I am using the OCC Exam Procedures as an audit program and select a sample of "high risk" accounts (usually high cash volume such as convenience store/check cashers)and get the most recent 3 months statements. I then compare the statements to the large cash transaction reports and investigate differences.

If the bank uses batch processing for cash-outs, I am finding that these items are not aggregating by tax ID on the large transaction report (we've observed this on at least 2 well known core processing systems). In several recent audits where banks (or maybe just a smaller branch) batched many of their cash items, I observed that this has resulted in significant underreporting on a regular basis.

One bank we worked with adopted the procedure where no cash out items of $1,000 or more would be included in a batch. (I thought this was a reasonable solution.) I was wondering how others were tackling this problem.

Also, if a bank identifies this weakness in their system, knows that CTRs may have been underreported for years, and chooses just to improve the process for more accurate reporting without refiling/amending, what could the implications be in the next BSA exam? Would the bank likely be subject to harsh criticism &/or fines for not amending past reports?

Thanks!

[This message has been edited by SLC (edited 04-13-2001).]

SLC; I don't think you would be subject to fines or backfiling. I don't have the BSA regs right now, but as I recall, the requirement is that the bank aggregate if it has knowledge of multiple transactions exceeding $10,000, but there is no requirement that you must have a computer program to do that. That being said, for all practical purposes banks have developed aggregating systems. With all systems, there are glitches. If you have identified a problem as you describe you should make reasonable efforts to correct the problem. As to whether you would be subject to backfiling, I rather doubt it because at the time the bank did not have knowledge of those transactions. Kudos on a good catch.

As to how we handle the situation - our aggregate report will not catch batched checks either because there is no imput of TIN. Cash outs appear on the report for each account identified by TIN and due diligence is conducted to determine if any of those checks cashed would total more than $10,000 and if so - who were the payees. If a single payee is identified as benefitting from a cash transaction exceeding $10,000 a CTR is filed.

This is strictly my opinion offered in good faith.

Thanks for your opinion A.D. It seems that most automated systems handle batches this way. With the new BSA exam approach, I would expect banks to have these types of findings where they did not before.

We had to take this a step further after our last exam. Our system aggregates by TIN, however, it doesn't catch the customers who have business accounts -same owner - (ie sole proprietorships) but different TINs. We had to go to a manual system to do that. I know there are programs that will do aggregation both ways, but our data center is not moving in this direction. Does anyone have any ideas on this one?

Aggregation of cash out transactions has always been troublesome. In most cases, there's a good argument that nothing should be aggregated or reported--unless I'm missing the meaning of "batch processing for cash-outs". What types of cash out transactions are being conducted and by whom?

Richard, an example would be:

Customer A cashes a check for $20,000. There is a debit to his account for $20,000 and a corresponding cash credit for this transaction. This would aggregate under the customer's TIN on a large cash report.

In a batch situation, this check would be included in a "batch" with, say 20 other checks. Each of the 21 checks would be individual debits with one corresponding credit to cash for the total.

Sorry, I'm not much of an accounting type, but hopefully this better describes the situation.

I'm not making the connection between aggregation and a single cash out transaction that triggers a CTR. Aggregation won't be necessary to determine your responsibility to file. If currency was involved, it was counted out by an employee who has been trained to file CTRs for single cash in/out transactions exceeding $10,000.

Is the real issue how to handle multiple small cash outs (each less than $10,000) posted during one business day? If so, you're struggling with a perennial problem--how you know that some or all of the small cash outs were conducted by or on behalf of a single person. Payroll accounts are the best example of this problem.

I wouldn't (and didn't) aggregate and report multiple cash outs unless I could prove there was a connection among the payees.

[This message has been edited by Richard Insley (edited 04-17-2001).]

[This message has been edited by Richard Insley (edited 04-17-2001).]

Some banks continue to maintain logs of cash paid checks within a certain dollar range so that this information can be reviewed at a central location to determine if a CTR should be filed. This may work for a small or medium sized community bank, but would be out of the question for a large bank.

Our report captures deposits by TIN. In addition, cash paid checks also appear on the report. If we see a cash paid check for more than $10,000, we look it up on the optical disk to see who the payee is. We then verify that a CTR has been filed by the branch that conducted the transaction. We do not investigate checks under $10,000. This would be a full time job in our bank. Instead, we do regular training to remind our employees that they are responsible for filing CTRs when they HAVE KNOWLEDGE that checks have been cashed on behalf of or by an individual on the same business day.

Dolly Nugent
VP/Compliance Officer
Citizens Business Bank

When using a batch system for cashing checks, don't you tran code the items cashed for posting to the account before items coming in through clearing?

OR are you talking about cashing items not on us?

Dolly- The bank I worked for had 250 branches, but we handled cash-outs in virtually the same way you described. The only difference was that our automated CTR data capture system (not our employees) read the tran information off of GL entries. We, too, had trained our employees and entrusted them to monitor (to the extent possible) multiple cash payments to a single payee. In cases where they knew that one person had cashed multiple checks totaling more than $10,000, they keyed the necessary info into our system to trigger the CTR.

Richard, I think that you and I agree that training is an important element to a successful BSA program!

Dolly