The dole que is this way...
Citibank accidentally wired $500m back to lenders in user-interface super-gaffe – and judge says it can't be undone
A judge has ruled that Citibank can't claw back more than $500m (£360m) it mistakenly paid out after outsourced staff and a senior manager made a nearly billion-dollar (£700m) user-interface blunder. The error occurred on August 11 last year, when Citibank was supposed to wire $7.8m (£5.6m) in interest payments to lenders who …
COMMENTS
-
-
-
Friday 19th February 2021 09:38 GMT Scott Broukell
I knew somebody who worked for a major bank in the 70's and inadvertently transferred c.£3M into one of her own accounts. It took place rather late in the day on a Friday, she was frazzled from a stressful work week and had plenty going on in her personal life as well. She said that she awoke, with alarming suddenness, sweating profusely, in the middle of a somewhat troubled Friday nights slumber and immediately alerted her manager to the error. Happily she retained her position with the bank, because she was good at her job.
-
Friday 19th February 2021 16:51 GMT Dave314159ggggdffsdds
I once managed to walk off-site with what were essentially a set of keys to a major bullion vault. It wasn't noticed til the middle of the night. I was woken by police cars outside with flashing lights. I wasn't under suspicion, they were worried about my safety if nefarious types had found out.
-
-
-
-
Friday 19th February 2021 18:15 GMT Anonymous Coward
Re: UI Screw Up
The PHB probably told them not to worry, the next "update" would fix the problem.
I remember seeing a similar error years ago in some clinical software that calculated the energy used as someone walked, it needed to know the subjects leg length and expected that the PT would enter the leg lengths in metres - that worked fine until an engineer sat down to enter the numbers - he knew that the gait measurements were all done in millimetres so he entered the leg length in millimetres. The software never checked the entries.
-
Friday 19th February 2021 18:31 GMT Dave559
Re: UI Screw Up
See also: Covid: Man offered vaccine after error lists him as 6.2cm tall
-
-
-
Wednesday 15th September 2021 21:47 GMT nobody1111
Re: UI Screw Up
Except...it didn't go to the wrong accounts and the amounts were exact to the penny on what was owed.
Yes, the defendants are being somewhat weasels but legally so. Citibank borrowed money and (inadvertently) paid it back early (which in the US is usually allowed.) It is Citibank's claim of "oh, I didn't mean to pay you, now loan it back to me" that the judge denied.
Go and pay off your car/house loan then ask for your money back. Your lender may be a bit hesitant too.
-
-
Friday 19th February 2021 10:07 GMT Chairo
"six-eyes" policy
A fine example that more eyes don't necessarily mean better quality. Instead people tend to click through all those nasty work-items without really giving them a glance. We had this in production as well. If you put several check stations down the line, you just end up with the line workers getting sloppy, as they know someone downstream is still checking it, while the check guys get sloppy, because there is still a station behind them and the last one in the line sleeps away, because - hey, there have been already so many check before, right?
That said, that amount of money should have rang some bells. I really wonder how someone can get casual about transfers of hundred of millions of dollars. I am definitely in the wrong line of work, it seems.
Edit: And no, I don't think 3 persons checking such a transfer is excessive. Just that adding even more people probably wouldn't help.
-
Friday 19th February 2021 10:41 GMT Azamino
Re: "six-eyes" policy
Wholly correct about staff growing sloppy Chairo. I worked for a bank which went from two approvals in Change Management to five and the quality dropped like a stone. I once accidentally submitted my template rather than the actual change and it went thru!
The old system was not perfect though. They had the most awful habit of sitting on change requests until they had expired, so no approval but no rejection either. This helped them hit some ill-defined performance SLA that made no sense to anyone.
-
Friday 19th February 2021 11:10 GMT Anonymous Coward
Re: "six-eyes" policy
I approve financial transactions, and I can tell you, I have neither the time nor the will to check the source documentation for each and every transaction to make sure each one has been entered correctly, I just assume it has been.
I'm checking to make sure that the person who put the transactions in, isn't slipping in one for themselves or their mate. I.e. I'm not checking for mistakes, I'm checking for deliberate theft.
-
-
Monday 22nd February 2021 08:49 GMT Anonymous Coward
Re: "six-eyes" policy
> "Out of curiosity, how does this work? You either approve or reject but not edit the transaction?"
Correct. I don't have authority to enter the transactions, only approve the list.
The finance committee approves a pre-prepared list of transactions, which is minuted. I have the dubious pleasure of actually logging into the online banking system and approving the transactions.
If I find something fishy, I report that back to the financial committee
-
-
Friday 19th February 2021 23:54 GMT jdiebdhidbsusbvwbsidnsoskebid
Re: "six-eyes" policy
Where I work, someone in IT support has a very similar surname to mine. I often get emails about support tickets that should go to them instead. I know them quite well now and it's an in joke between us. Sometimes I get purchase orders that should have gone to them, and I've approved them rather than go to the effort of getting someone in purchasing to understand that something has gone wrong and they need to redirect the purchase order. It's all small fry, a laptop battery here, some new software there, but all evidence of the culture that if the "system" says so, who can argue?
-
-
-
Friday 19th February 2021 13:28 GMT Cuddles
Re: "six-eyes" policy
"That said, that amount of money should have rang some bells."
There was a fuller explanation on Ars Technica. The problem is that the entire process is a nonsensical mess. Instead of having a system that can just transfer a specific amount of money just for the interest due, they instead have to transfer the entire value of the loans. So to make the actual payments they direct some of the money to creditors, and then point the remainder at an internal account. That's what's shown in the screenshot - the main transfer is already set up, and these are settings to override some of it and divert it to the account number shown. But it turns out that in order to do so you have to override three separate fields instead of just one, which is what they failed to do here.
The problem is that there's no obvious indication that anything has gone wrong. $900 million was supposed to be transferred, and $900 million was transferred. It was only later when they tried to take the bulk of it back out of the temporary account that they noticed it hadn't gone to the right place. There weren't any alarm bells because everything was correct right up until some of the money ended up in the wrong place, which isn't something approval before the transfer would catch.
Obviously the terrible interface deserves some criticism, but the real problem seems to be having a system that requires such a convoluted and inevitably error-prone process in the first place. The judge says this problem has never happened before, but I'd be willing to bet it has happened many times, just not with such large sums or without being able to fix it.
-
Friday 19th February 2021 15:26 GMT wub
Re: "six-eyes" policy - get the PDF!
Anybody who's read this far should really get the PDF and read that. You can skip the financial stuff, but that's fascinating too - at least one bank thought CITI was deliberately buying the loan as part of an aggressive refinancing scheme that had triggered a lawsuit that was almost ready for filing when suddenly the loan got paid off!
This is going to be a case study in business school, whatever comp sci morphs into, and law school. The facts of this particular situation meet a specific New York legal exemption to the rule that misdirected wire transfers have to be returned to sender, based on an almost perfect fit to a 1991 precedent. The court actually sounds reluctant to reach the "you get to keep it" verdict.
The Russians really don't need to do anything with the information they are collecting through their SolarWinds project. They can just sit back, grab a brewski or two, and watch as we spontaneously disassemble ourselves.
-
Friday 19th February 2021 15:46 GMT MacroRodent
Re: "six-eyes" policy - get the PDF!
> grab a brewski or two
Surely you meant a vodka or two. - Many, many years ago I was in a shop in a then-socialist Eastern European country, in front of the booze shelf, and noticed the cheaper Russian vodka bottles did not have screw caps, but caps of the old kind that you used to find in lemonade bottles, where you rip a tab to open it. You were clearly meant to consume the bottle in one sitting...
-
Friday 19th February 2021 21:30 GMT Anonymous Coward
Re: "six-eyes" policy - get the PDF!
Beer is drunk a lot in Russia, with kiosks selling it by the bottle all day long. It's a common sight on the Moscow metro to see people drinking a beer on their way to work, and in the Summer a beer at lunchtime from a kiosk is normal. For ordinary Russians, vodka is mostly drunk with friends, and served with pickles to nibble on.
The old foil caps on vodka bottles died out in the 1990s, a leftover from Soviet times. They also resued bottles, so it was common to see the same brand on a shop shelf but in different shaped bottles. I still have a green bottle that would have originally held mineral water, but was then reused as a Stolichnaya vodka bottle.
-
-
Friday 19th February 2021 18:38 GMT Claptrap314
Re: "six-eyes" policy
That process is not quite as insane as it sounds, given history. This is clearly an implementation of double-entry accounting. Well, an implementation. Done quite poorly.
"You sent a code monkey to do the work of a software engineer."
Someone said, "People adding all these numbers make mistakes. Bring in a programmer to fix that." And so they did--the "programmer" took their manual double-entry process exactly, created a UI that matched the accounting book, dropped the data into a database, and voila! No more accounting errors! And look how fast it reconciles the books!
Of course proper software engineering, especially in the fifties or sixties, would have said things like "When we set up a loan repayment system, we'll prepopulate all the fields with the correct data. Any changes will be flagged alert checkers." And, "The purpose of double-entry accounting is to prevent human error. Since we're using systems whose chance of error in computation is minuscule compared to the chance in human error in data entry, we should design this system to do the double entry itself. The operator should never get the chance to enter a value which can be computed."
-
Monday 22nd February 2021 14:14 GMT PeterM42
Re: "six-eyes" policy
Ah! - UI - the downfall of many a piece of what could have been decent software.
Having spent many years on application support, I have seen CRAP design far too often.
When ready for testing - give it to someone who has NO IDEA what the software is supposed to do.
Then take it back to development!!! (usually)
-
-
-
Saturday 20th February 2021 01:14 GMT J.G.Harston
Re: "six-eyes" policy
"it's just figures on a screen."
From the screenshot there doesn't even seem to be any figures on the screen. If it at least had a status bar saying something like:
TOTAL FUNDS TRANSFER: $1,122,763,920.00
there at least would be an indication of what was about to do down.
-
Friday 19th February 2021 19:42 GMT shah27
Re: "six-eyes" policy
You'd think large values like this should ring a bell but it doesn't.
The whole reason why there was sign off was to avoid the whole bell going off. This is because there are transfers of millions happening on a single transaction.
What should have happened is the system should have been set up in advance where only specific accounts are allowed to get certain payments. But that requires more operational, system development and training costs. Why would you spend millions to create a system that does the job when the current one is cheaper and does the job. The human user should pick up on mistakes... right?
-
Friday 19th February 2021 20:10 GMT Rol
Re: "six-eyes" policy
Ball park reality check!
As was always advised by my maths teacher - do a very rough calculation using numbers rounded off for ease and see if the answer is in the same range.
In this case, one of the managers who signed off should have had a spreadsheet on hand listing all the anticipated payments for the week, and then just checked the figures were close.
A col for the principal, one for the rate, another for the date, and the expected payment in col D.
Yes this would be a bit rough, but would have instantly flagged the error.
I'd say the guy who did the error be given a good finger wagging, and the rest be given the finger pointing to the exit, as they were clearly defrauding the bank of their wages.
-
Sunday 21st February 2021 05:24 GMT Anonymous Coward
Re: "six-eyes" policy
Also sounds like grounds for them to be investigated for possible discrimination against people that could be blind or just have one eye.
To reinforce your point about people expecting others to do the proper checks. Sounds like Ford and the PowerShift transmission issue. Some of the material showed that the engineers were under pressure to send it up the chain. So rather than push back since they knew there were issues with the transmission, the expectation they had was that further down the QA chain it will be rejected. It wasn't. I wonder how many different groups that "approved" it had the thinking, the previous group wouldn't have approved it if there were issues.
-
-
Friday 19th February 2021 10:34 GMT fidget
Double keying already used in some banking applications
Some years ago major publishers digitised journal back-issues for HTML publication. They did not use OCR. They used double keying. I found this definition of double keying on the web (emphasis added).
The process used by operators when they enter the information twice or when two separate operators enter the data at separate times. The two entries are then compared with each other to ensure that they match. This process is used in military and BANKING APPLICATIONS to detect intended falsification of information.
https://www.yourdictionary.com/double-entry-or-double-keying
See also https://journals.openedition.org/jtei/739, from a Text Encoding Conference.
-
Friday 19th February 2021 11:21 GMT Roger Greenwood
Re: Double keying already used in some banking applications
Anyone studying data processing would have learnt about double entry/double keying decades ago. I am surprised it isn't used more for high value systems/transactions, after all the comparisons will be automatic. We do it for our wages system as folks get very fed up if they aren't paid ( < 50 people)....
-
-
-
Friday 19th February 2021 22:33 GMT John Brown (no body)
Re: Double keying already used in some banking applications
"Citi bank do seem to care about their $500m, however."
And yet, it;s not, and never was, their money. They borrowed it to lend to someone else. Then they "accidentally" paid off the lenders early and now they are appealing a court decision so they can force those lenders to lend it back to them again. Weird.
The only thing that's slightly odd is that the lenders loaned the money to Citi on the understanding it was at a certain interest rate. By getting it back early, they lose out on the future interest payments. I can only assume they now think, having got it back from Citi, they can use it more profitably elsewhere.
-
Monday 22nd February 2021 13:41 GMT Cynic_999
Re: Double keying already used in some banking applications
"The only thing that's slightly odd is that the lenders loaned the money to Citi on the understanding it was at a certain interest rate."
Yes, but then Citibank managed to do a "debt restructuring" and they were only going to pay back 60c in the $. So at that point the lenders were looking at a big loss, and getting it all back would have been very welcome indeed!
-
Monday 22nd February 2021 18:35 GMT xehpuk
Re: Double keying already used in some banking applications
It's even more complicated. The bank was just the agent. A cosmetic firm had borrowed the money. The lenders had already sued to get their money back in advance since the cosmetic firm is in trouble. So when the money dropped in they argue it's totally believable that lender had caved in before court hearings. So now the bank might end up having to use their own money to cover it. So suddenly it is the banks money at stake.
https://www.google.com/amp/s/www.bloomberg.com/amp/news/articles/2020-08-15/citi-s-900-million-blunder-raises-stakes-in-revlon-showdown
-
-
Monday 22nd February 2021 20:07 GMT Michael Wojcik
Re: Double keying already used in some banking applications
Clearly blaming Wipro is a popular trend in the comments, and I'm not interested in defending them as such; but it seems to me the problem here is clearly in the application which fails closed -- that is, defaults to sending more funds outside the organization. I wouldn't be surprised to see this error from any data-entry staff, working anywhere in the world, for any employer.
And while double-entry would probably make it less likely, studies have shown that independent redundant data entry can result in the same errors. And this is an application which is just begging for this particular error.
It's like that old bomb-defusing joke: "Cut the red wire. But first, cut the blue wire.". If you design a system where extra vigilance is required to obtain the correct result in the normal case, you've designed it to fail.
-
-
-
Monday 22nd February 2021 15:55 GMT Adelio
Re: Double keying already used in some banking applications
Many moons ago, circa 1983/84 on an ICL ME29 system we used to enter timecards using DDS, two people enter the same stuff and it cross checks for any differences. By god, those women were quick. and it is a simple way of performing some basic validation.
-
-
-
Friday 19th February 2021 11:12 GMT ARGO
There's some background missing here. The loans were in the process of being refinanced; some lenders had agreed to the new terms, others had not. The ones who kept the unintentional repayment are all in the latter group. The loans were trading at ~40 cents in the dollar, so full repayment is a bit of a result.
-
Friday 19th February 2021 12:06 GMT Dave Pickles
The other pertinent point is that the loan was subject to New York state law. In that jurisdiction there is a "discharge for value defense"; a lender who receives full payment for a debt is under no obligation to refund the payment if it was made in error.
The judge's decision also explains why the 'six eyes' barrier failed. Everyone involved had the same incorrect understanding of how the software worked.
-
Friday 19th February 2021 18:11 GMT vogon00
Failure in understanding
> "Everyone involved had the same incorrect understanding of how the software worked."
Very true.
However, I can't help having the slightly different opinion that "Everyone involved had different and incorrect understandings of what the end result should be"
This isn't a snipe at Dave or any of the individuals concerned, just more of an observation that unless you understand the underlying reasons for the boxes you tick in the UI (No matter how crappily presented!), something like this is bound to happen.
Also, it appears the oversight (Human and automatic - there *was* some sanity check in the code, right?) failed here too.
-
Friday 19th February 2021 20:33 GMT A Non e-mouse
Re: Failure in understanding
"Everyone involved had different and incorrect understandings of what the end result should be"
No. All three CitiBank employees had the same understanding: The transaction was only to make an interest payment hence the special account code being entered into the "Internal GL" field and the "Override default settlement instruction" box being ticked.
The problem is that to achieve the desired outcome not only did the "Principal" field need overriding, but also the "Front" and "Fund" entries.
Also, it appears the oversight (Human and automatic - there *was* some sanity check in the code, right?) failed here too.
The computer was asked to make a money transfer and it did what it was instructed to do: Make a money transfer. The problem is that the UI was so lame that the computer was asked to make the wrong transaction by the meatbags. No computer fault: All human.
-
-
-
-
-
Friday 19th February 2021 11:42 GMT jmch
Re: Isn't it the banks themselves....
Yep, that's right. If there is a mistake, they can make an equal, opposite transfer (reverse) to rectify but AFAIK the transfer itself cannot be undone. If my bank mistakenly transfers a few million to my account, I believe they can even just make the reverse transfer themselves without my knowledge / verification (might vary according to jurisdiction).
If it's from a different or foreign bank, they probably have to ask nicely for the transfer to be reversed, as happened in this case. If refused, as happened in this case, they can go to court, and would most probably win if it was a clear case of money transferred for no apparent reason other than error. If a bank mistakenly transfers any amount to someone, that person's best bet is to politely say "Sure, I will return the money, here's my fee for the administrative effort".
In the case of Western Union etc, as you say, not knowing the recipient means no way to get the money back.
-
Friday 19th February 2021 16:20 GMT Dave314159ggggdffsdds
Re: Isn't it the banks themselves....
In addition, consumer bank accounts have different terms to business accounts, let alone money managers.
(If you - usually a large business - have lots of millions in cash, you don't put it in anything approaching a bank account. You have people who e.g. place it overnight wherever it gets the best (safe) return. Then you lend it out to different places during the day.)
Retail banking customers have a whole bunch of terms and conditions - and statutory protections - attached to their accounts that you don't get with really large amounts of money.
-
Friday 19th February 2021 17:53 GMT Henry Wertz 1
Re: Isn't it the banks themselves....
Yup, and if this was just a deposit, they could probably do this to. The sticky wicket here is that it's a loan repayment, and load repayments have rent rules (including, in New York where Citibank is based, one specifically saying loan repayments cannot be reversed.)
-
-
Friday 19th February 2021 11:31 GMT jmch
Interesting legal argument there....
So if I'm reading the article correctly, one of the judge's arguments was that if I don't expect a big bank (that has never before made an error in my favour) to make an error in my favour, then I am entitled to expect that it was't an error?
I understand there are other factors, particularly that it was, to the cent, money that was due to them anyway, just repaid earlier. But I would have thought the latter argument is enough without the former, since the former argument can set a dangerous legal precedent.
-
Friday 19th February 2021 11:48 GMT RPF
Re: Interesting legal argument there....
I think it's slightly more nuanced than that. He's saying that because it was such a huge amount, it would be reasonable to expect the bank to have thoroughly checked it. Let's face it, even with about 6 zeroes knocked off this sum, I would check it carefully!
-
This post has been deleted by its author
-
Friday 19th February 2021 14:02 GMT Stephen Beynon
Re: Interesting legal argument there....
There was also that the payment amount repaid the outstanding loan and interest to the penny, it was not a random payment amount where, or a decimal point shift so it was reasonable for the receiving party to believe that it was an intentional repayment of the loan rather than fat fingers on the amount transferred.
-
Friday 19th February 2021 19:48 GMT Eclectic Man
Re: Interesting legal argument there....
I have just paid three years worth of service charge on my flat (apartment for USAfolk). The invoice included the bank sort code and account number for the transfer, but I still checked them, sent an email to the accountant to say the money should be in the specified account (numbers and all) the next day (I had delayed the payment even though it was exactly the same account as 3 years ago). And all for a mere £2600.
-
Saturday 20th February 2021 01:20 GMT J.G.Harston
Re: Interesting legal argument there....
Ditto, and I even do that when paying a fiver to my window cleaner. I'm slightly paranoid about paying money into the wrong bank account after 30 years ago finding my bank had inadvertantly given my bank account number to another customer by giving them the wrong sort code, so we were both receiving and spending money from the same account.
-
Saturday 20th February 2021 22:10 GMT tfewster
Re: Interesting legal argument there....
When I bought my last car, instead of waiting for a cheque to clear or getting a bankers draft or even using a card terminal as previously, I was asked to transfer the money to their account.
So I started with a £1 transaction, and insisted they confirmed receipt and could identify the payer before transferring a larger amount.
The car salesman seemed bemused. Why should he have to make it safe and easy for me to give him my money? He almost lost the sale right there.
-
-
-
Friday 19th February 2021 12:59 GMT JacobZ
Size doesn't matter Re: Interesting legal argument there....
The combination of the large size of the error *and* the fact that it was the exact amount due is critical here. Neither part *by itself* sets a precedent, dangerous or otherwise. Indeed, the judge relied strongly on *existing* precedent.
The relevant New York law says that a transaction made in error is not reversible iff
(a) It is (or appears to be) payment in full for a debt that is owed
and
(b) The receiving bank has done "due diligence" to verify that the payment was correctly made
So the size of the loan is not the important thing by itself, but rather its supports the receiving banks' belief that it couldn't possibly be a mistake, i.e. it is one part of satisfying (b). Several of the banks also checked with each other to see if they had also been paid and when the answer came back yes, said to each other "Welp, I guess Revlon refinanced with somebody else". And the judge said that was a reasonable inference under *all* the circumstances, size included.
-
Friday 19th February 2021 17:37 GMT a_yank_lurker
Re: Interesting legal argument there....
The payees were owed the money from the bank's client. So X instead of Y showing up in the payee's account were X is the full balance would not necessarily be questioned by the payee's staff at least immediately. So the error was not the payee got money not due to them (the error most are talking about) but one were loan was paid off versus an interest payment. The nuance of the ruling is the payee was owed the money but the bank screwed up and paid off the loan instead of the interest payment. The other situation is people are talking about is the payee is not owed the money though they got it in error. In this case, the bank usually can get the money back if they ask for it in a timely manner (see the local law on this point) as it is not really the payee's money.
-
Friday 19th February 2021 18:17 GMT Henry Wertz 1
Re: Interesting legal argument there....
It's not that dangerous a precedent -- I think the judge is just making the point that it's not like they got $100 trillion or something in their account where it's obviously a mistake, at which point it would really not be reasonable to try to keep it. It's a large enough amount that it's reasonable to assume the bank would verify it (which apparently they did, the verification simply failed), but is the amount of their loan so also perfectly reasonable to assume it was an intentional payment based on the amount.
I recall Bank of America got sued a few years ago (well, their awful so they've been sued for many things but here's one...), they were cashing checks straight away, then sitting on the money (so they could get free interest) until 1 day before the legal limit. Their computers were down over a weekend, so they deposited some checks in like 9 days when the limit is 7. They got sued, they were like "Well, we can't help computers going down", the judge rightfully pointed out "Well, that's why you don't wait until the legal maximum to hold these checks", and the bank reluctantly paid $1000s in damages to the account holder for violating the law.
-
-
Friday 19th February 2021 11:31 GMT Pascal Monett
"all believing the settings were correct"
And nobody checked. A billion dollar transfer and nobody checked.
Three different "managers", and they all blindly clicked OK without checking themselves, or checking that the previous guy had did the job properly.
Rubber-stamp management is easy.
Citibank certainly does not deserve to get the money back.
-
Friday 19th February 2021 15:36 GMT wub
Re: "all believing the settings were correct"
Part of the problem was training, but the design and implementation of the software truly sucks.
They input the correct amount owing on the loan, because the software design made them PRETEND they WERE paying it off. Then they were supposed to say "the principal stays in a "wash" account inside the bank" while the interest, which they did have to pay, goes to the creditors.
Problem is, telling the software that was convoluted, and the outsourced employees and the manager did not really understand how to do that. They all believed that they had correctly asked the system to do what they wanted. Instead, it did what they told it to do.
And - it DID try to want them about the possibility of a mistake, with a warning that said funds we leaving the bank. Sadly, whoever wrote the warning did not mention HOW MUCH money was leaving, and the whole point of a wire transfer to get the money out of the bank.
Breathtaking. Read the PDF. It's just a clusterfuck of epic proportions. The judge actually sounds disappointed that the 1991 precedent ties his hands, requiring him to let the recipients keep the money.
-
Friday 19th February 2021 11:42 GMT Archivist
Ha ha ha
I have dyscalculia and through life have found ways around most problems this creates. One I have not, is copying a bank account number. I'm seriously anxious when I need to do this to pay someone because I know that if I make a mistake, it is very unlikely the bank will agree to claw back my money.
It's great to see the boot on the other foot!
-
Friday 19th February 2021 13:31 GMT heyrick
Re: Ha ha ha
For some reason, I find it easier to copy the bank number starting from the end, not the beginning. Maybe because of one starts at the beginning they're liable to read ahead, but doing to backwards one number at a time, anything following can be completely ignored as "done".
(fellow dyscalculia sufferer, though I thank Acorn User's Yellow Pages (type in listings) for greatly helping me deal with reading and reproducing numbers as a teenager)
-
-
Saturday 20th February 2021 11:58 GMT Anonymous Coward
Re: Ha ha ha
"[...] a holder name to match the account number [...]"
That has now appeared in UK online banking. Apparently on entering a new payee account number you get a status report of a match, near match, or no match. You can then choose whether to proceed with the transfer.
Apparently there have been problems with accounts that do not match the sender's expectation of the appropriate name.
Not sure if my bank has also changed their policy of automatically deleting a payee's account reference after about 12 months of no further payments. Most annoying to have to find and re-authenticate the same number when repeating a payment outside that window.
-
Saturday 20th February 2021 18:08 GMT Ken Moorhouse
Re: on entering a new payee account number you get a status report of a match
The crazy, crazy thing about this is that NS&I are not on that list. You really can't get more legit than National Savings & Investments, and when you are being told that it is my risk if the payment goes "somewhere else", it really spooked me. I suspect that the banks were really miffed that millions of people were transferring funds there because the rates were better. To add further substance to that theory the bank "forgot" to put the account number as reference on the payment, so it got bounced. That happened twice!
It's like the time I went into the bank and was told that I couldn't pay money into an account because of money laundering regs. I had to point out to the muppets that the payee was HM Revenue & Customs.
I think it's time for my meds.
-
-
-
-
Friday 19th February 2021 11:55 GMT 45RPM
Years ago, just after leaving University, I worked for **********. I took calls from clients, passed their requests to the traders with the details that the clients provided, and then sent out cheques or certificates - basic, boring, I need a job, clerical work. Perhaps unsurprisingly, I often enjoyed a lunch of a largely liquid nature - and I don’t mean soup.
One afternoon, a client got in touch with a wish to sell his holding of something or other - I can’t remember the details - but I can remember the value of the cheque. A little shy of a million pounds - and this thirty years ago. I filled in the details, and for cheques of this value, the cheque that I raised was sent to a senior director for signing, before being sent back to me for sending out.
Imagine my surprise when the cheque landed on my desk, ready to be sent out, correctly signed and with an accompanying letter addressed to the payee. Which was me. Somehow, whilst in my cups, I’d filled out the form in my name, and the director hadn’t noticed. Choices, choices. How long would a million last me in Brazil? Honesty prevailed. I went, cap in hand, to the director concerned and told him of my error, handed him the cheque and then raised a new one - with the correct details. I didn’t lose my job (which wasn’t very well paid), but in the following months I did occasionally wonder… What if?
-
Friday 19th February 2021 12:59 GMT juice
I've posted before about the time I opened my payslip to discover an unexpectedly large lump of money sitting there; I'd recently claimed for overtime and hadn't realised that you were meant to claim for "units" rather than per hour.
E.g. instead of claiming a single "8+ hours overtime" unit, I'd claimed for ten of them...
I was good and gave the money back, but given that it had all been approved and signed off, I did sometimes wonder about the same "What if..." question ;)
OTOH, I had it the other way around a few years earlier; I'd filled in the sort code for my bank incorrectly when filling in the forms for my new job. Which meant that come my very first payday, nothing landed in my bank.
Presumably though, someone else had a very pleasant surprise that day!
Thankfully, the bank just fixed it and the money appeared a week or so later, but that "what if" was far more stressful to contemplate!
These days, I think there's more error checking around such things, but I'm still very twitchy whenever it comes to submitting financial details...
-
Friday 19th February 2021 13:34 GMT heyrick
"still very twitchy whenever it comes to submitting financial details..."
A recently had to transfer a large (for me) amount of money from my usual account to the one with the chequebook. So all the details were set up, and my first transfer was for one centime.
When I saw that had arrived, I could then transfer the rest. It helps to check and double check.
-
Friday 19th February 2021 22:55 GMT Giles C
I do the same but am a bit more generous, as I do the initial transaction for £1.
People still think it is strange but it makes sure the money goes to the correct person.
Mind you the banks have started filling in the account name when you put the number in (certainly for business accounts)
-
-
Friday 19th February 2021 22:54 GMT Giles C
Someone did something similar to that to me.
I should have been paid £400 as a single on call payment.
But the payroll processed it as 400 hours. A roughly £20k payslip that month.
When I pointed it out they said to me write the company a cheque for the overpayment. I told them they needed to pull the money back and repay me the correct amount, also sorting out the mess with the inland revenue so I didn’t get screwed for tax.
-
-
-
Friday 19th February 2021 12:33 GMT mark l 2
Hmm I don't agree with the judges decision, a mistake was made and money was transferred to someone who wasn't entitled to it, the recipient was informed it was a mistake but they are refusing to return it, surely that is just theft?
I expect this will get overturned on appeal as it opens up a whole host of problems going forward if it is allowed to stand. What happens if it were the judge making a wire transfer and he accidentally transferred to the wrong person? From his own judgement that person wouldn't be required to return it and I am sure he wouldn't lie down and take that if it were his own money on the line.
-
Friday 19th February 2021 13:01 GMT JacobZ
FCOL
"Random internet person thinks he understands financial law better than highly experienced judge who write 100-page opinion citiing authorities, statute, and precedent".
What is with some people that they think they can just wake up, fall out of bed, and pronounce on complex matters of law?
-
Friday 19th February 2021 12:33 GMT JimC
I suppose the big thing about this one
Is that the recipients were entitled to the money, if not actual right then and there. But when there's serious money you get serious fights. A ballsup at a place I worked saw the entire month's payroll - and it was a very big payroll - paid to a single account. The mistake was spotted as it happened, and the recipient never saw the money appear in their account. But the building society that received the funds refused to reverse the transfer until all is were dotted and Ts crossed. In the meantime my employer had to borrow the money to duplicate the entire payroll and pay it to the correct people, and it was a decidedly non trivial sum in interest.
-
Friday 19th February 2021 22:59 GMT John Brown (no body)
Re: I suppose the big thing about this one
"In the meantime my employer had to borrow the money to duplicate the entire payroll and pay it to the correct people, and it was a decidedly non trivial sum in interest."
That sounds like a bit of a hand to mount situation if the company can only afford the payroll from that months income.
-
Monday 22nd February 2021 04:53 GMT KSM-AZ
Re: I suppose the big thing about this one
"That sounds like a bit of a hand to mount situation if the company can only afford the payroll from that months income."
Depends on your business. If your business is running a coffee shop with 10 folks working there.. . OK. If your business is oh say a workforce provider, where you basically are skimming payroll money, the you basically would need to come up with your entire revenue stream for a month. If you churn out say $1m a month in payroll, you know a small provider, 1000 ish workers. . .
'Kinda like most people, I'm no different, we love to talk on things we don't know about' -- "Ten thousand words"
-
-
Saturday 20th February 2021 12:06 GMT Anonymous Coward
Re: I suppose the big thing about this one
Many years ago my father found a cheque on the pavement outside a bank. It was "pay bearer" for a fairly large amount - presumably to get the cash for a small business payroll.
He took it into the bank and presented it to a teller - who proceeded to count out the cash before my father could explain the situation. Presumably the bank contacted their customer - but my father heard nothing more about it.
-
-
Friday 19th February 2021 12:54 GMT Sykowasp
I bet the bank already has a new UI with explicit actions for the common actions:
[_] Pay Interest (this is probably what you want)
[_] Pay Principal and Interest (loan matures X/Y/20ZZ)
[_] Advanced
Sign-off often doesn't have visibility of the original work order / story, so they will just get the request without context. As long as it looks right, it's approved. A good request includes an authorisation credential limited to the scope of the work order (permission to pay interest only) but that's a lot of work to get in place.
This is why approval requests should include the original work order so the sign off can compare the intent with the action. And the action has to be human readable, not just the underlying codes to perform the action, which are unintelligible to everyone but those in that particular knowledge silo.
Sucks for the bank, but who's crying for them eh? Good for the creditors who clearly have the law of their side in NY and gained 100% back instead of 40% for the refinancing. Bank loses $300m.
-
Friday 19th February 2021 13:20 GMT PerlyKing
Re: I bet the bank already has a new UI
I'll take the other side of that bet. Although they may have scheduled a pre-meeting to start roughing out the agenda for a meeting to decide who to blame for the cock-up.
Less cynically, these things take time. But I'd expect some training sessions in the very near future.
-
Friday 19th February 2021 15:48 GMT wub
Really - read it. The PDF is pretty easy to follow for a court document.
But it is SO much worst than just the failure in training - which also applies here.
The reason they input the exact amount owing is the software design for the electronic fund transfers forced them to do that! WTF?
Then, they're supposed to say, "OK, leave the principle here, in the bank's "wash" account, and send only the interest to the creditors". They believed they had done that, but they only did the first step of three (3) total steps needed to keep the principle at home. So, when they sent the interest, the principle went too. The miracle is they apparently hadn't done it before.
And the software tried to warn them that money was leaving the bank. However, the author declined to show HOW MUCH money was leaving the bank in the warning.
I'd say they're likely to be building a new tricycle about now, one that makes different mistakes. Ooops, I mean one that works properly.
-
Friday 19th February 2021 17:01 GMT DevOpsTimothyC
Re: Really - read it. The PDF is pretty easy to follow for a court document.
"And the software tried to warn them that money was leaving the bank."
That statement is a little disingenuous. It's not really a warning if it's an expected outcome now is it ? If you're making a payment to anyone then you expect some money is going to be leaving the bank.
If the software had tried to warn them that more than##% of the principal was leaving the bank then I'd agree that it was a warning
-
Friday 19th February 2021 21:47 GMT wub
Re: Really - read it. The PDF is pretty easy to follow for a court document.
I agree with you completely. Of course money was leaving the bank - it's the wire transfer application. But this is from the Court's decision:
Raj then proceeded with the final steps to approve the transfers, which prompted a
warning on his computer screen — referred to as a “stop sign” — stating: “Account used is Wire
Account and Funds will be sent out of the bank. Do you want to continue?” Raj Aff. ¶ 21; see
also PX294. But “[t]he ‘stop sign’ did not indicate the amount that would be ‘sent out of the
bank,’ or whether it constituted an amount equal to the intended interest payment, an amount
equal to the outstanding principal on the loan, or a total of both.”
A little extra effort could have indicated how much money was leaving (since the only way to do this transaction was to pretend the loan was getting paid off) and actually made the "warning" useful. The entire procedure seems overly complex and confusing. What would have happened if they had somehow miscalculated the EXACT amount that was currently outstanding on the loan? From the decision, it sounds as if the transfer would have failed. But that difference would have resulted in the possibility of CITI recovering the money. As it stands now, they apparently have zero chance of a successful appeal.
-
-
-
Monday 22nd February 2021 17:48 GMT Neil Spellings
You've clearly never worked in a large bank.
They will still be looking for the original application owner (who probably left 5 years ago) and who took over from them (left three years ago) to get the approvals in place to modify/retire the application (which nobody knows how it works) only that people far aware use it on a daily basis, so has to be kept running 24/7 on that fifteen year old server in the corner.
-
-
Friday 19th February 2021 19:58 GMT Eclectic Man
Ergonomics: had it happened before?
At my first 'proper' job we had a team of ergonomics experts to check things like user interfaces. The screenshot shows just how easy making a mistake would be. I do wonder whether anyone had caught the same error before in the 'six eyes' checking procedure.
I must admit that if I was expecting only 40% of my original investment back and the bank refunded me 100%, I'd keep it. The bank can ask or sue all they like, but in English law they cannot enforce a contract unless both parties can be seen to benefit, I would not benefit from less than 100% back on my investment, so I'd take the money and my custom elsewhere. (You may gather that I am not an over-enthusiastic fan of the banking sector.)
Last century there was an accident when, I think it was Lloyds Bank (could have been another, I guess), accidentally repeated a whole set of SWIFT transactions one day. Hastily telephoning their banking counterparts who were in receipt of twice the amounts they should have got that day, they got almost all of it back, except from a couple of minor banks a long way overseas, which probably wouldn't mind not doing any business with Lloyds again. So the 'old boy network' does have its uses, sometimes.
-
Friday 19th February 2021 21:02 GMT A Non e-mouse
Re: Ergonomics: had it happened before?
Please go back and re-read the article.
The underlying company (Revlon) were, financially, in the shit. CitiBank were "helping" to refinance Revlon's debt. All the debters were aware of this, so a debter having their debt repaid, to the penny, wasn't wholely unexpected.
You're arguement about enforcing a contract is completely moot. The companies already had a contract! It would have said something like: "We loan you $X dollars for Y years. You pay us interest at Z%. You can repay the full amount at any time."
-
-
Saturday 20th February 2021 09:33 GMT AndrueC
Based on the look of that UI (especially the icon on the button) it must date back to the last century. Probably written using one of the Borland tools. Borland C++ probably. Dunno if that's bad because no-one has bothered to invest and keep it up to date or impressive because apparently it's still (sort of - depends how often this kind of thing has happened over the last 20+ years) fit for purpose.
-
Monday 22nd February 2021 10:30 GMT conel
From the pdf; relating to a seperate lawsuit:
"The UMB Complaint accused Revlon of improperly amending the 2016 Loan Agreement to avoid paying the Lenders, and alleged that Revlon had defaulted on the loan. More significantly for present purposes, it alleged that Revlon was unable to repay the loan because it was “insolvent and facing a severe liquidity crisis.”
"The non-returning lenders believed, and were justified in believing, that the payments were intentional," Judge Jesse Furman ruled."
"Indeed, to believe otherwise — to believe that Citibank, one of the most sophisticated financial institutions in the world, had made a mistake that had never happened before, to the tune of nearly $1bn — would have been borderline irrational."
I think the lenders would've known it was a mistake on the basis that the debtor wasn't liquid enough to repay the load. Don't think the judge's reasoning here holds up. I'm sure the lenders were sophisticated enough to realise bad loans don't get repaid in full, ahead of time, intentionally.