8–9 Minutes to Read
Underwriting credit risk involves two distinct activities: gathering data and developing predictive models. The role of gathering data is largely a matter of opportunity—access to the data. The role of developing predictive models, on the other hand, is a technical skill.
Often, the party that has the best access to data is not the one with the best technical skills and vice versa. Local banks, credit unions and merchants have great data to underwrite small businesses and consumers, but they often don’t have the technology and expertise to match the most sophisticated, specialized non-local lenders. Likewise, the sophisticated non-local lenders with the best data analytics, often do not have the best data.
The typical solution to this problem is for each party to purchase what it doesn’t have. The local lenders buy underwriting models and the non-local lenders buy data. The problem is that the purchased models aren’t optimized for the lender’s exact data, and the purchased data may be incomplete or out of date by the time a lending decision is made.
Today, lenders with data and lenders with expertise rarely cooperate in the loan origination process. Both parties are seeking to originate loans from the same customers with sub-optimal results.
The solution to this problem is “symbiotic underwriting” in which two underwriters, one with the best data and one with the best expertise, partner to originate loans.
The challenge with symbiotic underwriting, as with symbiotic relationships in nature, is for the benefits to be mutual, as opposed to parasitic. Each party must benefit.
Imagine the following arrangement: a local lender has unique access to data on local small businesses. A non-local lender has unmatched expertise in underwriting small businesses. They partner to create a platform in which both parties make loans to the local lender’s customers using the combined data and underwriting expertise of each party.
So, what are the risks?
The primary risk for the local lender is that the non-local lender will seek to poach their customers. The local lender will be concerned that non-local lender will make one loan to a customer and then seek to continue to lend to the customer outside of the partnership. Another big risk for the local lender is that the non-local lender will not provide a good underwriting experience for its customers, which might jeopardize the relationship.
The risk for the non-local lender is adverse selection. The non-local lender will be concerned that the local lender will cherry-pick the best borrowers.
If the partnership is not structured properly, the mutual benefits of sharing each entity’s unique assets will not be realized.
So how do you structure a mutually beneficial symbiotic underwriting relationship? Every situation is different, but here are some suggestions:
- Loan ownership: Each party should own its loans directly, rather than through a partnership entity. This will allow for maximum flexibility in relation to volume adjustments between the parties.
- Cooperation, not competition: The loan selection and pricing process should be cooperative, rather than competitive. Consider giving the local lender priority in making loans at or below a price set by the non-local lender. Trust and the volume capacity of the local lender will be key factors in setting the ordering rules for which party makes which loans.
- Partner wisely: Local lenders should consider the business growth strategy of the non-local lender. Does the non-local lender want to eventually market directly to borrowers, such that they may be a competitor in the future? For banks, third-party risk compliance requirements will also play a major role in partner selection.
- Restrict solicitation: The non-local lenders should be restricted from directly soliciting customers of the local lender who were met through the partnership arrangement.
- Customer experience: When practical, the non-local lender should be required to provide pre-approved offers that can be accepted without any additional underwriting. Use a servicing platform that is branded for the local lender. Agree upon servicing and collections procedures at inception and require approval of the local lender for any changes.
- Data sharing: The local lender should be required to include data from all potential borrowers within the target customer groups, without the ability to pick and choose.
- Model improvement: Loan performance data for both parties should be shared for purposes of improving the underwriting model. Representatives from both parties should participate in the model improvement process.
- Maximize data: The local lender should seek to become their customer’s centralized store of data. Personal financial management tools for individuals and financial accounting and treasury management tools for businesses are great ways to deepen the relationship, as well as automate the data sharing function with the non-local lender.
For more information on opportunities to partner with other lenders or to learn more about these types of arrangements, please contact Shane Hadden.
For more information on how community banks can increase revenue in the face of new technology, read Now is the Time for Community Banks and Full Circle: Technology Brings Loan Origination Back to Community Banks.
Upcoming articles will include a detailed discussion of profitability, examples in the market today, partner selection and management, compliance, and Community Reinvestment Act issues. To stay informed on BlackLine articles, please sign up here.