Ricochet is the best place on the internet to discuss the issues of the day, either through commenting on posts or writing your own for our active and dynamic community in a fully moderated environment. In addition, the Ricochet Audio Network offers over 50 original podcasts with new episodes released every day.
Pictures Worth a Thousand Words – Posting Checks
I’ve written here about the steps in my commercial bank operations career that ultimately led to some major developments in the conversion of the paper-based payments system to electronics. Those steps were (1) the tellers’ line, the proof department, and the bookkeeping department. I managed all three for a period of time before becoming an analyst/programmer and converting the backend processes to automation.
The picture shows a lady operating an NCR Post Tronic bookkeeping machine. This is the equipment we used to post all demand deposit checking account transactions. The tray where her left hand is flipping through ledger cards contains all the accounts assigned to her. Stacked in front of her on the machine are the checks, deposit tickets, credit and debit memos for those accounts. In her right hand, it appears that she has finished posting transactions for that account, and is looking to replace the ledger in its proper location and get the ledger for the next account having transactions in the stack on the machine.
In our case, we had I think somewhere between 24 and 36 machines in total, of which I was the manager of the last half of the alphabet, L-Z, as we processed accounts alphabetically. In the tray, you can see the magnetic stripes on the back of the ledger cards that are used to register the ledger in the machine when posting transactions. At the beginning of the monthly statement cycle for a given account, the blank ledger cards are imprinted with the name and address of the account owner and the first thing done with that blank ledger at the Post Tronic machine is to transfer the balance from the statement period that is ending. That ledger card has two full sets of information about the month’s activity that will be cut in half, one half mailed to the account owner with paid checks enclosed and the other half retained by the bank.
Each of these machines would generally have two people assigned, sometimes three, and the first action by these people at the start of the day would be to sort all the day’s transactions into alphabetical order. When an account ledger was entered into the machine, the magnetic stripes were read and the balance picked up, the transactions entered one by one and, when completed, the ledger was dispensed with all those transactions printed along with the new balance.
Since not every account would have transactions daily, not all ledgers for a given machine were processed daily, so a running balance was kept for the machine and periodically what was called a trial balance was run where every account ledger would be entered into the machine and a grand total checked against the running balance.
The L-Z half that I managed probably had 30 employees handling both personal and business accounts. Lots of additional management issues like handling overdrafts, signature verifications, amount discrepancies, etc.
Once we got all the mainframe computer programs written, tested, and the work converted, all these jobs I’ve been describing went away. But the checks were still there, so there was still a lot to be done.
These machines were in use from the late fifties until the seventies. I had stints managing people operating these machines in 1961 and 1964.
If you’re not bored enough yet, I might do another one explaining the proof department operation.Published in General
Hey Bob – I can’t see the picture you say you included with this post. I think that’s because its in the png format.
I got it off these site:
http://www.dbmurphyusa.com › ncr-postronic.html
I guess the technology has bypassed me. That doesn’t go there either unless you go from DuckDuckGo.
Well, you got it now. Kinda kills my title if you can’t see the picture.
You capitalist pig.
Although certain aspects of digital banking are beneficial, there is a down side.
Over the last 4 or 5 years, the electronic processing, digital wiring that now replaces a paper check with a simple digital conversion of the check’s banking numbers and its amount issued to the Payee have made it impossible to have a record of what is going on.
I need the physical copy of a check that I sent to a clothing firm, to show that company held it so long that I then incurred a late fee. This was written on the bank account which doesn’t have a copy paper.
Bank is declaring the check made on my account with them is not possible to recover. They make this claim based on it being digitally converted. Not sure where to go from here.
It seems to me that there is still a requirement that the bank retain paid check images for some time for certain legal processes involving forgeries and tax audit purposes and maybe others. Exactly what this requires between the bank and its depositors I’m not sure.
Yeah, I am thinking that the day I went into the bank to recover the check, or ask for it to be recovered, the bank mgr was snowed under with paperwork. I think if I make an appointment to see her when she is not busy, there could be a different outcome.
I was always under the impression that a signed and endorsed check with the memo filled in was a contract with court admissibility.
I went to my bank for a copy of a check and was told I couldn’t get one. Then I got a copy of the front of the check. Then I had to ask for the back endorsement side. Then I got a copy of the endorsement but the bank account number to which it was deposited was inked out. Then I had to get a copy of the front and back on one page but with the receiving account number and signature visible. Then I needed to get the xerox copy signed by the bank manager certifying it as authentic.
Digital copies of checks do little to nothing even if you can get them. You can’t do a forensic analysis on a xerox.
AND the manager was surly and huffing and puffing to whole time. I cancelled my account there, but still I never see copies of my own cancelled checks with any other bank, either.
So that’s how that worked! Thanks for the explanation. Yes, do more if you have ’em.
This post reminds me of how quickly technology advances. While the NCR Post Tronic doesn’t seem to use punch cards, those were once considered cutting-edge as well. So much so that in 1964, the Phoenix Financial Center was built to look futuristic, but it already looked like an anachronism 15 years later when I became a teen.
That’s right — it’s a gigantic punch card.
When I look at my check debits online a click brings up an image of my check, front and back.
Yes please do tell about the operation of the proof department.
Postronic machines became popular on both sides of the Atlantic after standards for magnetic characters were adopted by the American Banking Association. This standard emerged from work by Bank of America and the Standard Research Institute earlier in the decade resulting in the ERMA. Wiki page for both is informative
I believe there was a competing standard develop by Chase.
In any case, the Postronic was important for NCR to open more sales in the banking sector, where they had been selling accounting equipment for some time.
Just a few years ago I bought a little handmade something in an ancient store in ancient Kyoto. The sales lady was equipped with an abacus.
What happens when the ledger card is full? Does the operator just insert a new one and the total is carried over via the magnetic stripe?
Also, does the stripe only include the running page total, or does it also include detail amount for each line?
To go to a new page from a full one it was basically the same as the month-end process where the full ledger is removed and the new ledger inserted and the data transfers.
I’m a little sketchy on the technical details here on both the printing and the details of what was actively stored in the magnetic stripe.
The operators knew more about this than I did. And I actually had three supervisors, who grew up on these machines, between me and the operators so I’m not capable on this level of detail.
Thanks. Also, I wrote about ERMA several years ago: Robot Emeritus.
Bob, this takes me back. I got my start in commercial programming in 1979 writing POD software on IBM midrange systems. After a couple of years of that I shifted to teller line automation on early networked hardware. I spent a lot of time in banks.
I pretty quickly moved from there to the technical/networking side, microcomputers, Tandems, and eventually the embedded machine control and robotics stuff I do now. But I remember well, even fondly, the old check sorters, RPG-II on the System34 and 38, and the arrival of the AS400 (by which time I was moving on to micros).
I miss a lot about the early days of the computer revolution.
I miss a lot about the later days, ha ha! Most of my work was in a batch environment. Even when putting government payments on direct deposit we still shipped out tapes.
I worked in a bank after school and summers in high school and college (late 80s and early 90s), and it was a great financial education. A couple summers I ran a proof machine so that the regular operators could take vacation. The bank had just gotten a network, so the proof machine dumped all the transactions to the network at the end of day. I enjoyed the challenge of when grocery stores brought a huge stack of checks in just before closing time and they had to be run as fast as and as accurately as possible so that the proof machine could close and then the bank could close before the courier came to get the checks. I was never as fast as the regular ladies, but I got pretty good.
They also used me to “guard” the tellers when they took cash to the ATM. The ladies carried a bag with tens or hundreds of thousands in cash to refill the ATMs, and they asked me to go with them to protect them. I don’t know what I could have done because we were clearly instructed to give the cash up immediately if robbed, but I suppose they felt better having a guy walk beside them.
Regarding batch processing, I was right at the end of it, or at least in the transition to the end of it, for a lot of transaction processing. Most of the stuff I wrote was part of the nightly batch processing run, and it wasn’t until I started doing teller line work using cobbled together networks of NCR PCs and check printers that it began to feel like real-time.
Talking to the group CIO of a global bank recently, he told me they still ran D+1 balances (ie as of last night).