The system for verifying
and clearing checks
(I'm American, spelling shift!) has actually caused some notable fraud in the past. My favorite example is a gentleman who had hit upon an incredibly efficient scheme, which all by itself was a great deal of the motivation for CheckPoint
and other electronic check-verification
systems (note that check verification is not the same thing as check clearing).
In any case, a bank noticed that a customer was requesting a close on an account that had only been open for three months but had accumulated nearly $10,000 in it, in nearly regular infusions. This alone would not be cause for any interest, since the pattern was roughly that of a salary, had it not been for the fact that the bank had just done an internal audit and found that it was short almost exactly the same amount. Normally, this wouldn't cause notice either; especially at a large bank where a $10,000 difference is penny ante compared to the amounts moved through it. However, this was a smaller bank, and the manager who noticed this checked up on things. He found that each of the deposits had been for a check from a different payor, despite the regular schedule. Tracing back the checks deposited, he found that none had been returned.
Herein lies the beauty of the scheme. Up until this point, the bank verification process was fail-open; that is, unless the processing bank received notification of a problem with the check, they would automatically advance the monies to the depositor, and expect to recoup them when their accounts with the originating bank cleared. This could take days, and could also involve a whole separate department. Thus, there was a window between the end of the three-day clearing period and the time of the actual account clearing where the money was available to the depositor without being yet retrieved from the issuing bank.
What the guy was doing, it came out, was destroying the checks in transit. He'd come up with a chemical solution that, when painted on paper, took nearly exactly two days to activate, and then reduced the paper it was painted on quietly to ash. So the banks would receive his (fraudulent) deposited checks and mail them to their purported bank of issue. In transit, the check would dissolve. The issuing bank would get an envelope with some ash in it but nothing more, and throw it out.
A week to two weeks later, our friend would come along and move the money that was now in his account, or close the account. The processing bank (his target) wouldn't even know they'd been had until the next account clear, and wouldn't know unless they happened to correlate the discrepancy in their books with this particular account. One of the gentleman's main slipups was obsessively keeping his transactions under $10,000 US (this is the size at which a transaction must, by law, be reported to the IRS). Patterns like this, with lots of activity just under that limit, naturally attract attention at the bank and among Law enforcement.
They've fixed the system. Now, usually, the processing bank verifies the paying account is open and has funds in it using an electronic system such as CheckPoint before actually mailing the check and crediting the payee's account three days later.