Well GonzoBankers, finding reliable statistics on check fraud losses in the banking industry has proven to be more difficult than pitching a perfect baseball game. At least Randy Johnson, ace 40-year-old left-handed pitcher for the Arizona Diamondbacks, showed us that can be done.
Give this a try: put “check fraud statistics” into your favorite search engine and see what you get. Google returned more than 269,000 hits. After reviewing the first few it is obvious that:
Attempts to reduce check fraud have historically relied on training both bank and business employees. On the front line of businesses accepting checks for payment, spotting the bad check depends on the check-out clerk examining the document and identifying the customer. Most businesses have relied on a national or regional check system to validate presented items. The gain initially recognized by this process is quickly being diminished by the increasing numbers.
Bank tellers have been the front line for reducing check fraud. Much of the same training provided businesses applies to the teller. “May I see your driver’s license or other form of identification” is a common request for an unknown customer. Suspect customer lists and potential fraudulent check lists are helpful but are often set aside under the pressure of long lines and rising customer temperatures.
Manual processes that rely on employees to examine documents and forms of identification are obviously not the answer. A few automated lists of bad check writers aren’t enough either. The inherent problem is clear:
Perhaps the billions of dollars the financial industry has invested in technology will provide the means for reducing check fraud. A number of maturing technologies coupled with recent legislation may be the key.
Exceptionally fast, affordable computers, cheap high volume storage, high speed image cameras, image processing software technology, and Check 21 have all collided at this time. Products marrying these pieces are beginning to appear in service offerings from vendors.
We have witnessed image item processing and power proof make an impact on the item processing business. Image proof is possible because of the very sophisticated image processing software developed to interpret the courtesy or legal written amount. Initial read rates were disappointing but now often exceed 75 percent on all work and more than 90 percent on printed business checks. Very complex software routines involving the execution of millions of computer instructions are required to “read” the amounts. Processing large volumes of checks requires the use of extremely fast computers. This technology has been honed on one particular part of the check; the same type of logic can also interpret the check’s signature and compare it to a sample stored as part of the old paper signature card. With this technology, it becomes possible to compare signatures on every check processed instead of only the “high dollar’ items.
One of the best anti-fraud strategies, positive pay, depends on an “issue” file from the bank’s customer matched against paid items. To date, the only comparison has been check number and amount. Image technology now makes it possible to validate signatures, check stock and other identifiable features. Add an encrypted, discrete image on the check face that can be decoded to provide “issue” information and the need for an issue file disappears.
Fiserv’s “FraudGuard” products from its ImageSoft business unit are providing this capability today. Other vendors’ products with similar capabilities are becoming available or are in the development stage.
Check 21 is the catalyst to move most checks to image format over the next few years. Today’s paper jam will likely become an image item super highway in the next few years. Unfortunately, most of these products are deployed in the back office and may not provide the opportunity to stop the bad check before it ever enters the payment system.
The real win will push these capabilities to the front office. Think about this scenario:
Oh, I forgot to mention, if this is a bad item, it can be declined and never enter the bank payment system.
The good news is that some real solutions to check fraud may be on the way. The bad news is it will be a while before mainstream vendors will provide well integrated solutions to their product sets. Quite likely the banks (or credit unions) with deep development pockets will spend their integration dollars in an effort to reduce fraud. Their efforts, and the many arrows that will be protruding from their backs, will be reflected in updated versions of future products.
Keep your hats on; this is going to be interesting. You should have a little talk with your payment processing, risk and IT groups. They need to be watching this technology and be prepared to invest when the investment can be justified through reduced losses due to check fraud.