Ritchie explains that banks create money for free again

And gets it wrong again, as you would expect:

This is just not true, and so it seems appropriate to give a blog I’ve used twice before another airing. I first wrote the blog reproduced below in September 2007: Northern Rock was falling over at the time and needed bailing out but it’s just as relevant today. The point is a simple one and is that the reality is that confidence is all there is to money because it literally comes out of thin air:
Instead what the bank does is a conjuring trick. They agree to give you a loan. They do it by opening two accounts for you. One is a current account (for ease, let’s assume you haven’t already got one). The other is a loan account. If you borrow £10,000 they mark your current account as having £10,000 in it. You’re now free to spend that however you like.

They also mark your loan account as having £10,000 in it. You now owe that to the bank.

Add the two together and they add up to nothing. One you apparently own (the current account) and one you apparently owe (the loan account). But if you decided to cancel the deal you could straight away repay the loan using the current account and there would be nothing left. Which is why I mean they add up to nothing.

Note there’s no cash involved in this process at all. It’s just an accounting trick. Nothing more.


The great gorgeousness of this is that if banking did work this way then Northern Rock would not have gone bust.

The bank does not just open the two accounts and then forget about it. It must go off and finance that loan from somewhere. It can do that from money extant internally in the bank or from money external to the bank.

If internal it could be the bank\’s capital, or bonds it has issued but most likely it the deposits that other people have made in the bank. I lend money to the bank by making a deposit, the bank lends money by lending that deposit to someone else.

It can also be done externally: as Crock often did. Jim and John get their mortgage, N Rock hands over the money to the solicitor. They also create that loan account for Jim and John: their mortgage. But by 4 pm that day Crock has to find that money from somewhere. Borrow it from another bank was their usual method. On the over night market. Then every few months they would bundle up all those mortgage accounts and issue them as another Granite bond. Take the money in for the bond and pay off all that overnight money they have been rolling over. Then do it all again.

If Crock simply invented the money it issued for mortgages then there would be no Granite bonds. But there are, tens of billions of £s worth of Granite bonds. So Ritchie must be wrong here. Further, if Crock just made up its own money then it wouldn\’t have gone bust. For what actually happened was that the overnight markets refused to roll over their loans. So, Rock was left with no way of funding the mortgages it had already issued but were not yet in Granite bonds. And as they had no funding for those loans they were bust.

This is what is just so amazing about the Murphmonster. He knows that the Granite bonds exist because he\’s written about them. But that they exist shows that banks don\’t simply invent money. They have to find funding for the loans that they make: otherwise, why the fuck do the bonds exist?

38 thoughts on “Ritchie explains that banks create money for free again”

  1. He’s a fucking moron. There’s no point attacking him any more to keep proving this.

    You need to start attacking the morons that are paying his salary…

  2. I thought this man was supposed to have operated as an accountant, yet he doesn’t understand even the basics of banking.

  3. Tim, it is not LOANS that have to be funded. It is DEPOSIT withdrawals – of any kind, including the deposits created in the course of lending. The whole point of lending is that the associated deposits (which are created ex nihilo) are spent – after all, no one borrows money in order to sit on it. The settlement accounting for the drawdown of a loan is exactly the same as it is for any other sort of deposit withdrawal. You and Richard both need to stop thinking in terms of funding loans, and think in terms of funding deposit withdrawals. Then you will get the funding side right.

  4. The comments on his blog are fucking brilliant.

    Someone: Debiting a debtor ultimately means you either credit another debtor, or credit a creditor. Unless banks use single-entry accounting, of course.

    Ritchie: The money is deposited by someone else The system balances

    Someone: So the system balances because there is another deposit? The bank can safely give me money because someone else will give them money? How is that relying only on thin air? You just ridiculed the suggestion that banks can only lend out of deposits…

    Ritchie: But the deposit is made out of thin air

  5. This is the really important debate and it needs to be got right since it questions the whole basis of capitalism and all. Frances Coppola is obviously the go-to person for elucidation since she finds fault with both Murphy (whose account sounds OK to me) and Tim W, so she might be considered unbiased. It would appear from her account that banks’ deposits are created out of thin air (so Murphy-like) .But that the banks have to cover for them when they are withdrawn?(so unMurphy -like) But would n’t they be spent and return to the banking system as new deposits or as increases to existing deposits? Would she care to go through it again more slowly ?Speaking personally, I have learned more about ,for instance QE , from her than from big(ger) -name commentators.

  6. Frances is correct. It is not the creation of a loan that needs to be funded, nor even the drawing of the loan if the proceeds remain on the deposit at the same bank, but the transfer of that money to another bank. You see it’s … oh what the hell. i really can’t be arsedat 4:30 on a Friday afternoon.

  7. Look, I know nothing about banking whatsoever but even I can see that Richie’s logic runs out as soon as the borrower withdraws the fresh air out of his current account.

  8. I wish somebody would just write a definitive what have you on this. The internets is full of people arguing about banking and shit, and no two of them agree how it actually works.

    I mean seriously, come on. This isn’t some part of the natural world we don’t fully understand yet. It’s just a friggin’ accounting system. Can’t be that hard.

    But I can guarantee that if one person anywhere on the internets says, “banking, what happens is X” somebody else will say, “no, you’re talking cock, what happens is Y” and then somebody else will say, “that’s a load of pants, are you fools, it’s Z” and then DBC Reed will arrive and say it’s all house prices, and round and around it goes.

  9. @DBC Reed

    I think Frances is technically correct, but for most practical purposes Tim’s take is also correct.

    Consider Mr A, who borrows 100K to buy a house.

    The bank does indeed effectively set up two accounts, writes -100K in one, and calls it Mr A’s debt, and adds 100K to the other, which is Mr A’s spending money.

    Richard Murphy is almost right at this point, the bank have “created” money, in a much as 100k which didn’t exist is now in Mr A’s bank account, however they have also (sort of) created anti-money (Mr A’s 100k debt).

    However, which Tim is pointing out, and the Murthmister doesn’t seem to get is that few people borrow 100k so they can look at their current account figures and go ‘isn’t is nice, I’ve got loadsofmoney’. No, they go out and spend it, probably more or less in one go.

    So, they instruct the bank to send their 100k to someone else (Mr B), in return for a house (or whatever). Now, the bank has to actually rustle up 100k of real pound notes, to hand over to the bank of Mr B, in order that Mr B may get his money. The bank has to find (i.e. either take in deposits, or borrow) an actually 100k at this point.

    So to summarize – as usual, Murphy is pretty much completely wrong, Frances is totally correct, and Timmy is correct for most practical purposes.

  10. Richie is an idiot.

    A word about securitisation. This is simply a way of providing credit which takes the banks out of the picture all together.

    The lender is simply an investor who wishes to make a loan in return for an attractive rate of interest – how attractive the interest rate is depends on the lender’s appraisal of the borrower’s credit quality. For a risky borrower, the lender may demand an interest rate of 8 or 9%. For a less risky project, the lender may be happy to accept 5%.

    In the securitisation process, banks simply sell the loans that they had on their loan books to investors. And to make it efficient with economies of scale and to give the investors diversification, they tend to package up these loans and sell them in packages of several thousand. Mortgage loans are a major source of this sort of lending as they tend to be large secured loans.

    When a bank does a securitisation, what happens is that the loans which were on the banks balance sheet are transferred into a separate company called an SPV. The bank takes a fee for this and it is no longer liable for the loans. They do not belong to the bank anymore. This is very important. Northern Rock do not own the Granite bonds.

    The money that was lent by the banks originally is paid back to the banks out of the initial money paid to the SPV by the new investors and the loans are now between the SPV and the investors. The original bank is not involved.

    This was a very popular business before the crisis as cash rich Asian investors were keen to buy assets with attractive interest rates. At the same time, banks were happy to reduce their balance sheets creating capacity which could then be used to make new loans.

  11. Also, one reason why these days only 3% of money is hard cash is because of credit cards and debit cards. We do not need to carry around wads of it.

    The BoE is constantly pumping in new money through its open market operations. QE is the same process in a very big scale.

  12. Well, what do you expect from an accountant? He looks at the bookkeeping, sees numbers, imagines it’s all made up.

    What he doesn’t seem to get is that every loan from a fracti0nal reserve bank increases their asset holding obligations. Which are there to cover bank runs (for a while). So it isn’t free or invented. I somehow doubt banks can write whatever loans they like and create money. How they document the transaction is completely irrelevant. Column A, column B, who cares. At the end of the day (literally) they have to be able to put their hand on their hearts and say they have this level of assets to cover defaults and withdrawals.

    This is the really important debate and it needs to be got right since it questions the whole basis of capitalism and all. Frances Coppola is obviously the go-to person for elucidation since she finds fault with both Murphy (whose account sounds OK to me) and Tim W, so she might be considered unbiased.

    Two problems with this DBC
    1. This is not the basis of capitalism! Capitalism existed long before joint stock companies and FRB banking. It was all that did exist .

    2. Just because someone finds fault with two other, diametrically opposed people, it doesn’t mean she’s right. What if one of them are right? Evaluate the arguments, not the perceived value of the person.

  13. This is the point at which capital requirements also kick in. As FC quite correctly notes the money is required to pay deposits – and as theprole notes this is where the money goes to a third party – the deposit of loaned money is used to pay for a house.

    NR borrowed most of this money, either via securitisation or via direct borrowing from the capital markets. But they are not allowed to borrow all of it. Because this is risky, and thus we have bank capital requirements. This is the Basel or BIS thing. If you read the headline numbers, this says that NR had capital of 15% so of every

  14. “When a bank makes a loan, it simply adds to the borrower’s deposit account in the bank by the amount of the loan. The money is not taken from anyone else’s deposit; it was not previously paid in to the bank by anyone. It’s new money, created by the bank for the use of the borrower.”
    – Robert B. Anderson, Secretary of the Treasury under Eisenhower, in an interview reported in the August 31, 1959 issue of U.S. News and World Report

  15. “Do private banks issue money today? Yes. Although banks no longer have the right to issue bank notes, they can create money in the form of bank deposits when they lend money to businesses, or buy securities. . . . The important thing to remember is that when banks lend money they don’t necessarily take it from anyone else to lend. Thus they ‘create’ it.” – Congressman Wright Patman, Money Facts (House Committee on Banking and Currency, 1964)

  16. “…banks extend credit [i.e. make loans] by simply increasing the borrowing customer’s current account…That is, banks extend credit [i.e. make loans] by creating money.” Paul Tucker, Deputy Governor of the Bank of England, Speech on 13th December 2007, p9.

  17. “When banks extend loans to their customers, they create money by crediting their customers’ accounts.”

    Sir Mervyn King, Speech to the South Wales Chamber of Commerce at The Millenium Centre, Cardiff on 23rd October 2012.

  18. I lot of this ISTM is confusion about what the word “money” is referring to.

    Originally, money was a commodity, let’s say “gold” (in the same type of usage as Ricardo used “corn” for all agricultural production). So, you have your dungeons and dragons economy where money is lumps of gold. Money is just standardised lumps of gold. A pound is a pound of gold.

    So, banking. Once the banking system develops, the gold stops being spent as money. It just moves from bank to bank and “the money” is receipts for gold. Money is now not gold. It is a contract for gold.

    The current system evolved from the gold system. The commodity now is not gold. The commodity is bank reserves. It is the figures on the Bank of England’s big account ledger, which they update every day with the quantity of reserves- the “virtual gold”- in each bank’s virtual vault.

    So “the money”, which was a contract for reserve gold, has now become a contract for reserve money, the M0. But the M0 isn’t money, because nobody spends that. They spend the broad money, the contracts for M0, the M2 or 3 (we’ll call that Mx, then).

    So with that understanding, it does become apparent that the Mx in commercial bank accounts can’t be dependent on other Mx. It depends on the reserve money, the M0. The regulation states that a bank can have some multiple (10 for simplicity) of its Virtual reserves out on loan.

    So on that basis, it would appear that when banks are making loans, they aren’t getting the money from anywhere else, as such. They’re just keeping within a simple arithmetical limit set by Basel. The reserves can increase from a variety of sources; from customer transfers of Mx between banks “dragging” reserves from one BoE account to another, from interest, or from gambling on the stock market, etc.

    So, there does not seem to be a direct connection between money loaned out and money paid in. On that basis, the charge that banks “create” money from “thin air” is valid to some degree, dependent on your use of language. The implication in that usage however is that the money creation is arbitrary and unlimited, which it is not. But the “canonical” descriptions of FRB as you find on Wiki etc are not really, as Frances pointed out before, how the system really works. It seems to me. YMMV, and all that.

  19. IanB

    You’re getting there, but you’re not quite right. Two points:

    1) The Basel limits are on capital, not reserves. Borrowing – of any kind – is NOT Tier1 capital under Basel rules. Tier1 capital is shareholders’ funds – in any other sort of company it would be called “equity”. Basel specifies the proportion of total (risk weighted) assets that must be backed by equity not debt. Does that help?

    2) Availability of reserves for settlement does not restrict bank lending in any way, since as a last resort central banks can always create reserves to settle transactions. There is no reserve “multiplier” as you suggest.

    That’s the difference between a fiat currency system and a gold standard, really. Central banks can’t create gold, so in a gold standard they restrict the amount of borrowing banks can do (note borrowing, not lending – it’s back to funding again) against the available gold reserves. That is the “multiplier” – and the size of the multiplier really depends on the velocity of money. Under a fiat money system this restriction is not necessary and is never applied in practice even when there is a positive reserve requirement (as in the US, for example).

  20. It is possible to agree with the statements of Mervyn King and still disagree with Ritchie. Ritchie claims that there is no limit on the loans that banks make and he seems to suggest that the interest is free profit.

    Can we agree that every penny of loans made by a bank is accounted for as an asset on the balance sheet and is balanced exactly by a deposit which are accounted for as a liability.

    The bank cannot lend more than it has on deposit plus what it has borrowed.

    It can also have more deposits than it has loans. In this case it has excess reserves.

    Banks tend to make loans first and to then fund the money for the loan from its deposits. If those deposits are not there then they will use the interbank or debt markets to get the money.

  21. Frances-

    It seems to me that the conceptual difficulties we all have are down to the reality that banks don’t simply get money inflow from deposits. They can increase their assets by a variety of means, at which point the “canonical FRB” description becomes less and less useful.


  22. Where I strongly disagree with Tim is his claim that Granite bonds prove that banks require finance to offer the loans. They don’t – to a large extent, banks can offer the loans without the bonds (and indeed, they did). This is all a simplification of course (despite its length) but I hope it’s useful to somebody.

    “Money” is “created” by banks in the following sense:
    (1) Imagine Bank1 has

  23. Not sure why my last post got cut off – if it doesn’t all post this time I’ll give up. Apologies for spamming the forum.

    Where I strongly disagree with Tim is his claim that Granite bonds prove that banks require finance to offer the loans. They don’t – to a large extent, banks can offer the loans without the bonds (and indeed, they did). This is all a simplification of course (despite its length) but I hope it’s useful to somebody.

    “Money” is “created” by banks in the following sense:
    (1) Imagine Bank1 has £1m. On day 1, it lends this to Borrower1, who buys a house or something. The seller (Seller1) deposits the £1m in Bank2.

    (2) On Day 2, Bank1 borrows the £1m from Bank2, and lends it to Borrower2, who buys from Seller2, who deposits in Bank2.

    (n) On Day n, Bank1 borrows the £1m from Bank2, and lends it to Borrower n, who buys from Seller n, who deposits in Bank2.

    After n days, Bank1 has £n million of assets, and Bank2’s deposit holders have £n million of deposits. This is all backed up by the same £1m of cash, basically going around in circles. So “money” has been “created”. In practice, the money doesn’t have to move – most of this can be paper transactions. This could in theory go on forever, so regulation puts limits on how much banks can lend based on how much capital they hold (so that if deposit holders actually want to withdraw they can – the theory being that not everybody will demand their money at once….). Bank1 therefore cannot create an infinite amount of money overnight (its ability to create wealth is limited by its liquidity and capital requirements) but that doesn’t change the fact that money is being created by the banking system.

    I’m treating Bank1 and Banks2 as separate entities; but at the same time, the mirror image of the above is occurring (i.e. Bank2 is lending to customers, while Bank1’s deposit holders accrue deposits).

    So why did NR use Granite bonds? A banks ability to lend is still limited by its liquidity, and by the regulatory capital requirements. Say you have lent money in a loan portfolio with principal of £100m; over the lifetime of the loans, you expect to receive interest of £100m. For regcap purposes, only the principal of £100m might count towards your capital base; however, if you sell the portfolio to a third party for £150m, then that £150m of cash is all allowable as regcap and can be used to fund further loans. The bet that the bank is taking is that the new loans will make more money than the interest they gave up.

    Securitisation is therefore not a necessary part of the lending process – it was used by NR to speed up their growth.

  24. Securitisation was used by NR to grab a market share that was much larger than their deposit base allowed and to therefore allow them to be a major player.

    They earned fee income and other insurance income from providing mortgages even if they did not warehouse those mortgages on their balance sheet.

    Their funding model was flawed – they relied too much on short term funding. Remember the banks do something called maturity transformation which means that short term deposits are matched against long term loans.

    What blew up NR was the fact that after Lehman blew, no-one wanted to lend. Many banks were able to survive this period as they already had a large part of their funding needs locked in. But NR did not. They NEEDED money every week or so. And the markets would not lend for longer than that and they would certainly not lend to NR as the poor loan quality of NR’s loan book was known. After all, they had been providing 120% mortgages.

  25. Oops! Ignore what I said about Lehman. Lehman blew a year later. NR could not get funding as their loan book was viewed as weak.

  26. Frederick,

    You’ve forgotten about the Bear Sterns hedge funds. Northern Rock had become increasingly dependent on short-term wholesale funding because the securitisation market was drying up – it was taking longer to get the mortgages packaged up and sold on, and they were generating less income. Markets were aware of NR’s funding fragility and its cost of funding rose steadily throughout the summer of 2007. But the proximate cause of NR’s liquidity crisis was the failure of the Bear Sterns hedge funds in July 2007, which caused the closure of the ABCP funding market. Germany’s Deutsche Industriebank IKB failed first, on July 30th, followed by NR in September. The ECB kept the rest of the Euro area banks going with large amounts of liquidity, the Feds did the same in the US, and the FHA bailed out loads of mortgage originators that experienced the same sort of finding crisis as NR.

    The best narrative on the European dimension of the 2007/2008 crisis is in the Liikanen report:


  27. What is money? [Insert textbook definition here.] QED, banks do not create money. Arguably, they may create currency, but obviously that’s not the same thing.

  28. One useful definition (M4, the BoE’s measure of the UK money supply) is circulating cash plus bank deposits. Under that definition, banks do create money.

    What Murphy may not have grasped is that a bank can create additional money for the whole system, but not specifically for itself. Because, as Frances and others have commented, withdrawals do have to be settled.

  29. Of course, he’s right in accounting terms.

    But this is an example of Interesting but Decidedly Irrelevant Observational Truth Syndrome.

    When your African guide points to the crazed herd of elephants charging toward you and says “They are all grey”, you don’t thank him for it do you?

    Perhaps the best way to refute his crazed argument is to ask your neighbour if he feels guilty that he didn’t lend the Crock

  30. PaulB at 31.

    Banks can create money for themselves – loans to borrower, borrower buys shares in bank – but this method tends to lead to jail.

  31. Dave, I don’t know what textbook you are using, but according to me, if I can buy you a drink with it then it’s money. Any drink I buy you will be paid for with money from my overdraft, which has of course been lent to me by the bank….Banks create money. QED.

  32. Would it be correct to summarise it as “Banks can create money out of nothing for other people, but not themselves, but can profit from the interest on that created money”? With the caveat that they still have to fund the withdrawal of the created money by some means, be that deposits, or loans from other sources?

  33. I recall from school economics that money represented a potential claim on goods and services. If it buys a drink in the pub, or a house, it seems like money to me.

Leave a Reply

Your email address will not be published. Required fields are marked *