Defaulting borrowers: How well did you know them?

Recently The Wall Street Journal’s small business section carried a story about the correlation of loan defaults and how well a lender knows his customer. What the author had to say makes an interesting jumping-off point for discussing the remaining human factor in commercial lending—indeed, any bank lending.

Automation versus eye-to-eye

The article’s source data were portfolios of SBA loans. The author made the distinction between loans underwritten based exclusively or primarily on a credit score vs. loans underwritten by a credit analyst or loan officer.

The author noted that a preponderance of loans made in urban areas were based on credit scoring systems, while loans to borrowers in exurban and rural areas tended to be underwritten manually (the “old-fashioned” way).

Default rates on “scored” loans were about 23% higher than loans individually underwritten. So, statistically, loans between rural borrowers and rural lenders were of a higher “quality,” as measured by the sole criterion of default.

This led the author to two interesting conclusions, one obvious and the other less so:

• Lenders who knew their borrowers directly, face-to-face, so to speak, had lower defaults based on that closeness and first-hand knowledge and personal judgments of creditworthiness.

• Loans between borrowers and lenders who had stronger social ties had lower rates of default.

One might argue that these are different sides of the same coin. How well I know my borrower is a function of how closely knit relationships tend to be in rural areas where they certainly tend to be firmer and more interrelated than in urban settings.

If I know my borrower from Sunday school and through attending Friday night football at the local high school, then it follows that I very likely know him from face-to-face business interactions as well.

Direct vs. indirect lending

There are potential lessons to take away from this discussion.

First, if I know my borrower, he’s less likely to be a collection problem for me.

Any old-school lender knows from experience that he will have a qualitatively superior result from a portfolio of directly negotiated loans than he will from customers whom he hasn’t met or doesn’t know very well.

This is the difference between what lenders call “direct lending” and “indirect lending.” 

Many years ago my bank in Albuquerque bid on a portfolio of credit card debt that had been sold and originated on an affinity basis. We bought the portfolio and the right to underwrite all new credit in New Mexico and West Texas that was originated by the affinity program.

It was a classy and respected brand name—but the business was lousy.

Really lousy.

We underwrote according to our normal credit card standards but experienced lower approval rates based on our expectations. The losses were not expected. 

Virtually all credit card underwriting is indirect. We seldom have a first-hand relationship with the borrower.

Indirect auto lending has been a feature of big bank credit for years. The car buyer selects the car, the dealer sends the contract to you or me for underwriting. If we approve the credit, the buyer drives the car home and never visits the bank. The loan is closed in the showroom.

Though I’m not experienced in that area of credit, I’ve participated in many discussions relating to portfolio quality and where the indirect portfolio simply doesn’t quite perform as well as the direct portfolio.

Implications for banking broadly

There are potentially significant issues to consider for the future of consumer and small commercial loan underwriting. As bank credit business models move away from the individually negotiated credits, the personal ties with borrowers becomes increasingly tenuous.

This is a profitable business model for a great many banks and it’s here to stay for most, as they can build the cost of losses into the rate as a cost of doing business.

But it has important implications for community banks that seek to develop strong relationship-based credit portfolios.

There’s no reason that an urban or suburban community bank couldn’t act like its country cousins and seek to develop the ties that bind the customer more firmly to the bank.

It’s more labor intensive, to be sure. But the results should be predictably better in the long run.

In my many years as a lender, I’ve seen the customers become more distant, more removed from the bank. But I can also say that the customer prefers the closer relationship and usually is pretty vocal about letting you know how he appreciates the difference.

Would you start out as a banker today?

In the long run, our credit portfolios have probably suffered a degree of degradation as this trend towards the indirect has ground along relentlessly.

It’s probably not a safety and soundness issue per se, but it may be a reflection on how and whether many of us would choose prospectively to be career lenders.

I personally wouldn’t care to be the central underwriter. I want to be the borrower’s consultant, mentor, and friend and that’s the social dimension to the job and to the inherent quality of the credit.

Now, the current direction of things is something community bankers can exploit. There are obvious implications of the prevalent credit business models. Community bankers have a possible new niche to explore and one the customer will not resist. He or she wants this sort of relationship as much as we do.

Maybe the future of community banking and community bank lending lies building on what’s worked so well in the past. The personal touch will never be obsolete.