The Meaning of Bank Deposits - Edwin Cannan (1921)

I HOPE I am not succumbing to the fashion of supposing a golden age in the past, but I cannot help thinking that the nature and functions of deposit banking were much better understood forty years ago than they are now. We had not then become convinced that nothing in economics can be both simple and true, and the young were taught that the theory of deposit banking was very simple. The banker was a man or a collection of men who undertook to keep money safely for its owners until they wanted it, and who made the business pay by lending out a good deal of this money to other people who wanted temporary loans. 

The Political Dictionary of I845 says, " People may deposit small sums of money at a bank, which the banker lends. Thus a bank is a means of facilitating the loan of money from the possessor of money to the farmer or manufacturer who has goods but wants ready money. The lending of money is the operation of banking, and a bank is a centre which facilitates this lending; it enables people to lend through a banker and his connection, who could not lend without that." William Ellis, in his bright little Outlines of Social Economy, I846, intended for school children, says: "The Banker . . . receives and takes care of the money of his customers with the understanding that he is to be prepared to pay on demand whatever they may call for." He asks what inducement the banker has to go to this trouble, and answers that of the money " part is employed at interest by the banker, and the interest thereby earned not only suffices to pay all the expenses of the establishment, but yields in addition a surplus profit sufficient to induce the banker to persevere in his business."

Mrs. Fawcett, in Political Economny for Beginners, 1870, and Jevons, in his Primer, Political Economy, 1878, say just the same.

The conception was a perfectly simple one, and I think it was and remains a perfectly true one. There is nothing really mysterious about the nature of banking " deposits." The term " deposit " seems very appropriate as the name of the verb which we use to describe the action of placing an article with some person or institution for safe custody. We " put things down " anywhere-our spectacle-case and our gloves, and often fail to find them again, and to " deposit " a thing is etymologically nothing more than to put it down; but the latinity of the word seems to give it a tinge of solemnity suggestive of the rites we go through when we entrust our bag to the cloakroom clerk instead of " putting it down " onthe platform.

With one exception we deposit things for safe custody with some person or institution in the full expectation of receiving again, when we come to claim it, the identical article which we deposited. We deposit our bag in the railway cloakroom on the distinct understanding that this bag and not merely an equally good bag will be restored to us when we demand it. True, if the railway company loses the bag owing to the inadvertence or dishonesty of its servant, it will tender compensation, and, the bag being irrecoverable, we shall have to accept fair compensation; but compensation impliesthat the contract has been broken: the contract was to restore the same bag and nothing else.

Moreover, with the same single exception, the things whendeposited may not be used by the person to whose custody they are entrusted. We should be seriously annoyed if we found the cloakroom attendant using our umbrella to keep himself dry in a shower of rain, and it would be wholly irregular for the wife of the chairman of a safe-deposit company to appear at a ball decked with the jewellery deposited by the Duchess of Blank. If the thing is to be used by the person to whom it is temporarily entrusted, it is not said to be deposited and to be " a deposit ": it is said to be " lent."

The one exception to both rules is money. Money is more homogeneous than bags and their contents. The substitution of one halfcrown for another will not affect us in the same way as the substitution of even our dearest friend's toothbrush for our own. Consequently, if we have deposited a half-crown, we are content to receive back another half-crown, or even " half-a-crown " in the different shape of two shillings and a sixpenny piece. No one, except some very small child, expects to receive the identical money which he deposited. Consequently persons and institutions receiving money on deposit have almost invariably mixed up the amounts received from various depositors. Deposit vour hat in the hotel cloakroom and the attendant will not expect you to be content to receive back the first hat he can lay hold of; but deposit your money in the hotel office and you will only expect to get back " the sum " for which you hold a receipt, and it will probably be paid to you in cash deposited by other depositors, or even in cash received in payment of guests' accounts. The homogeneity of money has always stood in the way of any objection being raised by the depositors of money to their deposits being used by the persons to whom they are entrusted. If you happen to meet the hotel ostler riding the bicyclewhich you deposited with him, you recognize it and complain ; but if in a shop you are given in change a ten-shilling note which you deposited in the hotel office a few hours before, you probably do not recognize it, and if you do you will not dream of going to the hotelkeeper and asking him why he presumed to spend your ten shillings: he did not undertake to keep that ten shillings for you, and he has another ten shilling note ready for you.

This explains why the depositor of money, unlike all other depositors, requires to pay nothing for the accommodation which he gets, but on the contrary nea'rly always receives something either in incidental services or in interest over and above the advantage of having his money kept safely for him. You will pay ninepence for depositing two bags and a rug for a day or two, but you can deposit a million pounds for a good deal less than nothing.

There is nothing in this one difference between money and other goods to suggest that the person with whom money is deposited can lend out more than he possesses in his own right plus what is deposited with him. The most abandoned cloakroom attendant cannot lend out more umbrellas or bicycles than have been entrusted to him, and the most reckless banker cannot lend out more money than he has of his own plus what he has of other people's. This is true even of a note-issuing banker: such a banker will no doubt lend his promises to pay on demand so long as there are people who will take them in exchange for goods and refrain from presenting them for payment; but these people are in reality making him a loan without interest. The extent to which he can borrow in this way limits the extent of his lending.

If it were not true that a banker's power to lend is limited by what he owns and can borrow, we should have the extraordinary result that a small bank with small deposits could lend as much a, a big one with many millions of deposits. Yet banks seem to regard it as of considerable importance to acquire depositors! The nineteenth century writers, taking it for granted that no one would suppose that a banker could lend more than he had got of his own and other people's, were in the habit of saying rather loosely that he could lend some proportion, such as two-thirds or threequarters of what he had obtained from depositors. But away back about 1730 Cantillon (in a passage cribbed like much else of his, by Postlethwayt's Dictionary before his Essai was published) had explained quite clearly that the banker had to forecast incomings in the shape of deposits and repayments of advances and set them against his forecast of outgoings in the shape of withdrawals of deposits and advances which he would not like to refuse, and thatdifferent bankers dealt with different classes, so that what was sufficient for one might be wholly insufficient for another-one might require to keep in hand about a tenth of his deposits and another would not be safe with less than a half or two-thirds. And anyone can see that the proportion would vary from time to time with the same banker as well as between banker and banker at the same time. If only a banker could arrange to make his incomings exactly correspond with his outgoings, he would obviously have no reason for keeping any stock or reserve at all.

It was never supposed by the simple-minded nineteenth century economists that anyone would make a difficulty about the aggregate of deposits (1) exceeding the aggregate bf cash held by the banks, or even (2) exceeding the aggregate of all the cash held by all persons and institutions, including the banks.

1. It was naturally supposed that a single banker could have a million of deposits and lend out, say, £750,000, that two bankers could have two millions and lend out £500,000, and the whole number of bankers together could have deposits equal to four times the amount of their cash in hand. No one saw any miracle in the aggregate of deposits being, say, a thousand millions when the cash held was only £250,000,000; this was looked on simply as another way of saying that the banks had lent out three-quarters of the money lent to (alias deposited with) them. No one supposed that they had " created" the £750,000,000. If cloakroom attendants managed to lend out exactly three-quarters of the bags entrusted to them, we should not be surprised to find that the number of bags on deposit was exactly four times the number in the cloakrooms: we certainly should not accuse the cloakroom attendants of having " created " the number of bags indicated by the excess of bags on deposit over bags in the cloakrooms.

2. Nor used any difficulty to be made if the aggregate of bank deposits was seen to exceed even the total of cash in existence. It is true that bags not being " currency," a means of payment or medium of exchange which passes easily from hand to hand, bags could only be lent on hire to borrowers who wish to use them personally, so that the number of bags on deposit would be less than the total in existence. But when the bags or other things deposited are currency, the situation is different. Borrowers in this case do not borrow with the intention of retaining the article borrowed till repayment, but with the intention, which they carry out immediately (simultaneously very often) of parting with it in exchange for other things. Consequently, though they owe the sum of money lent them, they do not hold currency to that amount. If you have borrowed a bag and not yet returned it to its owner, you probably have it still: if you have borrowed a thousand pounds it is most unlikely that so much will be found on your person or in your drawer at home. Thus the amount of the currency does not limit the amount of money which can be lent whether by the banks to customers (borrowers) or to the banks by customers (depositors). If the total of bank deposits is three times as great as the total of coin and notes in existence we need no more suppose that the banks have " created money " to the extent of double the coin and notes than we need suppose that because the National Debt is ten times the amount of all the coin and notes, the State has " created money " to the extent of nine times the coin and notes. No ordinary lender supposes he creates money by lending it; why should the banks ? Just as the amount of the State debt or the total of all individuals' debts is only the sum of what the State or the individuals owe, so the total of bank deposits is simply the sum of what the banks owe. In no case is there any reason for boggling over the fact that the totals greatly -exceed the currency in existence.

All this is much too simple for the present age. Instead of the old doctrine that capacity to lend is based on the possession of valuable property, and that banks accordingly can lend out of their own capital plus what solvent customers lend to them (alias deposit with them), we have journalists and popular writers and chairmen of large joint-stock banks persuading the public that banks have themselves created, or to use Mr. Hartley Withers' own word, "manufactured," thousands of millions of pounds by lending something which did not before exist to borrowers, who proceed to pay it to other people, who in their turn deposit it in the banks, and who could not have so deposited it unless the banks had lent.

This curious inversion seems to be partly due to the practice of considering how large a proportion of his deposits a banker can safely lend in the form of a question what ratio his advances. should bear to his cash. No doubt when you have a million of deposits to deal with, it comes to much the same thing whether you ask what ratio your cash should bear to your liabilities, or what ratio your advances should bear to your cash. But to compare the cash held by the banker with the amount lent by him without any reference to his aggregate command of money is very apt to be misleading. When, for instance, Mr. Withers remarks (Meaning of Money, 1918 issue, p. 35), " A banker who has £10,000 in gold or notes at his command would be running too great a banking risk if he advanced ten millions to the most unexceptionable customers," he may have meant that a banker who had £10,010,000 at his command would be very foolish to lend out as much as ten millions and keep only ten thousand in cash. But it is likely that someone among hisreaders will rub his eyes and say, " Wonderful thing it is to be a banker! Now I have got £10 in my pocket, and yet nobody warns me not to lend £1,000 on the strength of it. Prudence be blowed! my trouble is that I cannot lend a penny beyond my £10 because I haven't got it. Anyone who borrows from me will want to take the money, but these banker fellows seem able to find borrowers who don't. Withers says on the next page that the banker who lent the ten millions to the unexceptionable customers ' would give them the right to take out ten millions in gold and notes, and if even a thousandth part of the right were exercised, the banker's gold and notes would be all gone.' Somehow or other the money lent by the banker seems to stay in his possession, so that he can 'lend' ad lib. provided he isn't asked to lend in gold or notes."

The error of this inference clearly arises from leaving out of sight the fundamental fact that the banker is able to lend X, Y and Z more than his own capital because A, B and C are allowing him the temporary use of some of theirs on condition that he will let them have what thiey want of it when they ask for it. The "customers" of a bank include both lenders to the bank and borrowers from it, and though some of them are borrowers to-day and lenders to-morrow, there are at any single moment two distinct classes, between which the banker is the intermediary who arranges for the capital of the lenders being used by the borrowers.

I do not think Mr. Withers anywhere denies that where there are a number of banks the power of each bank to lend is limited by the extent of its "resources "-its power, that is, to command moneynor that he anywhere asserts that it can directly increase its resources by lending. If it could, every little bank would soon be a big one; but he does. seem to hold that all the banks in a particular country (what is a country?) taken together, and any real or supposititious single isolated bank increase their power of lending by lending. The real inspiring text for this doctrine seems to have been the saying among bankers, "Loans make deposits." Indirectly, no doubt, it is true that the lending of money by bankers tends to make deposits, because it is a useful service to the community. Road-making, and any other useful service may similarly be said to tend to make deposits. This seems to be all that the phrase "loans make deposits " originally meant. A nineteenth century banker, W. Haig Miller, remarks: " Bankers increase their deposits by lending money to individuals, who by their loans become wealthy and increase the resources of the district.J" (On the Bank's Threshold, 1890, p. 69.) But latterly the proposition, sometimes hardened into " every loan makes a deposit " (Meaning of Money, p. 63), has been taken to mean a good deal more than this..

In Mr. Withers' chapter V, on " The Manufacture of Money," the reader is asked to consider himself a " prudent person " who has borrowed £1,050 from his bank to pay for a new motor-car, and is assured that his " borrowing of £1,050 has increased the sum of banking deposits as a whole, by that amount."

If the borrower's £1,050 was lent him by his bank simultaneously with the repayment to the bank of £1,050 by some other borrower, the proposition would be indefensible on the face of it : if it were true that replacing one borrower by another increased deposits, the total would long ago have reached astronomical figures. Mr. Withers must mean us to suppose that the £1,050 was an addition to the loans already made by the bank. The theory thus is that every addition to the total, of loans by banks makes an equal addition to the total of their deposits: and if there is only one bank, every addition to the total of its loans makes an equal addition to its deposits, for Mr. Withers later in the chapter introduces thesupposition of an isolated district with a single bank which has, according to him, increased its deposits from £100,000 to £1,500,000 by lending £1,000,000 and investing (which is treated as a sort of lending) £400,000. The only other items in the balance sheet are capital £100,000, and cash in hand £200,000.

Of course, a balance sheet, as an expert in currency has observed, must balance; we must expect to find assets and liabilities growing up together at the same pace. The only question is, "Did the amounts on each side rise, as the nineteenth and previous centuries believed, because monied persons had the power and the will to lend to (or 'deposit with') the bank, as time went on, more and more money, or, as Mr. Withers teaches, because, as time went on, the bank chose to lend more and more? "

The only reason Mr. Withers gives for throwing over the old view that it is the action of the depositors in depositing which enables the bank to lend, and adopting the new view that it is the action of the bank in lending which enables the depositors to deposit, is that the isolated locality " could not have deposited £1,500,000 without advances from the bank, because there never was such a sum in the place." Presumably " there never was such a sum in the place " means " there never was £1,500,000 in coin (or possibly Bradburys) in existence in the place at one and the same moment." But what difficulty does this fact present ? No one supposes that the depositors paid in £1,500,000 at one and the same moment. Their £1,500,000 was got together by small surpluses of amounts paid in over amounts withdrawn, spread over a long period. If they paid in an aggregate of 1,326 sovereigns per business day and withdrew only 1,000, they would accumulate £1,500,000 to their credit in fifteen years, and the bank by keeping twenty-two of the sovereigns per day would add  to its cash, and by paying out the other 304 to borrowers would add £1,400,000 to its advances in the same period. The depositors have deposited £1,500,000 more than they have withdrawn, and it is difficult to make any seilse at all of Mr. Withers' proposition that they have " presumably deposited £100,00, since the bank holds £200,000 in cash, of which £100, 000 may be taken as having been contributed by the subscribers of its capital." The depositors have a well-founded belief that the whole of the deposits have been deposited by them: are not they the depositors ? but Mr. Withers tells them gently but firmly that they are quite mistaken; they have only deposited one-fifteenth and "the rest of the deposits have been provided by the bank itself." "The broad conclusion," he says a few pages further on (p. 72), " is that banking deposits come into being to a small extent by cash paid into banks across the counter, to a larger, but still comparatively small extent, by purclases of securities by the banks which create book credits, and chiefly by means of loans from the banks which also create book credits."

It seems incredible that anyone should imagine that depositors cannot have paid in the past on balance more cash into their bank or banks than the amount of cash which the bank or banks possess at the present moment, but Mr. Withers does not stand alone. Mr. McKenna, in his speech to the London Joint City and Midland Bank shareholders in January, 1920, which was widely applauded, took just the same line. " In June, 1914," he said, ' the banks held £75,000,000 of currency, Last month this figure stood at £191,000,000. The banks, therefore, held more currency to the amount of £116,000,000 and to this extent the increase in the aggregate of bank deposits is accounted for by paymentsin of currency." The rest of the increase in the deposits, amounting to about £1,114,000,000, he attributed to " bank loans." We might have expected that the example of the Savings Bank would be sufficient to warn Mr. McKenna off the strange assumption that the amount of cash held by banks shows how much of their total deposits is accounted for by payments in of currency. It could scarcely be contended;that any but a small proportion of the nineteenth century Savings Bank-deposits was not " accounted for by payments in of currency," but on December 31St, 1899, the Post Office SavingsBank deposits amounted to £130,000,000, while the cash held against them was too trifling a sum to appear as a separate item in the Post Office accounts. In the previous nine years the total of deposits had risen by £62,500,000, of which £20,500,000 was accounted for by interest credited to the accounts by the Bank, and the remaining £42,000,000 was " accounted for" by the fact that the payments in by depositors, which must have been nearly all in currency handed across the counter, amounted to £280,000,000, while the withdrawals were only £238,000,000. Will anyone say that either the sixty millions or the forty millions were " created by the Savings Bank," or that they were " provided by the Bank itself," or that they should be " attributed to bank loans ? "

It would seem well to return to the nineteenth century doctrine that banks receive money from one set of people and lend it to another: that the total of this money at any moment is a total of the same nature as the total of the money lent on mortgage of property: that it is just as wrong to regard it as a kind of fictitious cash "created by banks " as it would be to regard the money out on mortgage as a kind of fictitious cash created by solicitors, and little, if any, less wrong than to regard it as a mass of gold. We should also revive the doctrine that deposits tend to increase when people become more numerous and richer, and that given a certain population and material welfare, they tend to vary with variations in opinion about the comparative desirability of direct individual investment and indirect investment through the medium of banks (in more familiar language, opinion about the comparative advantage of " putting your money in the bank " and " putting it in business or stocks and shares "). Recent experience suggests one addition which the nineteenth century never required to think of. This is that the total of deposits tends to increase with a diminution in the purchasing power of the unit of currency in which they are reckoned (and of course vice versa it tends to diminish with a rise in the purchasing power of the unit). If the pound sterling will buy less, people of the same wealth and people dealing in the same amount of goods as before will require, and be able to have, a larger number of pounds at their banks than before. Hence the enormous rise of deposits in this country since the beginning of the War, and the much greater rise in countries in which the unit of account is still more depreciated.