Spread across states and active in all industry sectors, over 5 million businesses in the US employ fewer than 20. Their total sales of over $4 trillion and total employment of over 20 million is roughly comparable to the entire Canadian economy!
These very small businesses present lenders a market opportunity and a challenge. An opportunity because generating sales and employment at that level requires credit. For comparison, the total non-mortgage business loans outstanding in Canada is about $350 billion (C$ 450 billion). And a challenge because they are too many, too diverse, present sparse financial data, and borrow in smaller amounts. Lending to these businesses can be profitable if we use more appropriate techniques than those used when dealing with larger loans to larger businesses. To a large extent, the burgeoning marketplace lending for small businesses story is about developing these more appropriate techniques for gaining process efficiencies and improving risk assessment.
Credit Opportunity Map
What’s better? An 8% default rate or a 6% default rate? How about a 7% default rate that has increased from a previous 5% or the same 7% that has improved from a previous 9%? To help answer such credit risk questions, we leveraged the Business Default Index (BDX) and developed two measures – forecasted default rate and default gradient. The default gradient captures the direction and magnitude of change in default rate. It allows us to compare two borrowers with an identical default rate based on where they came from.
By leveraging the granularity of the BDX, we segment businesses by state and industry sector into nearly 1,000 groups and then cluster them based on the two default metrics—forecasted default rate and default gradient. With this methodology, 5 million businesses turn into 5 opportunity sets.
WAIN Street’s Credit Opportunity Map helps lenders gain a better understanding of the very small business credit market.Mapping businesses
The two axes use a standardized scale centered at 100 with a standard deviation of 15. On both axes, higher values correspond to greater risk. Each dot on the map represents a specific state & industry sector-based segment of businesses. Five clusters are identified with an additional for capturing segments that represent “Special situations” or outliers. The clusters are colored based on their distance from the origin with the closest ones being better credit risks.
The results are intuitive. Looking down the magenta, light green, and dark green clusters, we observe business segments that achieved similar default rates but from different directions. Similarly, looking across the light green, steel blue, and purple clusters, we observe segments that have experienced similar changes in their default rates but have achieved different levels of defaults.
Using the map
For marketing, the map provides micro-segments based on credit characteristics. To construct a conservative portfolio, choose from the dark green cluster. To showcase your underwriting skills, seek businesses from the purple and magenta clusters. With each dot representing a specific state and industry sector, marketing efforts can be more personalized.
For underwriting, the map provides additional actionable inputs concerning credit risk. Based on the Feb 2016 BDX, VT Information sector and GA Utilities sector businesses are the most attractive – they have the lowest default rates that are forecasted to improve. DE and UT “Health care & social assistance“ sector businesses are riskier and lenders should pay greater attention to firm-specific diligence. Similarly, AK Mining sector and ND “Management of Companies” sector businesses are riskier and require greater underwriting attention.
For portfolio management, the map provides a high-level perspective that helps identify segments for reshaping a portfolio. For risk managers, the micro-segments offer an effective means for identifying areas of the portfolio that require additional review or monitoring.
Conclusion
WAIN Street’s Credit Opportunity Map blends old-style techniques with big data approaches to help solve basic challenges that all small business lenders grapple with—where to expand; where to contract; and what to keep an eye out for.