Report Suggests Anti-Fraud Measures Are Having Impact

20041122c4cdxgq4-1-112304checks.jpg

Though attempts to steal money from banks by using bogus checks increased sharply in 2003, actual losses caused by fraud dipped a little, according to a survey report issued Monday by the American Bankers Association.

Processing Content

Bankers and consultants said that the Washington trade group’s report indicates that banks are doing a better job of spotting fraud.

“This is not a bad story,” said Nessa Feddis, the association’s senior federal counsel.

The ABA found that the banks lost $677 million to check fraud in 2003, 3% less than in the previous survey, in 2001. But criminals attempted to pass bad checks worth $5.5 billion last year, up 28% from the previous study.

“You’re always going to have some fraud,” Ms. Feddis said in an interview Monday. “Fraud is dynamic. As soon as the fraudsters see a roadblock one place, they move.”

The American Bankers Association Deposit Account Fraud Survey Report is closely watched by bankers seeking to protect their customers’ assets and measure their results against their competitors’.

“I hope this is reassuring to customers,” said Shirley Inscoe, a senior vice president at Wachovia Corp. and its manager of strategic support for loss management. “This shows we are taking this very seriously, and taking steps to stay ahead of the crooks.”

The total value of attempted check fraud was up considerably, but the number of cases grew only 3%, to 616,469, as criminals attempted to pass checks of higher value.

Counterfeit checks had the highest median loss per case, at $3,059, but forgery was the most common type of fraud, with 24% of cases attributed to forged signatures and 7% to forged endorsements. Bounced checks that are never made good were the second-largest source of losses, at 17%, and counterfeit checks were third, at 15%.

Banks, especially the biggest, have spent heavily to fight fraud, the survey found. It said the median expense per bank ranged from $1 million to $9.9 million for money-center banks (defined as those with assets of $50 billion or more); $1 million or more for regional banks (assets of $5 billion to $49.9 billion); $50,000 or more for midsize banks (assets of $500 million to $4.9 billion); and less than $5,000 for community banks (assets of less than $500 million).

One in five money-center banks spent more than $20 million each on operations related to check fraud, not including actual losses, the survey found.

The incidence of account-opening fraud increased as a share of total fraud losses, after showing signs of decreasing in the 2001 report. The problem grew for banks of all sizes but was particularly troublesome for community banks, accounting for 44% of their check-fraud losses in 2003, versus 32% in 2001. New-account fraud also made up a larger share of the fraud cases, regardless of bank size.

Ms. Feddis noted that the survey covered all of 2003, before new federal rules that could limit this problem were implemented. Section 326 of the USA Patriot Act, which imposed new customer identification requirements designed to catch terrorists and money launderers, took effect Oct. 1, 2003. The survey was conducted between March and July of this year.

Survey participants said account-screening software for account opening was the most effective method of fraud prevention, followed by the use of credit bureau scores at account opening, refusing to cash checks for noncustomers, and developing a centralized fraud management system. Positive pay — a product for comparing a business check against a list prepared by the company of the checks it wrote — fell to No. 5 after being ranked as the most effective method in the 2001 survey.

The survey produced signs that fraudsters might be attempting to avoid the biggest banks (with the most imposing defenses) and attacking smaller ones instead.

“To some degree that has happened, but not to the degree I would have expected,” Ms. Feddis said. Smaller banks have spent less on computerized fraud-detection systems, but “they don’t necessarily need them.”

Instead of relying on automated systems, smaller banks may know their customers well enough that branch employees and other operations staff could recognize suspicious checks, she said. “It may be an individual rather than the computer” that raises a flag.

Bankers are also taking cooperative steps. Herbert H. Hilliard, the executive vice president of risk management at the $28.3 billion-asset First Horizon National Corp. of Memphis, said the Tennessee Bankers Association had spearheaded a program two years ago to encourage all its members to fingerprint noncustomers who were cashing checks in their branches.

The program has led to some prosecutions, as banks were able to identify perpetrators by their fingerprints, Mr. Hilliard said in an interview Monday. But that is not the chief value of the program, he added. “I think it’s more of a deterrent. If you are a criminal, the last thing you want is for us to have your fingerprints.”

Indeed, policy may be as important as technology in combating check fraud, said Ariana-Michele Moore, an analyst at the research and consulting firm Celent Communications LLC of Boston.

“A lot of fraud can be deterred by banks changing their policies and practices,” she said. For example, banks could limit the size of withdrawals at the automated teller machine that are made in conjunction with an ATM check deposit. “That right there reduces their check-fraud exposure,” she said.

The issues will probably change as banks accept more deposits in the form of check image. The technology should reduce fraud in the long run, Ms. Moore said, but, “I think there will be a gap where more fraud gets through.”


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