Key facts
- Mortgage lenders are considering new credit scoring models like FICO 10T and VantageScore 4.0.
- Alternative data, including rental payments and utility records, is being explored to assess borrowers with limited traditional credit history.
- The integration of Artificial Intelligence (AI) into mortgage processes, such as document processing and analytics, is under consideration.
- Lenders face challenges in balancing the cost of new technologies and data with performance, borrower access, and regulatory requirements.
- The discussion occurred at MISMO's Fall Summit, featuring executives from FICO, VantageScore, Experian, Equifax, and TransUnion.
Mortgage lenders are navigating a significant shift in credit assessment, with a focus on integrating new scoring models, alternative data, and artificial intelligence. At MISMO's Fall Summit, industry executives debated the extent to which lenders should adopt these modernizations.
Key to this evolution are updated credit scoring models like FICO 10T and VantageScore 4.0, which incorporate trended credit data to provide a more comprehensive view of consumer behavior. Panelists agreed that while more data can improve credit decisions, lenders must carefully consider factors such as model performance, cost, technological readiness, investor acceptance, and regulatory compliance.
Alternative data sources, including rental payments, utility bills, and consumer-permissioned banking data, are seen as crucial for evaluating borrowers with thin traditional credit files, such as those in the gig economy or with non-traditional income streams. Executives emphasized that the goal is to assess risk more effectively rather than simply increasing risk tolerance.
The application of AI in mortgage lending was also a central theme, with discussions around its use in document processing, data extraction, and analytics. However, questions were raised about potential licensing requirements if AI systems begin to function as loan officers, highlighting the need for robust governance and compliance.
Ultimately, the industry is moving beyond the question of whether to use more data and is now focused on how and where to deploy it within their workflows to improve decision-making without adding undue complexity.