WEEKLY ADVISER: Risk Models' Failure Shows'Character' Lending Viable

Much-publicized problems with hedge funds and the disappointing risk models of large banks should make community bankers more secure in relying on their more personal approach to lending and investing.

The stories of how banks-which are equipped with the latest computerized trading and investing tools-were unprepared for the volatility also pointed out how expensive this failure was. And the difficulties of Long-Term Capital Management and other hedge funds that relied on historical data to plan their portfolios further emphasize that computer models are nothing to bet the ranch on.

Sure, academic economics can point out that the problem often is simply a matter of time. If the shaky arbitrage positions had been allowed to remain open long enough, the recent wide spreads between risky and riskless assets would have narrowed back to traditional levels, and the hedge funds and banks betting on this trade would have made tons of money.

Being highly leveraged organizations, however, whose lenders wanted their money back as the securities used as collateral fell in value, they ran out of time. And like a roulette player who doubles up every time he loses, expecting eventually to break even, arbitrageurs may run out of capital and go bankrupt before the law of averages turns and works in their favor.

The reaction of community bankers to all this appears to have run the gamut from sadness for their fellow financial professionals to a private joy at their problems.

But where community bankers must have agreed was in feeling an affirmation for lending and investment practices that emphasize knowing individual borrowers.

Year after year the press and analysts talk about technical approaches to banking and how the community banker is behind the times. So it must have been pretty sweet to see these same pundits admit that the banks who felt rocket science should remain the province of NASA were not so out of date and poorly managed after all.

What is even more important to local bankers is the sense of relief at learning it is not a dereliction of duty if you don't play "follow the leader." All too often we have seen lenders get into trouble by assuming industry leaders know what they are doing.

Just following them is a familiar attitude, and doubting your own judgment if you disagree with what the leaders are doing is common, too.

I remember as far back as the Penn Central collapse about 30 years ago, when everyone had followed Citibank in supporting that ailing railroad. And when one young lender objected, the railroad people warned that they would report him to his management, which might fire him for disagreeing with his elders.

Some lenders who helped Long-Term Capital inflate its portfolio to $1.2 trillion must have lent out their money under the same philosophy-that it must be all right if the major banks were lending their own funds in such volume.

Sometimes supporting your own judgment takes courage.

One of my favorite stories is of the Florida bank that refused to write off a loan to a Holocaust survivor, as instructed by regulators, after his restaurant was demolished by Hurricane Gloria. Knowing that this man would meet his obligations whether there was collateral or not, bank officials preferred to leave the national bank system and take a state charter rather than do what they knew was wrong. How different this is from following a computer model either for arbitrage investment or, closer to home, using a credit scoring model to determine whether a loan should be made or not.

Bankers used to learn the four C's of credit-character, collateral, capacity, and conditions. Now I feel they face a world of three C's and a C-minus. Collateral has become tougher to repossess after a default, given the lax personal bankruptcy law and public sympathy for borrowers.

But maybe we should revise the four C's again and raise character from a C to an A. This is a lending tenet that many community bankers have used through the decades. And the hedge funds and banks using risk models have shown us, by their recent sad experience, that it is still a pretty good guiding star. Mr. Nadler, an American Banker contributing editor, is professor of finance at Rutgers University Graduate School of Management.

For reprint and licensing requests for this article, click here.
MORE FROM AMERICAN BANKER