It's Time to Kill Camels

Editor's note: This post originally appeared in the June issue of American Banker Magazine.

When a bank fails and the resulting loss to the Deposit Insurance Fund crosses a certain threshold, regulators are required to write a material loss report explaining why the bank in fact failed. Well over 100 such MLRs have been written in the past few years.

Each report includes a recent history of the Camels ratings—the supervisory regulator's assessment of Capital adequacy, Assets, Management capability, Earnings, Liquidity and Sensitivity to market risk—that had been assigned to the bank.

Examine the MLRs and you'll notice two interesting things about these failed banks. First, they share the common problem of having grown too fast, typically in markets where competitors also grew too fast. Second, the banks' examiners almost without exception gave these institutions good to excellent Camels ratings nearly right up until the point of failure.

Start all the way back at 1979—the year the Camels system was developed—and the total number of bank and savings and loan failures is now north of 3,400 institutions, including more than 470 that have failed just since 2008.

It is time to kill Camels. Unreliable, unpredictable, and outright misleading, this system must be replaced by something that is more than a clever acronym.

A good example of the uselessness of Camels ratings is a case study of Bank of Clark County in Washington state. Closed in January 2009 with a shade under half a billion dollars in assets, the bank had grown an average of 17% a year since its founding in 1999. A big slug of the growth occurred from 2004 to 2007, when Bank of Clark County's balance sheet ballooned 73%.

What were the bank's Camels ratings during that time? Answer: straight 2s, on a scale of 1 to 5 for each category covered by the Camels system, with 1 being great and 5 being brain dead. These were strong ratings, and they included a full-scope examination conducted by the Federal Deposit Insurance Corp. in June 2007.

By the time the state reported the bank's Camels ratings in November 2008, the scores had deteriorated to 5,5,5,5,5,4/5. Two months later the bank was closed, leading to a $131 million loss to the insurance fund.

Sadly, the pattern of scores seen at Bank of Clark County is typical of findings in virtually all the other failed banks for which MLRs are written.

This isn't a matter of implementation or execution. The problem is the system itself. The evidence of more than 3,400 failed banks in 25 years—and the ratings these institutions received shortly before folding—simply overwhelms any defense of the usefulness of Camels.

The United States needs a new, 21st century model for identifying, monitoring and reporting risk at banks. The model must be forward-looking, not backward-looking as the Camels ratings are; it must consider a bank's momentum as well as the emerging risks. It also should be transparent to bankers, depositors, and investors in bank equity and debt. Finally, the new model should be based on objective measures.

My analysis of failed banks shows that there are several factors worth focusing on. There's loan and asset concentration risk, of course. But two other important areas relate to "velocity," i.e. how fast a bank's loans are growing and how fast loans are growing overall in the bank's geographic market. A good rule of thumb: banks and markets with fast loan growth are more risky than banks and markets where loans are growing at a more constrained pace.

Though requiring great judgment on the part of regulators, perhaps the most important factor of all is the need to evaluate the skill and experience of a bank's board of directors to oversee strategy, management selection, risk appetite and controls. Banks that plan to take on strategic risk and grow through acquisitions should have superior board ratings.

We clearly need a process that takes all of this into account, to replace what we have now in Camels, and bankers and regulators should take the initiative and collaborate in designing it.

Richard J. Parsons, a former Bank of America executive, is the author of "Broke: America's Banking System —Common Sense Ideas to Fix Banking in America."