How can lenders ensure they’re making the most accurate and fair lending decisions? The answer lies in consistent model validations.
What are model validations?
Model validations are vital for effective lending and risk-based pricing programs. In addition to helping you determine which credit scoring model works best on your portfolio, the performance (odds) charts from validation results are often used to set score cutoffs and risk-based pricing tiers. Validations also provide the information you need to implement a new score into your decisioning process.
Factors affecting model validations
Understanding how well a score predicts behavior, such as payment delinquency or bankruptcy, enables you to make more confident lending decisions. Model performance and validation results can be impacted by several factors, including:
- Dynamic economic environment – Shifts in unemployment rates, interest rate hikes and other economic indicators can impact consumer behavior.
- Regulatory changes affecting consumers – For example, borrowers who benefited from a temporary student loan payment pause may face challenges as they resume payments.
- Scorecard degradation – A model that performed well several years ago may not perform as well under current conditions.
When to perform model validations
The Office of the Comptroller of the Currency’s Supervisory Guidance on Model Risk Management states model validations should be performed at least annually to help reduce risk. The validation process should be comprehensive and produce proper documentation.
While some organizations perform their own validations, those with fewer resources and access to historical data may not be able to validate and meet the guidance recommendations. Regular validations support compliance and can also give you confidence that your lending strategies are built on solid, current data that drive better outcomes. Good model validation practices are critical if lenders are to continue to make data-driven decisions that promote fairness for consumers and financial soundness for the institution.
Make better lending decisions
If you’re a credit risk manager responsible for the models driving your lending policies, there are several things you can do to ensure that your organization continues to make fair and sound lending decisions:
- Assess your model inventory. Ensure you have comprehensive documentation showing when each model was developed and when it was last validated.
- Validate the scores you are using on your data, along with those you are considering, to compare how well each model performs and determine if you are using the most effective model for your needs.
- Produce validation documentation, including performance (odds) charts and key performance metrics, which can be shared with regulators.
- Utilize the performance charts produced from the validation to analyze bad rates/approval rates and adjust cutoff scores as needed.
- Explore alternative credit scoring models to potentially enhance your scoring process.
As market conditions and regulations continue to evolve, model validations will remain an essential tool for staying competitive and making sound lending decisions.
Ready to ensure your lending decisions are based on the latest data? Learn more about Experian’s flexible validation services and how we can support your ongoing success. Contact us today to schedule a consultation.
