avatar

Wrong Way Corrigan Strikes Again

by George Selgin March 19th, 2014 5:20 pm

Over at Zero Hedge, "Tyler Durden" reposts a piece by Sean Corrigan, the link to which appears to be non-working, in which Corrigan imagines that he is having a "Gotcha!" moment, at the expense of yours truly, because he has discovered some reputable authorities who claim that (commercial) banks are in fact capable capable of making loans above and beyond any sum of deposits they acquire.  Here are the first few paragraphs:

Of late there has been much breathless wonder expressed at the Bank of England’s supposedly ground-breaking release. ‘Money in the Modern Economy’, in which it argues – shock! horror! - that banks do not lend out previously received deposits, but that they create the latter ex nihilo by first making loans. Alas, as Gunnar Myrdal waspishly observed of Keynes himself, this has been a reaction plagued with the ‘unnecessary originality’ of those who don’t know their literature.

As an example, some few months ago, I had an exchange with the disputatious George Selgin (he of the perfervid fractional free banking bent) in which I cited – after a good twenty minutes’ research – the following authorities to that very same effect:-

Roepke from a footnote (p113) to his 1936 work, ‘Crises & Cycles’:

The process [of credit creation] is now clearly explained in any text-book on economics, banking or money (especially recommendable is Hartley Withers’ Meaning of Money). A fuller treatment may be found in the following books: R. G. Hawtrey, op. cit.; J. M. Keynes, A Treatise on Money, pp. 23-49 : C. A. Philips, Bank Credit, New York, 1920; W. F. Crick, “The Genesis of Bank Deposits,” Economica, June 1927, and F. A. von Hayek, Monetary Theory and the Trade Cycle, London,1933.

 

Without an understanding of this process and of its limitations, no real insight into the working of our banking system and, consequently, of our entire economic system seems possible, to say nothing of the mechanism of business cycles. There may still be many people who can no more believe the story of the genesis of bank money than they can believe the genesis of the Bible, but on the whole it now seems to be generally accepted. A last but hopeless attempt at disproving it has recently been made by M. Bouniatian, Credit et conjoncture, Paris, 1933. [Emphasis mine and apparently NOT the last!]

Or as Hayek indeed noted in ‘Prices and Production’ above his own lengthy footnote (pp 81-2):

The main reason for the existing confusion with regard to the creation of deposits is to be found in the lack of any distinction between the possibilities open to a single bank and those open to the banking system as a whole.

Actually, Sean, I can assure you that I've read them all, and carefully; what's more, I've probably taught several thousand students about the process of multiple deposit expansion, being careful in doing so to make precisely the distinction Hayek insists upon between what individual banks on one hand and the system as a whole on the other are capable of making doing with any fresh deposits.* In fact, no individual bank involved can lend more than a part--usually the greater part--(the "excess reserves") of whatever fresh funds it is able to secure. It is precisely your misunderstanding of Hayek's point that was, I am inclined to think, the nub of our original disagreement!

*Here FYI is a screenshot of the relevant page of the class notes I give to my students:

NotesPartII_Page_010

30 Responses to “Wrong Way Corrigan Strikes Again”

  1. avatar Mike Sproul says:

    George:

    If you are thinking in terms of the Irving Fisher intertemporal choice model, then it is true that the amount of corn borrowed must equal the amount of corn lent. But you go beyond that and claim that no individual bank can lend more than a part of the cash deposits it has received. The deposits, of course, can be in the form of bits of paper and bookkeeping entries, and so the bank's lending is not constrained by its deposits, at least not in the sense that you mean.

    A private bank receives $100 Fed dollars on deposit, for which it issues $100 checking account dollars. You would agree that if the bank were to lend some fraction of its Fed dollars, which were re-deposited and re-lent several times, that the bank could end up with $300 of checking account dollars created by the loan expansion process, against which the bank holds $100 of Fed dollars as reserves, plus $200 of IOU's from the bank's various borrowers.

    But instead of this textbook process, suppose that the bank starts with the same $100 Fed dollars on deposit, and then makes 10 loans of $20 each by crediting $20 checking account dollars to the account of each borrower. Once again the bank ends up with $300 of checking account dollars, against which the bank holds $100 of Fed dollars as reserves, plus $200 of IOU's from the bank's various borrowers. Keep in mind that as one customer spends his $20, some fraction of that $20 will be redeposited by the recipient in the same bank, thus relieving the bank of the need to pay out the full $20 in Fed dollars.

    One might claim that the $300 of checking account dollars created in this "depositless" process cannot all be paid by the issuing bank, but that is equally true of the bank that issued $300 checking account dollars in the textbook fashion.

    The trick is to recognize that the first $100 checking account dollars issued (by either bank) were issued in exchange for "deposits" of Fed dollars, while the next $200 of checking account dollars were issued in exchange for "deposits" of IOU's. That's the only sense in which a bank's lending must equal its deposits.

    • avatar George Selgin says:

      I can only say, Mike, than in general the kind of redeposits your story relies upon don't typically go far in allowing banks to lend considerably beyond their actual reserves. C.A. Phillips considers the matter at length; and I myself explain in Theory of Free Banking (see the opening of chapter 3)why the redeposits don't alter maximum lending.

      • avatar Mike Sproul says:

        I presume you mean that money lent by some bank will likely be redeposited at some other bank, thus limiting the loan expansion of the individual bank, even as it allows loan expansion for banks as a whole. Of course that is not a problem for an economy with few banks or only one bank.

        But let's take the case of a bank that is very small relative to the market as a whole, so that hardly any redepositing occurs. Suppose the bank holds $100 Fed dollars against $100 checking account dollars. Then a single customer asks for a loan of $200. The customer probably does not want to borrow $200 Fed dollars, but $200 checking account dollars. So the bank credits his account with $200, and now the bank has issued $300 checking account dollars, backed by $100 Fed dollars plus the borrower's $200 IOU. The loan expansion process has worked, and the bank's lending has exceeded its deposits of cash. If the borrower wrote a check for more than $100, then the bank would have to sell some of its IOU's to cover it, but that would be the case if the bank had followed the textbook loan expansion process as well.

        Also, note the shift in your position in the two sentences below.

        " no individual bank involved can lend more than a part--usually the greater part--(the "excess reserves") of whatever fresh funds it is able to secure."

        "in general the kind of redeposits your story relies upon don't typically go far in allowing banks to lend considerably beyond their actual reserves."

    • avatar vikingvista says:

      "The deposits, of course, can be in the form of bits of paper and bookkeeping entries, and so the bank's lending is not constrained by its deposits, at least not in the sense that you mean."

      I don't see what difference this makes. It doesn't affect Selgin's argument. It makes no difference if the basic money is digital or, whether digital or physical, if the actors keep track of ownership entirely through their books without ever going into the bank vaults (if any). What matters is that the accounting is correct.

      "But instead of this textbook process, suppose that the bank starts with the same $100 Fed dollars on deposit, and then makes 10 loans of $20 each by crediting $20 checking account dollars to the account of each borrower."

      In the first case the bank is constrained by available funds. In the second case, there is no such constraint. If the second were true, why would the bankers care about deposits or loans at all? They can simply create money and be as rich as they want. So, in the first case, the $300 final state is not a matter of choice, it is a matter of attracting sufficient funds. In the second case, the $300 final state is apparently something banks deliberately do just to prove your point, since such a process imposes no such restriction on them.

      • avatar Mike Sproul says:

        "If the second were true, why would the bankers care about deposits or loans at all? They can simply create money and be as rich as they want."

        The banker that "creates money" is adding $200 of checking account dollars to his liabilities while he adds $200 of IOU's to his assets. This does not affect the banker's net worth.

        Think of a note-issuing bank. It starts by issuing $100 of its notes in exchange for $100 of coins deposited, then it prints and lends $200 of its notes to various borrowers, in exchange for $200 of the customers' IOU's. As long as there are people willing to hold those "extra" $200 of notes, the banker can keep them in circulation, and his lending of notes is in no sense constrained by the amount of coins he got on deposit. The real constraint on his lending is the the amount of IOU's that his customers are willing to "deposit" in exchange for notes lent. (The same is true if the bank issues checking account dollars instead of notes.)

        • avatar vikingvista says:

          "This does not affect the banker's net worth."

          I understand your argument. My question is, if a banker could do what you say, then why would she settle for not affecting her net worth? If people want to borrow the bank's fiat credit, then the bank's credit is money (widely accepted in trade). The banker could make her life a lot easier by simply printing up whatever credit she wanted, to buy whatever she liked, to make herself as rich as she desired.

          To say that you've identified an equivalent process to standard banking theory, allowing a bank to operate with essentially less than zero reserves rather than loaning only available funds while maintaining positive reserves, is rather like saying that walking from NY to London is the same as flying because you wind up in London either way. But if you try it, you will find out that you never will.

          • avatar Mike Sproul says:

            If the creation and lending of $200 checking account dollars increased your banker's net worth by $1, then she would do it. If it reduced net worth by $1 then she wouldn't do it. If it had zero effect, then she would be indifferent to doing it. Remember that "zero profit" only means that the firm is earning a competitive rate of return on its capital.

            It the banker spent the newly-issued $200 on $200 worth of new assets for the bank, then the bank's money would be adequately backed and would hold its value. If the banker spent the $200 on candy, which she ate, then the bank's money would lose value. There would be only $100 of assets backing $300 of the bank's money, making $1 of the bank's money worth $1/3 of a Fed dollar.

            The bank that issues $300 checking account dollars against $100 Fed dollars (+ $200 of IOU's) is operating on 33% reserves, not "less than zero" reserves.

          • avatar vikingvista says:

            “If the creation and lending of $200 checking account dollars increased your banker's net worth by $1, then she would do it.”

            Again, why on earth would she stop at $200? Why not print $1 trillion checking account dollars and buy up NY real estate? Isn’t NY real estate adequate “backing”? If anyone comes asking for debt repayment, just print up more checking account dollars to give them since they are readily tradable for anything priced in that nominal amount. Of course, that wouldn’t work if standard banking theory were true, and each issuance of such credit is entirely dependent ultimately upon the widespread belief in redeemability in basic money. But that is not your theory.

            “The bank that issues $300 checking account dollars against $100 Fed dollars (+ $200 of IOU's) is operating on 33% reserves, not "less than zero" reserves.”

            Except that your reserves are irrelevant to your example. They just sit there like a cat at a dog show. There is no reason in your theory, other than trying to demonstrate your example, why anyone would do that. Specifically, as you said…

            “suppose that the bank starts with the same $100 Fed dollars on deposit, and then makes 10 loans of $20 each by crediting $20 checking account dollars to the account of each borrower.”

            What is the point of the $100 Fed dollars on deposit? It never gets used in your example. Under your theory, fiat bank credit has the same nominal value as the Fed dollars. If the bank is capable of just “crediting 20 checking account dollars”, the example can proceed the same with or without the deposit. It is only in the textbook theory that the deposit is essential, since deposits are always the only source of funds to lend.

            So, if you ignore the vestigial $100 deposit, as any banker believing your theory would, then for that bank you have after the first set of loans: Loans = $200, Deposits = $0, and Reserves = Deposits - Loans = -$200. Or if you insist on counting the fiat checking account dollars as true deposits, then Reserves = $0. However, I can assure you that any competitive bank practicing this way would find the sum total of its fiat “deposits”, regardless of their pronounced nominal value, equal to $0 on trade.

            It seems the bank in your example is as likely to reach the textbook-equivalent 33% reserve ratio as a NY swimmer is to reach London.

          • avatar Mike Sproul says:

            "why on earth would she stop at $200? Why not print $1 trillion checking account dollars and buy up NY real estate?"

            For the same reason that a farmer won't produce a trillion tones of wheat. Any business will grow to the point where further growth is not profitable.

            "your reserves are irrelevant to your example."

            Reserves are there to satisfy the customers who prefer coins or paper dollars to checking account dollars.

            "So, if you ignore the vestigial $100 deposit..."

            The $100 is part of assets, as are the $200 of IOU's. If the bank spent the $100 of reserves on bonds, then there would be $100 of bonds +$200 of IOU's backing $300 checking account dollars. Then you'd have zero reserves. This is actually possible as long as customers are OK with holding checking account dollars instead of paper dollars. It was standard procedure with American colonial banks. The bank would hold a 100-shilling IOU (backed by 200 shillings worth of land) and would issue 100 paper shillings to the borrower. That bank has zero reserves, although it is still a 100% asset bank. Customers were OK with holding paper shillings rather than coins.

          • avatar vikingvista says:

            ME: why on earth would she stop at $200? Why not print $1 trillion checking account dollars and buy up NY real estate?
            DU: For the same reason that a farmer won't produce a trillion tones of wheat. Any business will grow to the point where further growth is not profitable.

            Except in your theory, profit is completely fiat. It is whatever the bank wants it to be. The bank doesn’t have to produce anything. All it has to do is create $1 trillion of checking account money and use it to buy up already existing NY real estate.

            ME: your reserves are irrelevant to your example.
            DU: Reserves are there to satisfy the customers who prefer coins or paper dollars to checking account dollars.

            If, for some very odd reason, there are people in your scenario who will not accept fiat checking account dollars in place of “coins or paper dollars”, in spite of the fact that the bank’s checking account dollars are widely accepted in place of “coins or paper dollars” at full nominal amount, then you have to wonder why the bank would be in the business of redeeming “coins or paper dollars” at all. It is chump change, in your example.

            Your theory has completely decoupled true deposits from loans. They are completely independent. The deposits have nothing to do with the loans. As such, each independent business, (1) the loaning of fiat checking account dollars, and (2) the receiving and redemption of “coins and paper dollars” stands entirely on its own. But the former business is absurdly more profitable.

            ME: So, if you ignore the vestigial $100 deposit...
            DU: The $100 is part of assets, as are the $200 of IOU's. If the bank spent the $100 of reserves on bonds, then there would be $100 of bonds +$200 of IOU's backing $300 checking account dollars. Then you'd have zero reserves. This is actually possible as long as customers are OK with holding checking account dollars instead of paper dollars. It was standard procedure with American colonial banks. The bank would hold a 100-shilling IOU (backed by 200 shillings worth of land) and would issue 100 paper shillings to the borrower. That bank has zero reserves, although it is still a 100% asset bank. Customers were OK with holding paper shillings rather than coins.

            Fine. This bank chooses to own $1 trillion of real estate. If that isn’t satiating enough for them, the next day they will choose to own $5 trillion of real estate. With those amounts available using fiat checking deposits, they certainly would have no interest in a paultry cash deposit/redemption business which has nothing to do with their fiat checking.

            This, of course, would not work with the textbook banking theory, because in textbook theory, the loans are completely dependent upon deposits--No deposits, no loans, and no profits. Your theory decouples that dependence.

          • avatar George Selgin says:

            I don't want to find myself once again engaged in the task of teaching people elementary banking theory, but I must say Mike here doesn't seem to understand the role of check clearings in constraining bank lending. His argument concerning why a bank won't want to write up an infinite amount of loans is, given this lack of understanding (and corresponding assumption that reserves are only needed when yokels withdraw actual hard currency) quite unsound. The wheat analogy is itself utterly inadequate, since wheat costs real resources to produce, while loans (according to Mike's understanding) can be created ex nihilo.

          • avatar Mike Sproul says:

            "All it has to do is create $1 trillion of checking account money and use it to buy up already existing NY real estate."

            This is a classic case of taking a reasonable proposition like "A bank issues $100 of new liabilities and uses it to buy $100 worth of new assets", and extending it to the point where one firm buys the whole world. It doesn't happen. The firm only expands as long as expansion is profitable. Banks, like wheat farmers, stop expanding when expansion is not profitable.

            "fiat checking account dollars"

            If you mean "unbacked checking account dollars", there is no such thing. All checking account dollars are the liability of the issuing bank, and are backed by the issuing bank's assets.

            "Your theory has completely decoupled true deposits from loans."

            That's because a cash deposit is not the only kind of deposit. A customer who wants to borrow 100 checking account dollars from a bank can "deposit" his $100 IOU, and that is enough for the bank to issue 100 new checking account dollars on loan. When I swipe my credit card, the credit card company creates 200 credit card dollars, not because they have received a $200 cash deposit from me, but because they have received my $200 IOU on deposit. When the American colonies issued paper shillings, either on loan or by direct spending, they did it either because they received the borrower's IOU on deposit or else they pledged future taxes to cover the shillings issued. But there was never any significant deposit of cash, even though the colonies issued large amounts of notes.

          • avatar vikingvista says:

            “This is a classic case of taking a reasonable proposition like "A bank issues $100 of new liabilities and uses it to buy $100 worth of new assets", and extending it to the point where one firm buys the whole world. It doesn't happen. The firm only expands as long as expansion is profitable. Banks, like wheat farmers, stop expanding when expansion is not profitable.”

            Of course it doesn’t happen. It doesn’t happen, because your theory is incompatible with reality. Please explain at what point using fiat checking account dollars stops being profitable for the bank. It is easy to explain for the textbook theory. For your theory, it is inexplicable.

            “If you mean "unbacked checking account dollars", there is no such thing. All checking account dollars are the liability of the issuing bank, and are backed by the issuing bank's assets.”

            Fine. Your backing theory doesn’t make sense, but I’m giving it to you for this example. If the bank creates checking account dollars to buy up New York real estate, then the NY real estate is “backing”, is it not?

            ME: Your theory has completely decoupled true deposits from loans.
            DU: That's because a cash deposit is not the only kind of deposit.

            No, it is because cash deposits are completely irrelevant to loans in your example. The only deposits that pertain to your example are the IOU deposits (as an inseparable part of any debt). But the only reason the borrower wants the bank’s checking account dollars is because merchants ALREADY generally accept them on trade, before the borrower’s IOU was ever even contemplated. So, why would the bank even bother with borrowers at all? Since people already accept them on trade, it can turn its fiat checking account dollars immediately into hard assets.

            This, of course, doesn’t work with standard banking theory, because all of the bank's debt is understood, and found by experience, to be redeemable in a specific nominal quantity of cash. That cash redemption reputation must be both established, and regularly maintained as its debts are cleared. Cash deposits are essential for textbook banking theory, but irrelevant to yours. All deposits wind up being irrelevant in your theory.

            “When I swipe my credit card, the credit card company creates 200 credit card dollars, not because they have received a $200 cash deposit from me, but because they have received my $200 IOU on deposit.”

            When you swipe your credit card, the issuing *bank*, not the credit card company, credits the merchant (redeemable by the merchant in cash), and adds a corresponding debt to your account. Yes, the bank is accepting an IOU from you, but that is an inseparable part of the debt coin. By definition, a debt is a trade of something in exchange for an IOU. That being the very definition of a debt, it is semantically impossible for any debt to not be “backed”, in your theory. Again, this just shows how your theory essentially turns the bank’s fiat checking account dollars into basic money at the bank’s disposal--which is how human beings would use it if your theory were correct.

          • avatar Mike Sproul says:

            "Please explain at what point using fiat checking account dollars stops being profitable for the bank."

            It's impossible to explain anything as long as you believe that checking account dollars are "fiat money". They are backed by the assets of the issuing bank.

            " If the bank creates checking account dollars to buy up New York real estate, then the NY real estate is “backing”, is it not?"

            Correct.

            "Cash deposits are essential for textbook banking theory, but irrelevant to yours. All deposits wind up being irrelevant in your theory."

            Cash deposits are just as relevant as a "deposit" of borrowers' IOU's or government bonds. They are all assets, and they all back the money issued by the bank that holds those deposits

            " it is semantically impossible for any debt to not be “backed”, in your theory."

            No sensible banker or credit card company would knowingly issue a checking account dollar (or paper dollar, or credit card dollar) without getting at least a dollar's worth of assets (backing) in exchange. That is not to say that bankers are always sensible, or that their assets don't ever fall in value.

            " the issuing *bank*, not the credit card company"

            Sometimes banks issue credit cards, and sometimes they are issued by credit card companies.

          • avatar vikingvista says:

            “It's impossible to explain anything as long as you believe that checking account dollars are "fiat money". They are backed by the assets of the issuing bank.”

            That’s the reason you won’t answer? I already said I would grant you your backing theory for this argument, I just didn’t say I would believe it.

            “Cash deposits are just as relevant as a "deposit" of borrowers' IOU's or government bonds. They are all assets, and they all back the money issued by the bank that holds those deposits”

            They are assets, but they don’t back the bank’s money. They affect the bank’s creditworthiness by enabling successful public history of redemption (or in some cases by enabling a favorable audit). But any assets beyond that purpose have no affect on the amount of basic money the bank will redeem for a particular IOU. If IOU x, all you will get for it on redemption is x, regardless of how minimally or incredibly profitable I am. But this is besides the point. I wasn’t arguing your backing theory, only your particular banking example.

            “No sensible banker or credit card company would knowingly issue a checking account dollar (or paper dollar, or credit card dollar) without getting at least a dollar's worth of assets (backing) in exchange. That is not to say that bankers are always sensible, or that their assets don't ever fall in value.”

            “Dollar’s worth” to whom?

            “Sometimes banks issue credit cards, and sometimes they are issued by credit card companies.”

            I’m not aware of any credit card issuers who aren’t banks, but I will happily grant you this point. Back to your banking example…

            Perhaps I’ve been too abstract. Tell me why the following example will not work:

            Bill brings a business plan to Lars Bank (an already established reputable bank, by whatever means you like). The plan is to buy up $1 trillion of NY real estate and generate profits from the rents. Bill wants Lars to loan him $1 trillion for this purpose. Lars (who doesn’t accept cash deposits) creates a $1 trillion checking account and trades it to Bill in exchange for Bill’s $1 trillion + 1$ IOU. Bill then buys $1 trillion of NY real estate. Bill, however, cannot make the principle payments back to Lars (perhaps because he only was able to purchase 50% of the properties he hoped as he drove prices up). Bill defaults, and Bill’s only creditor--Lars--gets the $1 trillion in real estate.

            Lars bank thus goes from minimal assets to minimal plus $1 trillion (or after prices fall post buying spree, $500 billion). Lars repeats this process over and over, acquiring many trillions of dollars in assets.

            It is easy to see why under textbook banking theory, this doesn’t happen (must have cash deposits to loan). How does *your* theory explain why it does not happen?

          • avatar Mike Sproul says:

            "If IOU x, all you will get for it on redemption is x, regardless of how minimally or incredibly profitable I am."

            Of course, because money is a fixed claim against the issuer's assets (like a bond), and not an equity claim (like stock). So money is backed by the issuer's assets just as bonds are backed by the issuer's assets, but the value of money doesn't have to change with every change in the issuer's assets.

            "Bill brings a business plan to Lars Bank..."

            For starters, nobody lends on such slender collateral. Everyone involved must have sufficient assurance against losses before they will do business.

            But let's take a realistic case. In the American colonies, the governor would lend 50 newly-printed paper shillings to a farmer who hands over his IOU for 50 shillings + interest, collateralized by 100 shillings worth of land. As more loans are made, more paper shillings are added to the liability side of the governor's balance sheet, but at the same time, an equal value of IOU's are added to the asset side of the balance sheet, so 1 paper shilling remains worth 1 silver shilling as more paper shillings are issued. Once the quantity of paper shillings is enough to satisfy the public's need for a medium of exchange, paper shillings start to pile up in private hands. The next farmer who wants a loan finds that he can borrow them from his over-stocked neighbors. The governor finds that nobody wants to borrow his newly printed shillings anymore, so the supply of paper shillings stops growing.

          • avatar vikingvista says:

            "For starters, nobody lends on such slender collateral."

            First, that simply is not true. Second, if they did require collateral, there must be a reason for it. With standard banking theory, the reason is clear--successful lending temporarily depletes the bank's loanable funds, and default wipes those funds out. But with your theory, loanable funds are as easy as printing up IOUs and trading them for someone else's IOUs. Cash is unnecessary for loanable funds in your theory--only IOUs are required.

            "Everyone involved must have sufficient assurance against losses before they will do business."

            It is entirely about risk management, and collateral may or may not play a part in that. But since your theory posits that the bank's printed IOUs are as good as cash in the marketplace whether or not they represent cash deposits, the risk for the bank in loaning the $1 trillion checking account dollars is next to nil. Those checking account dollars will buy at least a fraction of $1 trillion in hard assets that the bank did not have before. How can the Lars bank possibly lose?

  2. avatar Rob Rawlings says:

    I would like to seek clarification on the statement:

    "In fact, no individual bank involved can lend more than a part--usually the greater part--(the "excess reserves") of whatever fresh funds it is able to secure."

    If I understand the BoE paper correctly it is describing a system where the "fresh funds it is able to secure" may sometime actually be created by CB action after a loan had already been issued. The CB will do this if the bank can't get the funds in the interbank market without causing the interest rate that holds there to rise above target.

    While this does not mean that banks "create money of thin air" as Corrigan claims, it does mean that banks in a system like the one the BoE operates do not actually need to secure fresh funds before they can look for new customers to lend to.

    Is my understanding correct ?

    • avatar George Selgin says:

      Yes, Rob: today it is of course possible for a bank that finds itself short of funds for settlement (as it might were it to try to lend more than its available excess reserves) to secure such from a central bank. But it doesn't follow that banks generally either can or want to rely on such assistance, which may be refused in some circumstances, and to which a stigma often attaches. (In the U.S. case it certainly does--banks go to the discount window only as a last resort.) It would therefore be quite wrong to suggest from this that, even in a modern central-bank based system, banks generally need not worry about lending beyond their ability to attract funds from private savers.

  3. avatar vikingvista says:

    From the BoE link above...

    ON BANKING:

    "The principal way in which they are created is through commercial banks making loans: whenever a bank makes a loan, it creates a deposit in the borrower’s bank account, thereby creating new money."

    Is the ability to create that deposit at all influenced by whether or not that particular bank possesses, say from other deposits, that amount of money to loan? Well...

    "As ‘Money creation in the modern economy’ explains, though, banks cannot create money in this way without limit: how much banks lend will rest on the profitable lending opportunities available to them"

    ...apparently not. Commercial banks don't really need depositors' funds at all. All they need are profitable lending opportunities. But what incentive would banks have for seeking profitable lending opportunities if funds availability is not a constraint on them? I guess I'll have to read MCitME to find out.

    Perhaps Rothbard is right and the textbook money multiplier is just a silly fiction taught by economists who don't understand real-life banking.

    Can't we just get some real-life bankers or bank examiners to tell us if it is customary, in their experience, for a commercial bank's books to show that the bank's loans exceed its deposits from external sources? I mean, is "[the bank] creates a deposit in the borrower’s bank account" a shortcut for "the bank loans its funds to the borrower who immediately deposits them back with the bank", or isn't it?

    ON MONEY:

    "the article explains that money today is a type of IOU, but one that is special because everyone in the economy trusts that it will be accepted by other people in exchange for goods and services. It is because money is a form of IOU that banknotes still have the ‘promise to pay’ inscription: but money today is fiat or ‘paper’ money that is not convertible to any other asset (such as gold or other commodities)."

    So...money "is a type of IOU" in which IOU not a damn thing? Or is it an IOU because people accept it for goods and services, in the same way that gold coins of yore, or the labor I will produce for my employer tomorrow, are both IOUs?

    Thanks BoE. Your keen understanding of money and banking really clears things up.

  4. avatar RalphMusgrave says:

    George, I wouldn't bother even reading articles by Sean Corrigan. I've read several articles by him. They are invariably long-winded, verbose, pompous, rubbish.

  5. avatar Paul Marks says:

    If banks do not expand credit ("broad money") beyond real savings of cash-money then want "money supply" contracted between 1929 and 1933? And so many other busts going back to 1819 and before?

    Did evil "deflation elves" sneak into bank vaults between 1929 and 1933 burning Dollar bills and melting down coins? No - it was bank credit ("broad money") that contracted - because bank credit had indeed expanded (as it so often does) beyond real savings of cash-money, and it can do that even without a Central Bank (such as the Federal Reserve) although it is quite true that a Central Bank makes everything worse (makes the expansion of "broad money" worse - vastly worse). This is why Central Banking should be abolished.

    However, even without a government backed Central Bank (whether it is formally "privately owned" or not is not relevant - for example nationalising the Bank of England in 1946 made little practical difference) the ordinary commercial banking system (via its complex interactions) can expand lending (can expand credit "money") far beyond real savings of cash-money (although on a smaller scale than with the backing of a Central Bank) - it (the banking system) creates the false "boom" that must end in bust.

  6. avatar Keith Weiner says:

    George: I agree with you. I think this debate will go on endlessly until two ideas are broadly understood.

    First, the process of credit expansion requires a series of decisions by many actors. It cannot be taken for granted that a bank can (to use your reserve ratio example of 20%) just expand credit by 5X. Let's use the example of Andrew depositing 100oz gold. The bank lends 80oz to Barbara. She spends the money to hire Charlie to build her a workshop. Charlie deposits the 80 and the bank lends 64 to Diane. Diane spends it to buy some merchandise from Eric. And so on. Each must decide to participate. At no point can a bank take in a 100oz deposit and *just create*, ex nihilo, 500oz of checking deposits.

    Second, there is no money in our system today. Money is gold, nothing else (I would add silver, but not to quibble with JP Morgan!) There is no particular limit, either morally or economically, determining how much credit can be established on top of one ounce of gold. If there are opportunities to create value, and if people have a low time preference, lots and lots of lending will occur. This is not the old "100 claims to the same gold coin". It is a process of borrowing and spending and depositing and borrowing that can keep repeating, so long as all of the participants choose to let it go. Presumably, they will do this if and only if they think it's profitable to do so. In a centrally banked system, where the rule of law is disintegrating, maybe we could get to the point where the Fed could write a rule that says "If you're a bank named Morgan, Wells, BofA, or Citi, then you can print your own checking account balances as much as you want" But that has nothing to do with banking theory...

  7. avatar pkotak says:

    Would it be accurate to say that the banking industry has evolved with time? Commercial baking has become a big giant by interlinked with each other.

    The questions that I came across were the following:
     Banks have used the inter bank exchange for making themselves liquid to meet their lending needs and deposit requirements. There have been numerous reasons to believe these banks work with each other. As an example: recently JP Morgan Chase was heavily fined for its bad lending practices one of the largest fines (13 billion) payable by any firm in the history of USA. Yet the bank is solvent and opening new branched nationwide. They have also heavily invested in Wells Fargo Bank, as one of Wells Fargo’s Major Holders. Shouldn't one bank’s reputation reflect on the others?

     These Commercial banks borrowing funds from the Treasury. They use the line of credit or loan given from the Treasury to meet their lending requirements. Has these banks always used its own money to lend back? Isn't there an imbalance f their loans compared to the deposit ratio?

     The third and mostly used tool by most banks is to sell the loans to another lender or to the Wall Street. Bundling these loans into securities and selling as a package has been a trend in the lending industry.

    Their $100 in deposit sometimes grows to $400 - $800 in lending capacity if all the factors (Inter bank exchange, Treasury lending and selling their loans) are considered, than IOU is the big player here.

  8. avatar Martin Brock says:

    All of this talk about lending deposits baffles me as usual.

    You have a car, and I want a car, but I have no dollars or gold or silver anything similar. I have only my labor, but you find me credit worthy, so you accept my IOU for $10,000 in exchange for the car. Actually, you accept 1000 separate IOUs each promising ten dollars. I agree to exchange one IOU for ten dollars each day for the next 1000 days.

    In this scenario, you accept my 1000 IOUs practically as though each IOU were a $10 bill. If you will accept my IOUs this way, then your neighbors might accept them as well. If neighbors also accept my IOUs, then they're a general medium of exchange among you and these neighbors, i.e. they are money within this group.

    No bank creates the money in this scenario. I create it, and I don't create from it money deposited anywhere. I create it from my creditworthy labor. I create as much money as my creditworthiness supports.

    In reality, I might not succeed individually in circulating IOUs that circulate as money this way, but an institution certifying the creditworthiness of many individuals like me and pooling our default risk might succeed. This institution is a mutual credit bank of the sort that the original libertarian/anarchists in the nineteenth century advocated. State socialists, particularly Marxists in the Communist Manifesto, perverted this idea into a proposal for national banks, operated "for the working class", along the lines of the Fed.

    • avatar vikingvista says:

      "In reality, I might not succeed individually in circulating IOUs that circulate as money this way, but an institution certifying the creditworthiness of many individuals like me and pooling our default risk might succeed"

      It is easy to imagine all kinds of lending scenarios, however unlikely, particularly if an imposed central bank is thought to relieve lending institutions of market discipline. But a basic money free of any default risk will always be viewed differently than a money that entails such risk, with that differential varying by economic circumstances generally and specifically to the originator of the IOUs. In particular, in a competitive system, an institution will understandably feel more secure the less chance there is of a rival suddenly submitting an unexpected and unfulfillable number of its "IOU basic money" promises for redemption.

      Of course, 100%ers say the solution is to strictly limit redemption of bank IOUs in accordance with the redemption of the IOUs that the bank holds. However, even if that were possible, it can never obviate the need for reserve funds, given that default risk is always greater than zero. It is the nature of debt itself rather than demand deposits in particular that are the source of risk for banks.

      Then there is the problem of an institution emerging in the first place with widespread belief in its creditworthiness. This emergence will require a successful redemption history, and that will require long maintaining an adequate supply of irredeemable (basic) money. Basic money thus right from the start is an integral basis for acceptance of the bank's IOUs as money.

      There are then 2 issues, one empirical, and one theoretical. The empirical one is "How do commercial banks function in reality?" The answer, as any review of their books will reveal, is that they function as intermediaries of basic money lending. The theoretical question is then "Why do banks function thus?" The answer deals with institutional emergence and maintenance in a competitive regime.

      "State socialists, particularly Marxists in the Communist Manifesto, perverted this idea"

      Central banking itself is a central tenet of marxism. But it easy to see why the unwise power hungry, marxist or not, would advocate for it or for any other imposed monopoly under their control.

      • avatar Martin Brock says:

        A "basic money" like gold coin has no default risk, but it has plenty of risk. A coin's gold content could be misrepresented, or you could accept the coin from someone with a confederate waiting around the corner to rob you, or you could lose the coin through a hole in your pocket. The risk of an IOU defaulting need not be greater, particularly if an insurance pool exists.

        The standard of value promised by an IOU could be a standardized form of labor itself, and early libertarians advocated this standard. Furthermore, most money these days is electronic. Electronic money is always a record of entitlement to consume within some network. Even Bitcoin constitutes this sort of network.

        The "reserve" in a mutual credit bank is a reserve of the standard of value, but this reserve need not exist in a centralized bank vault. If labor is the standard, a "reserve" simply means that people exchanging the promise of labor in the future for goods consumed today must promise labor a bit more valuable than the goods they consume, so one when one of them defaults, the others make up the difference. Similarly, in a life insurance pool, the beneficiaries of a dead policy holder collect their benefit from premiums paid by people who do not die.

        If some people want a 100% reserve bank, they're welcome to use one, but I don't expect their warehouse receipts to be the most common medium of exchange without a state forbidding alternatives.

        "Intermediaries of basic money lending" doesn't seem to describe what banks do these days, since the "basic money" they lend are fiat currencies issued by a central bank either to regulate a fiat interest rate or to satisfy the whims of politicians. Sure, on its books, a bank "borrows money" from the central bank or another bank or the public before lending it, but this bookkeeping ignores the fact that money flows out of the central bank to meet the demands of creditworthy borrowers (ideally) and is not limited by any supply of more basic money.

        • avatar vikingvista says:

          “A "basic money" like gold coin has no default risk, but it has plenty of risk.”

          All forms of savings (deferred consumption) has risk, because the future is not completely knowable. But IOUs for basic money have all the risks of basic money, plus default risk.

          “A coin's gold content could be misrepresented,”

          An IOU can be fraudulent.

          “or you could accept the coin from someone with a confederate waiting around the corner”

          Some IOUs (such as traditional bank notes) suffer the same risk. Some IOUs (like a money order or coat check) reduce that risk by ensuring the basic money funds (or specific coat) are set aside, without reducing the security of their storage, but the IOU is no less at risk than the basic money it promises.

          “The risk of an IOU defaulting need not be greater, particularly if an insurance pool exists.”

          The risk of default is an added risk to that of basic money. You are right that pooling, as banks do, can be used to reduce the risk of bank loan defaults propagating into a bank defaulting on depositors. But insurance can also create moral hazard increasing risks. Insurance is best seen not as an overall risk reducer, but rather for insurance customers, a transformer of one kind of risk (<100% of unexpected loss) into another kind of risk (100% risk of loss to premiums). The pricing of that trade can either increase or decrease risky behavior. But otherwise the risk of a house burning down, or a bank loan failing, remains the same.

          “The standard of value promised by an IOU could be a standardized form of labor itself, and early libertarians advocated this standard.”

          Yeah, but I don’t see it working very well. Referring to “labor” is like referring to “food” or “people”. it is very heterogeneous, and one kind cannot be represented in terms of another kind. Although I realise such systems are used to a very limited extent around the world, they are doomed to be plagued by Gresham’s Law by at least an order of magnitude more than bimetallism. There is never anything standard about labor.

          “Furthermore, most money these days is electronic”

          Most money is electronic, but I don’t see how that constitutes a “furthermore”.

          “If some people want a 100% reserve bank, they're welcome to use one, but I don't expect their warehouse receipts to be the most common medium of exchange without a state forbidding alternatives.”

          I agree. Other than a handful of ideological Rothbardians, I can’t see much of a market for paid warehousing developing given the obvious benefits to depositors of putting their money to use in extremely low risk debt investments as happens with commercial banking. In particular, because there is nothing preventing such a system today other than the market. If a 100% reserve system were to rise to market dominance, it would have to be forcefully imposed, just like the ostensibly 100% reserve historical Bank of Amsterdam.

          “"Intermediaries of basic money lending" doesn't seem to describe what banks do these days, since the "basic money" they lend are fiat currencies issued by a central bank either to regulate a fiat interest rate or to satisfy the whims of politicians.”

          You’re characterization of fiat currency here is correct. And you’re right that the central bank unnecessarily maintains vestigial trappings of historical private banks and clearinghouses. However, central bank monopoly fiat money *is* the basic money of this country. The commercial banks function essentially as always--as intermediaries of basic money lending--only using irredeemable FedRes fiat money rather than irredeemable gold. Of course “essentially” excludes all the bailouts, crony regulation, distorting regulation, etc. The main problem with modern central banking is forced suppression of money competitors, and not the use of fiat basic money. The reason is that free competition would appropriately regulate a fiat money creation.

  9. avatar Mike Sproul says:

    Martin:
    You've got it right. The baffling thing is why people think that a bank can lend no more cash than it has received on deposit. The trouble is that people don't see that a $10 loan from a bank is made in exchange for a "deposit" of the borrower's $10 IOU. That's the only sense in which loans can't exceed deposits.

Leave a Reply

You must be logged in to post a comment. Login now.