the primary driver of withdrawal risk, while yield curve slopes act as amplifiers of depositor responses. Additional effects are found for seasonal patterns and for the Covid-19 crisis, highlighting the regime-dependent nature of non-maturity deposit stability. Overall, the findings support an integrated view of interest rate and liquidity risk, showing that deposit stability is a market-dependent outcome rather than a structural balance-sheet characteristic. The proposed framework provides a practical and forward-looking tool for risk management and regulatory applications, allowing banks to assess funding stability under IRRBB stress scenarios.