Signature Verification Evolution To Exception Check Review
Signature Verification Evolution To Exception Check Review
Background:
Two decades ago, most banks reviewed nearly 100% of the signatures of
check makers when they paid and manually filed checks in file drawers in the
branch. As banks grew in size and as the number of accounts, customers, and
checks processed dramatically increased, the need grew for more controlled,
reliable, and efficient methods for processing and paying checks. Accordingly,
check storage was centralized and automated in large processing centers where
the signatures of some, but not all, items were compared with the signatures on a
signature card, a microfilm copy of the signature card, or a stored digital image of
the signature card. With the move of the check pay-and-file process from the
branches into large processing centers and the increase in check volume,
signature verification changed from review of all signatures to a review of a
portion of the signatures – usually large dollar transactions, accounts with special
arrangement to check multiple signatures, and a random sampling of all checks,
regardless of amount.
Current Practices
Clearly, with over 65 billion checks written each year in the United States,
it is not practical or cost-effective to attempt to compare signatures on the
hundreds of millions of checks processed daily. Still, many customers and law
enforcement and judicial system officials are unaware of the processing changes
related to signature verification that have taken place over the last decade.
Some do not realize that no large banks and few, if any, small banks verify all
signatures. They may also be unaware of the trend among the largest banks
away from verification of random signature to a check review, which includes
signature verification, of only items identified as risky.
Banks have come to recognize factors beyond their control that prevent
them from identifying fraudulent items based on signature verification:
• Signatures change over time and no longer match the file copy;
• Use of facsimile signature plates makes detecting forgeries difficult and
subject to individual interpretation;
• New scanner technology makes forging signatures very easy, affordable, and
undetectable by normal verification.
These factors reduce significantly the success rate in detecting forged signatures
solely by comparing signatures.
In contrast, many banks, particularly large ones, have found that a review
of several components of the check, including signature verification, of only those
checks identified as suspicious or risky is more effective in detecting fraud than
random signature verification. Before comparing signatures and reviewing other
components of checks, these banks, using fraud filters, identify suspicious items
based on criteria developed from fraud experience.
2
Once identified by the fraud filters, suspect checks are given extra special
attention and thoroughly examined to determine their validity. As one of many
components of the check review, the signature on the check is compared to the
signature card or signature database to determine whether it may be forged.
Large banks have found this approach very effective in identifying fraudulent
items. In the survey of large banks noted earlier, participating banks uniformly
ranked this approach four or five out of possible five in terms of effectiveness.
One bank found that it reduced its check fraud losses $16 to $20 per dollar spent
using this approach. Not only do many banks find such programs more efficient
and accurate for identifying fraud, they find that they protect their customers from
fraud losses.
Conclusion