For decades, academics have tried to determine what they call the “optimal size” of a bank. There is no consensus on an answer, but several studies show that costs drop for a bank as it grows to a certain size, after which they remain stable until the bank becomes massive and costs rise again. The exact asset size of the point at which costs level off varies among studies. According to theory, though, small banks should have some very good reason for existing, because they’re comparatively expensive to run.
A study presented last week at a conference on community banks at the St. Louis Fed offers one good reason. Academic work in the past has found that smaller banks are better with “soft information,” unquantifiable data that’s opaque to credit rating tools, but can help a loan officer make a decision. The study presented at the St. Louis Fed goes a step further, looking into which community banks do this best. The answer: rural banks. Call it the Bedford Falls principle.
Much of what we know about banking comes from, for better or worse, Frank Capra’s It’s a Wonderful Life. There’s a classic bank run in the movie, set in the fictional town of Bedford Falls. George Bailey, the banker, stops the run through a combination of personal savings and what economists call “social trust.” People in Bedford Falls like George. They decide to take only the money they need, enough working capital to get through the week, and leave him with the rest. The bank survives. The deposits remain on the books.
The authors of Small Business Lending and Social Capital (PDF) discovered that this trust works better for community banks in rural areas. The study compares the performance of small business loans at banks with assets of less than a billion dollars. That doesn’t sound small, but it is. The authors restricted their study to the years between 1984 and 2001, before standardized credit scoring became commonplace, and the data come from the Small Business Administration, which offers a partial guarantee on new companies, without an extensive credit history. All this means that the banks in the study were relying almost exclusively on soft information and social trust to make their decisions.
Loans between rural banks and rural borrowers, the study found, were only 70 percent as likely to default as those between urban banks and urban lenders. The numbers get even better for what the authors call “hyper-rural” banks and borrowers. And when the bank and borrower are in different rural markets, the default rate rises. It’s not that rural loan officers are generally smarter, or rural borrowers more trustworthy. It’s that when people know each other, loan performance improves.
“These findings may also help explain why small banks, and small rural banks in particular, continue to exist in the U.S. in disproportionate numbers,” the report concludes, “and are able to operate successfully at less than efficient scale.” Or: George Bailey’s little life had meaning after all.