Check Fraud Isn't Going Away

I don't have to tell readers of this post that check fraud is not going away. The number of checks being issued has certainly declined, but run of the mill check fraud continues to present an ongoing threat to banks of all sizes.

Despite this, banks in general continue to apply rather rigid and quite frankly, arcane fraud detection rules. Most apply the standard suite of check fraud detection rules that establish a "red line” that once breached queues a check for review.

So why do investigators continue to pay fraudulent items? The answer to this question could fill a book (well almost). They failed to follow procedure. The fraud was too difficult to spot. The investigator did not spend the appropriate amount of time reviewing the item. However, sometimes, despite our best efforts to explain the error, the investigator just missed the item – plain and simple.

So really, why do investigators miss fraud? They miss fraud for all the reasons noted above. But, in addition, I believe that they miss fraud because they cannot see the "wood for the trees". The vast majority of check fraud suspects triggered by static fraud or "red line" rules are NOT fraudulent.

By way of example, I often met with new check fraud investigators after they had been with my department for about 3 months. During our discussions, new investigators would be concerned that they had not caught any fraud since joining the bank. At the other end of the spectrum, I would meet with experienced investigators that routinely identified check fraud and rarely, if ever missed a fraud item. They could tell you with absolute confidence why an item, an account, or type of check was fraudulent, and they could do so quickly.

So what caused the differences between investigators? Simply put, the new fraud investigator believes that all suspects are created equal and have an equal chance of being fraudulent. The experienced fraud investigator knows from experience that all suspects are NOT created equal and they act accordingly.

Some banks, can, and do revise check fraud rules to reflect the realities of what is actually, truly fraud. However, the vast majority of check fraud detection engines do not have the functionality to do anything more than establish a revised static rule for ALL checks (for example, serial number out of range). Fraud is not static, and does not always fit a neat set of rules. Why should fraud detection rules NOT be as dynamic as the fraud it is designed to catch? Fraud rules should reflect as much of the investigator's body of knowledge as is humanly possible. After all, don't we all want to learn from history before it repeats itself?

Make a Comment

* = Required

* Name:

* Email:

Website:

* Comment:

*

Fill in the code shown on the left