Assessing Systemic Vulnerabilities in the U.S. Banking Core: A Predictive Analytics Model for Mitigating Third-Party Vendor Risk and Ensuring Federal Regulatory Compliance

• Adaan Ahsun

  Covenant University

  Author

The U.S. banking sector's deepening reliance on third-party service providers (TPSPs) for core operational functions has introduced a critical systemic vulnerability that existing risk management frameworks fail to adequately address. While traditional vendor risk assessments focus on financial viability and compliance checklists, they lack predictive capability to anticipate vendor failures before they materialize. This study bridges this gap by developing and validating a predictive analytics framework for identifying high-risk third-party vendors and forecasting potential disruptions. Using a quantitative design-based research approach, the study analyzes a comprehensive dataset of 5,000+ vendor relationships across 200 U.S. banks and credit unions, applying machine learning algorithms including Random Forest, Gradient Boosting, and XGBoost to predict vendor risk scores. The XGBoost model achieved 89.4% accuracy (AUC-ROC: 0.92, p < 0.001) in identifying vendors likely to experience security incidents or compliance failures within 12 months. Key predictors included vendor cybersecurity posture, financial stability indicators, regulatory history, and subcontractor dependencies. This framework provides a replicable, data-driven approach that enables institutions to proactively mitigate third-party risk, reduce potential losses by up to 60% through earlier intervention, and align with the Interagency Guidance on Third-Party Relationships. The findings inform both practitioner risk management strategies and regulatory policy development for monitoring systemic concentration risks in the financial sector.

• PDF

Copyright (c) 2026 Adaan Ahsun (Author)

This work is licensed under a Creative Commons Attribution 4.0 International License.