Member Business Lending: Three steps to early identification of increased risk of default
You rely on the ability to quickly return decisions to your members as a competitive advantage, but do you have the right early warning system to monitor the portfolio and reduce risk after the loan is on the books?
No credit union will ever be 100% successful in identifying all of its risky members before they default. They simply can’t afford to have staff continually review every loan in the portfolio, but everyone needs to do everything in their power to identify potential losses early enough to take action.
Credit unions can identify risks and reduce losses with a rules-based early warning system. The systems and tools must display the results of the analysis in formats and reports that provide users with actionable data and insights. How it is created and monitored is key.
A rules-based early warning system (RBEWS) is a model (or set of models) that predict increased risk of default. There is no denying that creating an RBEWS is a modeling exercise, which can create uncertainty and hesitancy among lenders who don’t have rigorous modeling experience. It is possible to achieve meaningful results using readily available statistical software. The bigger challenge for most financial institutions lies in collecting the data for the analysis. The question becomes, “Where to start?”
Developing and implementing a rules-based early warning system can be broken down into three primary efforts:
Constructing the early warning risk models.
- Create an RBEWS model that is focused on the right risk characteristics, identifies the systems and tools needed for the RBEWS to function properly, and demonstrates how to review results and how to act on those results.
Merging the model results into current credit-monitoring practices and procedures.
- Construct an RBEWS based on a bundle of risk characteristics geared toward the entity type (individual, small business, mid-market, large corporate), loan-product type (commercial real estate, secured and unsecured, etc.) and geography; how to incorporate “Hard,” “Soft” and “Judgmental” data into the process; when to perform a review; the loan reviewers responsibilities when using this tool; and how relationship managers should approach borrowers after the system has issued a warning.
Working with borrowers to reduce the credit union’s risk while preserving member relationships.
- This includes identifying a list of risk characteristics; managing the increased frequency of data reporting and field exams without generating an unmanageable amount of data; and improving an RBEWS over time by, for example, tracking false positives, tracking the credits flagged as heightened risk, and tracking the loans not identified, but that later become non-performing.
Your RBEWS will need to aggregate the required data in a single data set. It is very important to minimize the manual effort in creating this data set, since manual inputs are time consuming, costly, prone to error and not scalable. Your RBEWS will need to interface with multiple data sources, which ideally will include:
- Your core system
- Your financial statement spreading solution
- Consumer credit bureau reports
- Consumer and commercial credit-scoring services
- Other third-party data aggregators
Reviewing your portfolio to identify areas of decreased risk-specific borrowers, industry segments or product types—represents improved lending opportunities for your institution. A rules-based early warning system can help reduce losses by automatically performing a comprehensive portfolio review (at pre-determined times or as needed) and identifying the short list of default candidates.
Click here to learn how to construct and implement a rules-based early warning system in this three part eBook series.
For more information about the D+H portfolio of lending solutions, please contact us at 800-815-5592 or click here.
Contributing author: James Hartzog
