There’s a much easier manner of explaining why banks don’t just create money. Which is to note that there are different types of money. There is M0, which is created by the central bank, there are M3 and M4 which are that after the influence of the banking system and the velocity of circulation of money. M0 is not the same as M3 or M4. Thus we are talking abut at least slightly different things. Perhaps narrow money and wide money but best described as money and credit. Money and credit not being the same thing, thus banks create credit, not money.
Or, we can have the long explanation:
Banks do not create money out of thin air
Pontus Rendahl, Lukas B. Freund 14 December 2019
In recent years, some have claimed that banks create money ‘ex nihilo’. This column explains that banks do not create money out of thin air. From an economic viewpoint, commercial banks create private money by transforming an illiquid asset (the borrower’s future ability to repay) into a liquid one (bank deposits); they would quickly be insolvent otherwise. In addition to bank solvency representing a constraint on private money creation, banks require access to liquid reserves in order to be able to engage in money creation.
In contemporary societies, the great majority of money is created by commercial banks rather than the central bank. Whenever a bank makes a loan, it simultaneously creates a matching deposit on the liability side of its balance sheet.1 This happens when, say, a new mortgage contract is concluded, but also seamlessly in everyday life. If, for instance, you pay for your morning commute coffee using your Barclays (or Deutsche Bank or Bank of America) credit card, you have just been handed a Barclays-IOU (or a Deutsche Bank-IOU or a Bank of America-IOU, but you get the flavour). We could also call this IOU a Barclays-pound. Conveniently, that Barclays-pound is denominated in GBP and is treated by your coffee vendor as trading with a one-to-one pegged exchange rate to British pound sterling. Practically speaking, you can make transactions using the Barclays-pound just as well as you could using a cash note printed on behalf of the Bank of England. Although this understanding of the money creation process is hardly new (e.g. Tobin 1963), over the past couple of years, it has been thrown into the public spotlight by the publication of several comprehensive guides on the operational realities of money creation in the modern economy by central banks such as the Bank of England (2014) and Deutsche Bundesbank (2017), as well as by campaigning by such groups as Positive Money.
Unfortunately, the fact that the money we use is mostly issued by private banks and that its creation operationally involves not much more than a few keystrokes has led to a widely circulating but erroneous belief that banks create money out of nothing. To mention but three examples, Richard Werner (2014) writes in a peer-reviewed article: “The money supply is created as ‘fairy dust’ produced by the banks individually, ‘out of thin air’”. Zoe Williams (2017) in The Guardian suggests that “[all] money comes from a magic tree, in the sense that money is spirited from thin air”. And David Graeber (2019) opines in the New York Review of Books: “There are plenty of magic money trees in Britain, as there are in any developed economy. They are called ‘banks’. Since modern money is simply credit, banks can and do create money literally out of nothing, simply by making loans”. This misconception may stem from the seemingly magical simultaneous appearance of entries on both the liability and the asset side of a bank’s balance sheet when it creates a new loan. But this is simply a reflection of double-entry bookkeeping. Economically, money creation by private banks is far from magic, nor is it out of thin air.
There are several ways in which banks’ ability to create money through lending is constrained, meaning that the idea of limitless money creation conjured up by the image of a ‘magic money tree’ is flawed. The aforementioned central bank explainers cover the operational details with much greater expertise than we could claim for ourselves.2 Instead, we want to draw attention to a fundamental economic point that is underlying these constraints. When banks create money, they do so not out of thin air, they create money out of assets – and assets are far from nothing.
A simple parable helps clarify how banks create money and what the role of asset-backing is in that process. Suppose a PhD student new to the British town of Cambridge – let’s call him Lukas – would like to celebrate a day’s worth of work with a pint at a local pub and pay for the drink by issuing an IOU. Unfortunately, the pub refuses to accept the Lukas-IOU. After all, the pub doesn’t know Lukas very well, and it can therefore not trust that he will have the ability to repay the IOU at a later point in time. Moreover, a third party, say a brewery, would not accept the Lukas-IOU as payment for their restocking of the pub’s beer inventory either – the Lukas-IOU is both risky as an asset to hold, and worthless for third-party transactions. Fortunately for Lukas, his supervisor – let’s call him Pontus – happens to have a lot of trust in Lukas, and is willing to accept the Lukas-IOU in exchange for a Pontus-IOU in return. Here is the crux – the local pub does indeed trust Pontus, and so do third parties (“oh, it’s a Pontus-IOU – that’s as good as the pound note in my wallet!”). Lukas can then get his well-deserved drink by paying with the Pontus-IOU. And the brewery can restock their inventory by paying with the same means.
Seemingly like magic, Pontus has just created money out of thin air! But only seemingly. Why would the local pub trust Pontus and treat his IOU as good as money? There are a few reasons. First, they trust his ability to screen Lukas’ repayment capacity, so he has a healthy ‘asset’ backing up his own IOU. Second, he also happens to have liquid reserves on his savings account. Thus, if the pub asked to have the IOU cleared well before Lukas is able to settle his accounts with Pontus, he can always honour his promises by using those reserves. Has money appeared magically out of thin air? No. Pontus has created an IOU that is treated like money by third parties out of Lukas’ repayment capacity, which is equal to a stream of repayments in the future. A stream of repayments is the same as a stream of dividends, so the money Pontus created was out of an asset. If, on the other hand, Pontus were to be so reckless to issue IOUs without the backing of solid assets, or if he didn’t have access to liquid reserves with which to immediately settle any transactions, the air would suddenly get very thick, and the pub and other third parties would soon find out, and his IOUs would lose their value altogether.
When banks create money, the process is very similar to that told in the parable. In the above example, when you paid for your coffee using a Barclays credit card, you gave them an asset – your future repayments – and they handed you a Barclays-IOU in return. These Barclays-pounds are accepted in society as a means of payment (i.e. money), as Barclays is trusted to hold healthy assets as well as British pound reserves issued by the Bank of England.3 That privately issued money is asset-backed represents a fundamental precondition for it to be traded at par with central bank issued money, which comes in the form of currency (which individuals and businesses can use to settle transactions) or reserves (which commercial banks use for the same purpose). In addition to bank solvency representing a constraint on private money creation, banks additionally require access to these public monies in order to be able to engage in liquidity transformation – or money creation.
Lastly, suppose that, for some reason, Barclays’ customers suspected that it held unhealthy assets – that is, there was doubt regarding the repayment capacity of Barclays’ debtors – or that it did not have sufficient reserves to settle transactions on their behalf. This would undermine the pegged exchange rate between Barclays-pounds and British pounds on which the banks edifice stands. Eventually, this would lead to a run on the bank, and it would quickly find itself illiquid, or even insolvent. This is indeed what happened to several banks during the Global Crisis, including Countrywide Financial in the US and Northern Rock in the UK. If banks were indeed able to create money out of nothing, why would we need to bail them out?
Bank of England (2014), “Money creation in the modern economy”, Quarterly Bulletin, 2014 Q1.
Bundesbank (2017), “The role of banks, non-banks and the central bank in the money creation process”, Monthly Report 2017.
Graeber, D (2019), “Against economics”, The New York Review of Books, 5 December.
Tobin, J (1963), “Commercial banks as creators of ‘money’”, Cowles Foundation discussion Paper 159.
Werner, R (2014), “Can banks individually create money out of nothing? – The theories and the empirical evidence”, International Review of Financial Analysis 36: 1-19.
Williams, Z (2019), “How the actual magic money tree works”, The Guardian.
The language we use matters - it provides clarity to our own thoughts and enables…
It is now generally acknowledged that the structure of the NHS needs to be overhauled…
In the film Apollo 13, a loss of oxygen causes the crew to start inadvertently…
There's an idea out there which seems intuitive but then so many ideas do seem…
When we think about the darkly opaque goals of modern central bankers as they relate…
As the papers recently filled with the distressing images of desperate souls looking to escape…
View Comments
That sounds like a clear and convincing explanation to me. But what do I know? :)
Yeah, sure, banks don't create money but something else, though it comes in the form of banknotes and you can spend them and get stuff. I guess the point of this is that banks are nevertheless constrained. And that alone does blow MMT to hell.
Thanks, this explanation helped me understand things a bit better.
However, it’s not true that bank lending is reserve or deposit constrained - they lend based on (perceived) risk and only stabilise reserves separately. This has been shown from a study in 2016 on banking software. The lending capacity isn’t attached to the reserves directly, the two are completely different mechanisms (several large and small banks were used in the study)
https://www.sciencedirect.com/science/article/pii/S1057521915001477
The synopsis is that nothing in the lending systems forces reserves to be topped up. Lending happens independently, and as far as the software is concerned, the money is created from nowhere - the reserves aren’t even considered in the code.
So although there are different mechanisms for ensuring transactional clearing can happen at the end of the “day” there isn’t any software to determine whether or not a loan should be issued based on reserves, or not. You can argue that separate stabilisation mechanisms exist, but as far as the code goes, private banks are “creating money” without any explicit permission from the central banks.
Sure, the bank's creating without reference to the central bank. That's not important.
The loan is indeed constrained though by the reserve and or deposit base. Because no one in a bank lends out money without the permission of the Treasury Department of that bank. That department being in charge of financing, at the end of that day, any loans that have been made.
Whilst it’s easy to accept there are some constraints on lending, by your argument, these are all down to individual psychology.
If the risk of lending is hidden, then it isn’t priced correctly. Similarly, if the constraints are based on imperfect information, then they aren’t actually constraints, but just feelings and tendencies.
Since feelings of what is a safe amount of lending could easily be changed by a call from a government official, where’s the link between reserves/deposits and lending exactly?
All loans must be funded. The bank's books must balance at 4.30 pm. Thus the regulation of how many loans the bank can make out is how much money it can pull in. This is what the Treasury Department in a bank does, funds the loans that have been made - or, obviously, lends out the excess deposits that have been taken.
This doesn't seem difficult to understand to me.
That’s precisely what the study doesn’t show. In reality, they balance so much as there is an accounting entry for the loan that was created, and a corresponding entry for the money that is owed to the bank. No adjustment to the reserve records happens. Therefore they have “created” money. I think you are confusing the theory with the reality of bank software. Please read the study - it shows without doubt that the reserves are not corrected at 4.30pm every day.
You're not grasping the basics of banking here.
I'm entirely willing to agree that banks just make loans. Whisk them into existence. But then - and this is important - they have to fund them. By 4.30 pm that day. This is why the interbank market exists. Because if a bank has made more loans than it received in deposits that day its book is unbalanced. It then balances said book by borrowing from other banks.
If this were not true then Northern Rock wouldn't have gone bust. But NR did go bust.
It doesn't matter what is being done down at the level of the banking software or what assumptions are made there. Banks balance their books every day. They have departments that raise the deposits to fund the loan book.
I see what you are saying, but if I understand it correctly the interbank market handles settlement transactions leaving the bank, i.e debits from accounts.
That’s not the same as it actually backing the loans the bank has made that day. There’s no empirical evidence that the books DO balance each day - in fact the studies seem to show the reverse.
This study specifically deals with an obvious issue - if the bank makes a loan to the customer, but the loan capital doesn’t actually leave the bank, then is the bank required to back it with deposits or reserves? It turns out it’s not. But it does show on the balance sheet.
If banks don’t create money, and have to back all loans with deposits/reserves at the end of each day, then the maximum they could need from the interbank lending market would be the total of the loans they made that day, right?
Of course that is not true in reality, and why NR got into such a sticky situation - they created the loans whenever they wanted to, but the funds can leave the bank much later, and only then would any deposits or reserves need to be called upon. You could create a huge glut of mortgage loans one week and not have to settle them until a month later, when the house purchase completion actually happened.
"but if I understand it correctly the interbank market handles settlement transactions leaving the bank, i.e debits from accounts."
Oh Jesus. No. That's interbank payment system. The market is where banks lend to each other. Also - inaccurately because it can be for longer terms - called the overnight market.
Let's try approaching this another way. You recall that there was something called Libor? (Now Sonia). The interest rate at which.....banks lent to each other. So, why would banks lend to each other? Because they must balance their books. Some will be, on any one day, short of the deposits to back their loans. Thus they attract deposits - borrow from another bank - from other banks. Libor is the interest rate at which they can do this.
Why is there an interest rate at which banks borrow from each other? Because banks do borrow from each other. Why do banks borrow from each other? Because they must balance their books each day. Why must they balance their books? Because each loan must be backed by either capital or deposits.
" and have to back all loans with deposits/reserves at the end of each day, then the maximum they could need from the interbank lending market would be the total of the loans they made that day, right?"
No, it would be - is - the difference between their loans out and deposits in that day. It's only the balance they must go and find elsewhere.
"Of course that is not true in reality, and why NR got into such a sticky situation –"
No, that's not why. And just to be picky about this I have discussed it with the then Chairman of NR, Matt Ridley. He agrees with me.
NR would issue mortgages. On issuance - or, by 4.30 that afternoon - they would be funded by the overnight or interbank market. When they had a nice pile of them then they would issue bonds. The money from these "Granite" bonds then financed the mortgages, tranche by tranche. All of the short term money from the interbank market was paid back when the bonds were successfully issued.
OK.
Then came the Crash. NR had a pile of mortgages issued, financed by overnight money market. They didn't have a big enough pile to market as a bond yet. Every day they go back to the interbank market and borrow for another 24 hours the finance for those mortgages. Then crash - no one would lend them the money. They had to pay back yesterday's 24 hour borrowings, couldn't sell the bonds, couldn't borrow to finance the mortgages for another 24 hours. They're insolvent.
If banks just create their own money then this couldn't have happened, could it? It could only have happened if banks must fund their lending. As it did happen therefore banks must fund their lending.
“No, it would be – is – the difference between their loans out and deposits in that day. It’s only the balance they must go and find elsewhere.”
But it’s not the difference between loans and deposits, it’s the money between deposits and withdrawals. The loans were not fully backed with liquid capital when they were issued, only when money needs to be paid out to other banks.
“Why do banks borrow from each other? Because they must balance their books each day. Why must they balance their books? Because each loan must be backed by either capital or deposits.”
Not each loan on their book - rather each withdrawal to another bank must be backed by deposits or borrowing. That’s not the same thing at all!
You could issue £10b of loans in a day and only actually pay out £1m to other banks, some of which may need to be covered by interbank lending like you mention, but fundamentally, the £10b of loans have still been issued and are on the balance sheet.
The reserves are not altered to show this fact - nothing happens until they actually have to move the money from the bank to elsewhere, in which they need that from somewhere, either deposits or the interbank lending market (or to be even crazier, selling their mortgage book tranches just to pay back their previous cashflow interbank borrowing).
My fundamental point remains - the reserves of the bank are not directly linked to the lending. They can create as many loans they like, when they like, providing they have sufficient cashflow to fund withdrawals to other banks. Of course, if they can’t fund cashflow they become insolvent.
Banks don’t need to fund their loan book, they need to fund withdrawals to other banks. That’s not quite the same thing, is it?
Look, you're not grasping the basic terms of art here. This is one of the things leading you into error.
" loans were not fully backed with liquid capital"
Whadda ya mean liquid capital? Sorta like gold coloured mercury or something? Capital is capital, it's the shareholders equity within the corporation, in this case a bank. You don't back loans with capital. Well, sure, you can, but you don't. It is the potential losses on the loan book that must be backed with capital - this is close to but not exactly the same as the reserve ratio.
All loans must be back by a combination of capital and deposits. If you lend out £100 you must find, from some combination of capital and deposits - recall, you're a bank, so borrowing from another bank is a deposit to you - £100 by 4.30 pm that business day.
That's not liquid capital either. And starting to shout about liquid capital backing for loans shows you've not really grasped the basic vocabulary here. Meaning that it's going to be difficult for you to grasp the points being made if you keep trying to use the technical language.
"only when money needs to be paid out to other banks."
That's not actually a difference with any useful meaning. For most people do borrow to go spend it, right?
"Not each loan on their book – rather each withdrawal to another bank must be backed by deposits or borrowing. That’s not the same thing at all!"
Why not actually look at the accounts of a bank? Loans equals deposits plus capital. It's just not a difficult concept.
As to "only if to be paid to another bank" makes no great difference. Tell me, last time you got an overdraft did you spend it on services from the bank? Did you only spend the money at places which bank with the same bank?
Further, what did I say up top? That the bank has to fund the difference between loans out and deposits in. That's exactly the same thing you've just said, isn't it.
You seem to think you've stumbled upon some great mystery at the heart of banking, something that changes the world. It ain't so, sorry.
"You could issue £10b of loans in a day and only actually pay out £1m to other banks, some of which may need to be covered by interbank lending like you mention, but fundamentally, the £10b of loans have still been issued and are on the balance sheet."
The balance sheet balances, right? So, sure, you offer loans, they don;t get spent outside the bank, so, you've loans issued and deposits to equal them. And? So what?
"The reserves are not altered to show this fact "
Eh? Reserves are the capital that you must assign to a loan. That the loan is still deposited in your own bank doesn't change this.
"the reserves of the bank are not directly linked to the lending.£
What are you talking about? Reserves are prudential. You lend to this type of customer you must allocate x% of the loan as capital against it. This other, less risky type, y%. That's nothing to do with the cashflow of actually funding a loan. You're mixing and matching terms to get into a muddle because you're not being accurate in your use of the terms.
"Banks don’t need to fund their loan book, they need to fund withdrawals to other banks. That’s not quite the same thing, is it?"
What did I say up above? Banks must fund the difference between loans out and deposits in by 4.30 pm every day as they balance their books. I did say that, didn't I? Yes, I can see it, that is what I said. Now you're coming back and saying "But, Aha! Banks must fund the difference between loans out and deposits in by 4.30 pm every day and that's different! Rilly!"
Well, no, it's not actually.
See that was my understanding, but then you see people saying things like this: http://www.youtube.com/watch?v=aIQY44LCIjc&t=6m0s
By my understanding that is simply wrong, because what's to stop the bank lending itself infinite money then buying dividend-paying shares using its infinite money? But Economics Explained certainly seems to be stating that the bank is creating loans from nowhere.
Sure, the bank creates the loan from nothing. But by 4.30 that afternoon must have funded it. The Positive Money peeps etc are forgetting that second bit. You ask the manager for an overdraft. He doesn't worry whether the bank has deposits to cover it. He's got guidance from central office - "You can led up to this amount on your say so" - and perhaps even some targets - "Try to lend £500k this week". He concentrates on lending, sure he does, and doesn't worry about the funding.
But there is a bit of the bank - Treasury - that does nothing but funding.
The system "as a whole" creates credit, sure it does. But an individual bank must fund loans.