RBS IT cockup: This sort of thing can destroy a bank, normally

But in this case the taxpayers just get hit again

Analysis The thing you have to remember about banking is that it's a confidence trick. As with all such things, once the confidence is gone the trick no longer works. That's what should be worrying the executives at NatWest and RBS over the shambles in their computer systems this week.

As to what actually caused the problems, I'm reading El Reg to find out myself [watch this space - Ed]. The general view of Reg readers seems to identity the major issue as an upgrade to systems held together with sealing wax and spit by contractors without a deep and complete understanding of the interactions between those legacy systems. And since all those with such knowledge had been recently laid off on the grounds of improved profits to be made from outsourcing, the bank was out of luck. Naturally, you all followed such observations with a hearty “Har Har, serves you right”. Perhaps they tried to move something into the cloud?

My bit is to try and divine what this means for the future of the bank itself. And that's where the confidence trick comes in. The basic thing about banking is that if you borrow short and lend long you're a bank – if you don't, you're not. By short we mean for a short period of time, by long for a long period of time. To a bank, our deposits in the bank itself (the balance of our current accounts) are loans to the bank. These are the bank's liabilities (some of them at least), while its actual assets are the loans the bank has made out to other people. Yes, some of these are short-term – an overdraft say, or a credit card balance. But many of these are for longer terms: a three-year loan, perhaps, or even better: a 25-year mortgage.

Liquid courage

The actual job we want banks to do is this maturity transformation: we want them to take that short-term money that is lying around the economy and turn it into the sort of long-term loans out of which an ever more glorious civilisation can be built. This is, over and above their provision of a basic payments system, the economic purpose of having banks at all.

But it is precisely this maturity transformation that makes banks fragile, at risk of a loss of confidence. Because the money has been put out there in three-year loans while our deposits are recallable anytime we want our dosh. So if everyone turns up at the same time to demand their cash, then the bank goes bust; not because it is insolvent, but because it is illiquid. It is worth the money that it owes, it just cannot lay hands on it right now.

Yes, agreed, this is very simplistic but it is in essence entirely true: true of The City just as much as Aunt Dora's current account. The vast rivers of cash that run through the wholesale banking system tend to be short term on their way into banks and longer term than that on their way out of them. There is always thus the risk of a bank being illiquid. And all that stops it being so is the confidence that it won't be so. Lose that confidence and you get a bank run.

This is in fact what happened to Northern Rock. No, the queues of people outside the branches were near irrelevant. It was the people in those wholesale markets who led the run. Crock financed their mortgages by borrowing short term (largely overnight) from other banks. Sally and John get their mortgage, buy the house, Crock pays the seller. Crock gets the money by borrowing it wholesale then adds, over time, thousands of such mortgages together until it has enough to make a bond issue. The bond issue is for 10 or 20 years say, about the average life of the pool of mortgages. At which point Crock's book is maturity-balanced: they're safe. They've borrowed money for about the same length of time as they've lent it and they cannot thus have a run.

Except, of course, what happens if the ability to borrow short term dries up while you're building that pool of mortgages to sell as a bond issue? Which is indeed what happened to Crock. At which point they're illiquid, the confidence fairy has disappeared and people are clamouring for their short-term deposits back – which they cannot have because they're now in those 25-year mortgages. Result: Northern Rock goes resoundingly bust.

This can happen to absolutely any bank, anywhere, at any time: once confidence goes, so does the bank. And that is what should be getting the NatWest peeps worried. Not that they've had a computer blowup but that they've right royally pissed off their customers and some fraction of them, as soon as it is actually possible to remove money from the bank, are going to remove all of their money from the bank. A bank run of some size in short.

A bank run, but without the bankruptcy bit...

No, this isn't going to lead to the bankruptcy of RBS (again). The first reason being that it's already pretty much government-owned and the solution to a bank run is government ownership to restore confidence. The second is that it's the wholesale markets which are the real worry. Back in 2007 most of this wholesale money went through the interbank market: it was Lloyds lending today's excess to NatWest and vice versa the next day. These flows are large (some £20 trillion a year back then) and very short term – overnight out to seven days for the majority of them. Crock went over so fast because that money could be – and was – yanked out very fast.

Today, to an acceptable level of accuracy, the inter-bank market no longer exists. Lloyds sticks its excess into the Bank of England, NatWest borrows to cover any shortfall from the Bank of England. So we've already got the government (erm, you and me actually, our taxes) guaranteeing this wholesale money. As long as the BoE keeps allowing NatWest to access the cash then they can allow any number of people to get their cash back out of the bank and the run, if it happens, won't collapse the bank. The BoE can (and certainly will) provide unlimited liquidity, meaning that RBS will not become illiquid and thus go bust as a result of people hollerin' for their money back.

Which is good really.

It also provides me with a certain perverse amusement. The reason that such a confidence-destroying fuck-up in the computing systems isn't going to cause a Great Financial Crash is that we've just had a GFC.

Taxpayers lose out

However, that's not quite the end of the story. While RBS can and will survive any set of withdrawals it's pretty obvious that it will do so as a much smaller bank (or series of banks if you prefer). The company as a whole has some 7 million account holders. Some small to large number of them will be so pissed off at this past week's events that they will finally go through the grunt work of moving their account to another bank.

It's a strange but true fact that while moving an account is quite easy (it's easier than moving house – yet people move house more often than banks), very few people actually do. The stickiness of personal banking is observable but not really explainable by standard economic theory. It just is: and that is what makes it all very profitable for the UK banks. They get that float in our current accounts for no interest and then lend it out at interest. Sure, your or my account might have £5 in it: but on average over all of the bank accounts in the country one estimate has it that this float earns the banks £20 billion a year.

Lose some number of those accounts and you're shrinking the bank and losing those profits. Which is what will be worrying the executives: how many of those customers are sufficiently pissed off to move? It should also rather worry us as taxpayers for we pretty much own the company and a smaller bank is worth less. And there's really not very much that anyone can do about this now. Near a week without being able to use a bank account properly: yes, that is going to make some people move. This is going to shrink RBS and given that those personal deposits are the cheapest and most stable form of borrowing that a bank can do in normal times the loss of some part of them is going to hurt.

Well, there is one evil thing that could be done: engineer a similar crisis at another bank this week. Then consumers will conclude that all are as bad as each other and thus won't move their accounts. But I have a feeling that managing such a feat deliberately would be rather difficult: as opposed to the ease with which incompetence seems to have achieved it.

The end effect of this is not going to be another bank collapse. Just a larger loss on the taxpayers' shareholding in RBS. Reassuring but not exactly anything to cheer about. ®


No, please, I don't want to hear anything about 100 per cent reserve banking, or the way that banks create money out of thin air. Whether all of this is true or not (it isn't, the banking system as a whole creates credit, not an individual bank's money) is irrelevant. Whether or not we achieve maturity transformation by banks, their purpose, the confidence fairy problem nevertheless continues to exist and RBS faces exactly the problems outlined above.

Other stories you might like

Biting the hand that feeds IT © 1998–2022