What is Money?
The Fortnightly Review, January-June, 1879
 Professor Jevons, in his excellent little book on "Money," tells us that the ingenious attempts that have been made to define money "involve the logical blunder of supposing that we may, by settling the meaning of a single word, avoid all the complex differences and various conditions of many things, requiring each its own definition." Without denying that this blunder has been sometimes committed, I think it misleading to suggest, as Mr. Jevons does, that the attempt to define a class-name necessarily implies a neglect of the specific differences of the things contained in the class. Indeed, when he goes on to say that the many things which are or may be called money - " bullion, standard coin, token coin, convertible and inconvertible notes, legal tender and not legal tender, cheques of various kinds, mercantile bills, exchequer bills, stock certificates, &c." - "require each its own definition," he apparently maintains the rather paradoxical position that it is logically correct to give definitions of a number of species, but logically erroneous to try to define their common genus. It is easy to show that several at least of these more special notions present just the same sort of difficulties when we attempt to determine them precisely as the wider notion "money " does. For instance, the distinction between bullion and coin seems at first sight plain enough ; but when we ask under which head we are to classify gold pieces circulating at the their market value in a country that has a single silver standard, we see that it is not after all so easy to define coin. The characteristic of being materially coined-that is, cut and stamped by authority - though it has always been combined in our own experience with the characteristic of being legal tender, is capable of being separated from it; so that we have to choose between the two in our definition. So again, we may ask, what makes a coin a token ? Does a seignorage sufficient to cover the expense of coining have this effect ? If not, why not ? and what further difference is required between the value of coin and the value of the metal contained in it ? Similarly, we may inquire whether by calling notes convertible it is merely meant that their issuer has promised to convert them into coin on demand, or whether a belief is affirmed that he would so convert them if required ? If the latter alternative be chosen, it must be evident that the legitimacy of such a belief must depend upon the nature and extent of the provisions made by the issuer for meeting demands of coin ; so that in order to define convertibility precisely we shall have to determine what provisions are adequate, and whether all possible demands should be provided for or only such as may reasonably be expected. Then further, how shall we  treat the case - which used to be common in the United States (1) - of notes for which coin will almost certainly be paid if demanded, but not without a serious loss of good-will to the demander? In short, we cannot escape the proverbial difficulties of drawing a line, if we attempt to use any economic terms with precision; and instead of seeing in these difficulties - as Professor Jevons seems to do - a ground for not making the attempt, I venture to take an exactly opposite view of them. I think that there is no method so convenient for bringing before the mind the "complex differences and various conditions" of the matters that it is occupied in studying, as just this effort to define general terms. The gain derived from this process (as I have urged in a previous paper (2)) is quite independent of its success. We may find that the reasons for drawing any proposed line between money and things rather like money are balanced and indecisive. But since such reasons must consist in statements of the important resemblances and differences of the things that we are trying to classify, the knowledge of them must be useful in economic reasoning, whatever definition we may ultimately adopt.
Let me then raise once more the vexed question - What is money? But first, we must observe that when proposed in this form the problem is fundamentally ambiguous; as it blends the two quite distinct questions, (1) What do we call money ? and (2) What ought we to call money? I am inclined to think that the "intellectual vertigo," which has been said to attack all writers who approached this fatal theme," may be partly traced to the want of a clear separation between these two very different issues, and the different methods of discussion respectively appropriate to each. The first point has obviously to be settled entirely by reference to the current use of language. In fact it is not strictly an economic question at all but a linguistic one ; only it is a linguistic question which it requires a certain amount of economic knowledge to answer satisfactorily. For though we have all of us something to do with money, most of us are even painfully conscious that our acquaintance with it is very limited. We commonly recognise that there are certain classes of persons, bankers, merchants, writers of city articles, &c., who are especially occupied in considering and discussing money and its relations from a practical point of view. Hence it is their use of the term which we shall naturally begin by investigating. If there is any one who ought to know what is meant by money, they ought to know.
At the very outset of our inquiry a curious phenomenon presents itself. There seems to be a tolerable accord among our monetary experts (3) in England, at the present time, as to the answer that ought to  be given to the question What is money? when they directly attempt to answer it; at any rate, the extent to which they differ is inconsiderable in comparison with the extent of their agreement. Unfortunately the answer so given is in palpable discrepancy with their customary use of the term when they are not trying to define it; and this discrepancy is not of a minor kind, but as fundamental as can well be conceived. When the question is expressly raised they have no doubt that by money they mean what they also call currency, that is, coin and bank-notes. They see the need of distinguishing the latter as paper money or paper currency; and they recognise the existence of a narrower definition which restricts the term money to coined metal, on the view that bank-notes are mere promises, to pay money, which ought not to be confounded with money, however currently they may be taken for it. But they are disposed to reject this view as a heresy; and though the narrower sense is that adopted by several economists of repute, I imagine that it would be regarded as at least old-fashioned by practical men; except so far as the word is quite technically employed in relation to the details of banking business. Again, our authorities allow that there is a certain resemblance between bank-notes and bills of exchange, letters of credit, promissory notes issued by private persons, &c.; but though they may perhaps regard these latter as constituting an "auxiliary currency," they do not consider them to be currency in the strictest sense, and therefore do not call them money. The only important point on which their utterances are doubtful or conflicting is the question whether notes issued by private banks and not made legal tender should be considered as money; the importance of this question, however, so far as England is concerned, is continually diminishing. But when bankers and merchants, or those who write for them, are talking of "money" in the sense in which, generally speaking, they are most practically concerned with it - of money which (or, more strictly, the temporary use of which) is continually valued and bought and sold in the money market, which is sometimes scarce and dear and at other times cheap and plentiful - they speak of something which must be defined quite differently. For though coin and bank-notes form a specially important part of money-market money, they certainly cannot constitute the whole of it in any country where deposit-banking is fully developed and payment by cheques customary; and in England, at present, they do not constitute even the greater part of it.
What has just been said will appear to some of my readers a truism. But there are probably more to whom it will appear a paradox; and for the sake of these latter it will be well to pause and illustrate pretty fully this use of the term money. For this purpose I shall take Bagehot's Lombard Street as my authority. I do this not merely on account of the marked popularity of this little  book, which is now in its sixth edition ; but because Bagehot united practical and theoretical qualifications for dealing with this subject, such as have rarely been combined in any single man. He was himself a banker; he was, as editor of the Economist in the habit of writing for bankers and merchants, so that "he that went by rail" might read; while, at the same time he was a master of abstract economic theory, thoroughly acquainted with the criticisms that theorists have passed on the common language and ways of thinking of dealers in money. Hence we may be sure that his sense of the term money is deliberately chosen ; not, perhaps, as the sense he would have adopted if he had assumed the linguistic liberty of a purely theoretical writer, but, at any rate, as a sense which he found so fixed in the ordinary thought and discourse of his readers as to render it inexpedient for him to try and modify it.
What, then, is the money of Lombard Street, the possession of which makes England "the greatest moneyed country in the world" ? The answer is very simple. It is a commodity of which the greater part exists only in the shadowy form of what is sometimes called bankers' credit, but may be more definitely conceived as bankers' obligations to pay money on demand ; such credit or obligations being not even embodied in bank-notes. It is true that Bagehot never says that he means this by money, but there are many passages in which it is clear that he can mean nothing else. Take, for example, the following : -
"Every one is aware that England is the greatest moneyed country in the world; every one admits that it has much more immediately disposable and ready cash than any other country. But very few persons are aware how much greater the ready balance - the floating loan-fund, which can be lent to any one for any purpose - is in England than it is anywhere else in the world. A very few figures will show how large the London loan-fund is, and how much greater it is than any other. The known deposits - the deposits of banks which publish their accounts - are, in
London (31st December, 1872)........................................................................ £120,000,000
Paris (27th February, 1873)................................................................................. 13,000,000
New York (February 1873) ....................................................................................... 40,000,000
German Empire (31st January, 1873)............................................................. 8,000,000
And the unknown deposits - the deposits in banks which do not publish their accounts - are in London much greater than those in any other of these Cities. The bankers' deposits of London are many times greater than those of any other city - those of Great Britain many times greater than those of any other country."
Here Bagehot clearly regards these bankers' deposits as "immediately disposable and ready cash," or, as he afterwards calls it, "money-market money." If, then, we ask ourselves where and in what form this money exists, it must be evident that, at any given time, most of it exists only in the form of liabilities or obligations, acknowledged by rows of figures in the bankers' books; and that it is transferred from owner to owner, and thus fulfils all the functions of a medium of exchange, without ever changing its form.  Most of us, no doubt, have had a vague impression that these figures in bankers' books "represent" sovereigns or bank-notes ; which, though they are not actually in the banker's possession, have yet passed through his hands, and exist somewhere in the commercial world. But I need hardly say to any one who has read Lombard Street that this cannot be Bagehot's view; since the main drift of that book is to bring prominently forward the fact that, in consequence of the "one-reserve system" upon which English banking is constructed, but little of this immense "loan-fund which can be lent to any one" could possibly be presented in the shape of coin or bank-notes. Of course some portion of the money lent by London bankers is continually taken from them in this shape. But a little reflection on the mode in which it is borrowed and used will show how comparatively small this portion is. Such loans are chiefly made to traders, either directly by the bankers or through the agency of the bill-brokers; and when a trader borrows from his bank, he almost always does so by having the loan placed to his credit in his banker's books, and drawing against it by cheques ; and the effect of such cheques, for the most part, is not to cause the money to be produced in the form of coin or notes, but merely to transfer it to some other person's account at the same or some other bank. The bank-notes and gold are merely the small change of such loans; and it is only when money is lent to manufacturers and farmers, who have large sums to pay in wages, that the amount of this change bears even a considerable proportion to the whole loan. It may seem that when cheques on one bank are paid into another, material money must pass between bank and bank. But by the system of the Clearing House the mutual claims of the different banks are set of against each other; so that, even when the balance daily due from each bank to others was paid in notes, the amount of these required was very small in proportion to the amount of liabilities transferred; and now no notes are commonly needed at all, as such balances are paid by drafts on the Bank of England, where the other banks keep the main part of their reserves.
But we may reach the same result more briefly by means of a few statistics, which I may conveniently take from Mr. Inglis Palgrave, whose Notes on Banking were published almost contemporaneously with Bagehot's book. Mr. Palgrave estimates the whole amount of deposits held by English, Scotch, and Irish banks (exclusive of the discount-houses) on the 12th of March, 1873, at about 486 millions the liabilities of the London banks alone being about 179 millions while he estimates the metallic circulation of the whole kingdom in 1872 at about 105 millions, and the note circulation at 43 millions. If we consider that more than 10 millions of notes and coin,on the average, were kept as reserve by the Bank of England, and that the provincial banks require a considerably larger proportion  of coin for their daily business than the London banks, we shall require no elaborate proof to convince us that the greater part of the "unequalled loan-fund" of Lombard Street can never emerge from the immaterial condition of bankers' liabilities.
The difficulty, indeed, is not to prove this, but rather to explain why this obvious truth is overlooked, or even implicitly denied in so much of what is said and written about money (4). Even Bagehot frequently uses language which suggests that what banks receive on deposit and lend to traders and manufacturers is entirely legal tender, coin or bank-notes. Take, for example, a passage (p. 142) where he is arguing that the rise in general prices in 1871 was due partly to "cheap money" :
"It might be said at first sight that so general an increase must be due to a depreciation of the precious metals . . . . . . And, indeed, there plainly is a diminution in the purchasing power of money, though that diminution is not general and permanent, but local and temporary. The peculiarity of the precious metals is that their value depends for unusually long periods on the quantity of them which is in the market. In the long run, their value, like that of all others, is determined by the cost with which they can be brought to market. But for all temporary purposes, it is the supply in the market which governs the price, and that supply in this country is exceedingly variable."
One cause of this variation, he goes on to explain, which operates during the depressed period that follows a commercial crisis, is that the
"savings of the country increase considerably faster than the outlet for them. A person who has made savings does not know what to do with them. And this new unemployed saying means additional money. Till a saving is invested or employed it exists only in the form of money: a farmer who has sold his wheat and has £100 'to the good,' holds that £100 in money, or some equivalent for money, till he sees some advantageous use to be made of it. Probably he places it in a bank, and this enables it to do more work. If £3,000,000 of coin be deposited in a bank, and it need only keep £1,000,000 as a reserve, that sets £2,000,000 free, and is for the time equivalent to an increase of so much coin."
This passage certainly suggests that "saving," as actually performed in England at present, consists either in depositing coin with a banker, or at least in doing something which has the same effect as depositing coin ; and that the business of a banker normally consists in lending about two-thirds of the coin thus paid in to posits him. But it must be evident that what each of us chiefly deposits when he saves is represented by cheques, dividend warrant, &c.; and that by these he merely transfers to his own banker the obligations towards himself that other bankers have incurred, together with the right of collecting corresponding sums of money from these other bankers. And the total effect of this process on the aggregate of banks cannot possibly be to increase directly the  amount of "loanable money;" it can only influence this indirectly, so far as the saving diminishes the amount of legal tender which the bankers' customers require for their expenditure from day to day, and by thus increasing the proportion which the bankers' notes and coin bear to their liabilities, induces them to extend the latter. No doubt, to a certain extent, increased saving leads to a direct transfer of coin and bank-notes from the circulation to the bankers' cash-boxes; and this is especially the case with provincial saving, which is largely performed by farmers and retail tradesmen. Still, the phenomenon of cheap money in Lombard Street, for which Bagehot is trying to account, should be explained by reference rather to a decrease in the demand for loans than to any positive increase in the supply of loanable money directly caused by the process of saving.
I need hardly say that I do not attribute to such writers as Bagehot any settled misapprehension of the real nature of what they call money. But I think that their language is apt to mislead persons less familiar with the facts; and that it further has some tendency to confuse their own reasonings. Thus even Bagehot seems hardly aware that he uses the phrase "ready cash" in two different significations. In one sense, as we saw, England is said to have "much more ready cash" than any other country. That is, she has much more of the immaterial money which exists in the form of bankers' obligations to render material money if required. For of this latter, as he goes on to explain, England has comparatively little; the amount of our "cash in hand" (in this sense) "is so exceedingly small that a bystander almost trembles at its minuteness compared with the immensity of the credit that rests upon it." The truth is that the same thing presents itself to him in the opposite characters of credit and cash, according to the point of view from which he regards it. When he is considering possible crises and collapses of credit, the difference between bankers' liabilities and their means of meeting them becomes only too palpable; so that the latter, as "cash in hand" is naturally contrasted with the former. But in ordinary times "book-money," as I have called it, is generally preferred as a medium of exchange to gold or bank-notes, involving as it does not only less trouble but less apparent risk; since a man is more afraid of having his gold or notes stolen, than he is of his banker breaking. Since, then, each depositor is aware that he only leaves his money in its immaterial condition for his own convenience, and that lie can convert any portion of his bankers' liabilities into gold or notes at will, he naturally comes to conceive the former as "ready cash" no less than the latter.
In this way we may partly explain the paradox which I noticed at starting, that money is expressly defined by most of those who write about it in a manner which implicitly excludes the greater part  of the medium of exchange, which (as they are aware) is commonly lent and borrowed under the name of money in England. But a further explanation may be found in the view which influential economists have taken of the current commercial use of the terms money and the value of money. This view contains, or at least suggests, an important element of truth ; but the statement of it, which I find (e.g.) in Mill's treatise, seems to me seriously misleading, and indeed calculated to shroud the whole matter in an impenetrable fog of confused thought.
Mill begins his chapter on the Value of Money (5) by "clearing from our path a formidable ambiguity of language," by which, as he explains, money is commonly confounded with capital.
"When one person lends to another," he says, "what he really lends is so much capital; the money is the mere instrument of the transfer. But the capital usually passes from the lender to the receiver through the means either of money, or of an order to receive money, and at any rate it is in money that the capital is computed and estimated. Hence, borrowing capital is universally called borrowing money; the loan market is called the money market .....and the equivalent given for the use of capital, or, in other words, interest, is not only called the interest of money, but, by a grosser perversion of terms, the value of money."
Now, I do not deny that there is an ambiguity in the phrase, value of money; but there seems to me a second equally serious ambiguity in the language that Mill uses in exposing the first. No doubt, when the value of money is mentioned in Lombard Street, it is not the purchasing power of money, measured in commodities, that is intended; but neither is it exactly the rate of interest, as Mill elsewhere uses this phrase, i.e. the average annual return to capital, subtracting insurance for risk and wages of management. It is, in fact, the value of the temporary use, not of capital generally, but of money in particular; estimated, as other values are commonly estimated, in terms of money. It is quite true that people often speak of the interest paid for the use of capital in other forms, as the "interest of their money;" but they are aware that it is money invested, and I do not think they really confound this with ready money. They must know that the interest of invested money, or capital generally, may vary comparatively little, while the price paid for the use of ready money is fluctuating through all the stages of a financial crisis; rising perhaps as high as ten per cent., and then falling as low as two per cent. But even if we admit that what is called interest of money should often be rather termed interest of capital, it is still misleading to say that by borrowing money we really mean borrowing capital; since, as Mill elsewhere observes, "loanable capital is all of it in the form of money," and therefore the antithesis is obviously inappropriate.
 The truth is that Mill, in his account of Money and Credit, is rather too much influenced by his desire to guard against two errors. In the first place he wishes to dispel completely the illusory assumption, which he regards as the basis of the old Mercantile System, that "wealth consists solely of money." Whether this illusion ever did really "overmaster the mind of every politician in Europe," I do not now inquire; perhaps we inevitably exaggerate the errors of our predecessors in the intellectual struggle that rids us of them; at any rate the doctrine is now happily defunct. But there is a more subtle confusion of the same kind which Mill also effectively exposes; the tendency to infer increase of wealth from what is merely increase of money, and to imagine that we are all better off when we have only got "more counters to reckon with." In the laudable effort to crush this fallacy, Mill is occasionally led to depreciate unduly the importance of money, and to speak of it as if it were not really "wealth" or "capital" at all; intermittently forgetting that "money being the instrument of an important public and private purpose is rightly regarded as wealth," and, since it is indispensable to the most effective production, as capital also. Thus, he tells us that "there cannot be a more insignificant thing than money, except in its character as a contrivance for sparing time and labour." It is not so much what is here said that is misleading, as the tone in which it is said. We might with equal truth affirm that there cannot be a more insignificant thing than a steam-engine, except in its character as a contrivance for generating and applying steam-power. But Mill's sentence certainly suggests that money is something that we could easily do without; whereas his real aim is not to depreciate the economic importance of the function of money, but merely to urge that this function will be no better fulfilled by a larger amount than by a smaller; provided that our habits and customs of distribution and exchange are duly adapted to the smaller amount. Similarly, the statement that "when one person lends to another, what he transfers is not the mere money, but a right to a certain value of the produce of the country, to be selected at pleasure," is unsatisfactory, though it is in a sense incontrovertible. A man only borrows money in order to buy something else, or to pay for something already bought; but what he actually borrows is money, and it is essentially inexact to represent him as borrowing anything else. The bad effects of this inexactness are, indeed, latent so long as we are dealing with metallic money; for when commodities are bought and sold for hard coin, it is impossible to ignore the fact that they are transferred by means of an instrument which is equal in value to the wealth that it is used to transfer. But when bankers' credit is the medium of exchange, it is easier to let this fact drop out of sight; and Mill continually does so. Thus he speaks contemptuously of an "extension of credit being talked of . . . . as if credit actually were capital," whereas it is only "permission to use the capital of another person." Now, in a certain rather strained way, we might say this of gold coin; its function is to "permit" or enable its owner to obtain and use other wealth. And it is only in this sense that Mill's statement is true of the credit or liabilities which a banker lends to his customers, whether in the form of notes, or under the rather misleading name of "deposits." This credit, no doubt, is a comparatively fragile and perishable instrument for transferring wealth; like the magic money of the Arabian Nights, it is liable to be turned by a financial crisis into worthless leaves of written or printed paper : still, so long as it is commonly accepted in final settlement of debts, it has not only precisely the same function as gold coin, but also precisely the same market value; viz. the interest or discount that is paid for the use of it. And probably those who have "talked of credit as capital" have never meant to imply more than this.
At the same time, as I said, there is an element of truth suggested by the statement that "borrowing money is really borrowing [other] capital;" and it is an element all the more important to bring out, as it is inevitably ignored in most of what is written about the money market. It is true that the existence of these vast amounts of bankers' credit depends on the concomitant existence of correspending amounts of wealth of other kinds, which are transferred by means of it. While it is needful for clearness of thought to insist that the "loanable capital" of Lombard Street is merely money - in the wider sense above explained,- and most of it immaterial money, it is no less necessary for completeness of view to bear in mind that this immaterial money is only kept in being by the continued exercise of its function. There would not actually be these millions of it in London, if London were not the greatest emporium in the world, and therefore a centre through which many more millions' worth of commodities is continually passing (6).
Let us now sum up briefly the different uses of "money" which we have found to be more or less current. We may conveniently arrange them in order, according to their width of meaning. First will come the narrowest use, which is also the earliest, to denote coined metal. Next there is the sense universally recognised in the definitions now given by monetary experts, which includes besides coin such "paper money" as is "legal tender." Thirdly, there is the wider use which such experts sometimes, though not most commonly admit, in which bank-notes that are not legal tender are taken in. Fourthly, there is the still wider signification, which wehave found to be current in the language of Lombard Street, though it is not often expressly recognised, according to which bankers' liabilities not represented by notes constitute the larger part of the so-called money. All these four - metallic money, paper money of both kinds, and "book money" - have the same exchange value, are lent and borrowed for the same interest or discount, and in ordinary times are currently accepted in final settlement of all debts - except, of course, the debts of bankers. It is by this latter characteristic especially that "book money" is distinguished from other kinds of credit which are not regarded as "money in hand;" in particular from bills of exchange, since the liabilities represented by these, though they may serve as a medium of exchange no less than bankers' liabilities, are always ostensibly to be liquidated at some definite time; hence they are not looked on as finally settling transactions, and so are not commonly regarded as money. The same remark applies to exchequer bills, as these are not absolutely convertible into legal tender to the amount they nominally represent, except at certain definite times. For similar reasons cheques are not commonly held to be money, though they perform some of the functions of money, since a transaction settled by cheque is not finally completed until the cheque has been paid in and money transferred from the drawer's banking account to that of the presenter of the cheque at the same or some other bank. In the case of cheques, however, there is the further difference that the cheques themselves do not exactly "circulate," though the liabilities transferred by means of them do. Still less, again, are securities, such as Government bonds or railway debentures or shares, regarded as ready money, since there is no time at which they are convertible into coin for a fixed amount : when taken in liquidation of a debt they must always first be sold like any other goods, or at least estimated at a continually varying market value ; though, no doubt, as being more conveniently carried and kept, and more readily exchanged than most commodities, they are better fitted for taking the place of money. We have, however, had occasion to notice a lax usage by which such bonds and shares are often spoken of as "money;" but here, as was said, the notion of "invested" money seems to be always implied. In this fifth signification money is almost, as Mill says, a synonym for wealth, but yet not quite, for we do not apply the term to landed estates or pictures in a picture-gallery; we confine it, in fact, to wealth that is readily negotiable, and of which the current value is more or less definitely known.
It is now time to consider the second question originally raised, "What ought we to call money?" But I must hasten to say that I have no pretension to answer this dogmatically. I should be quite content to accept any of the denotations above given; or indeed, since there are undoubted disadvantages in rejecting any established  use of language, I am quite willing to adopt them all at once, and to change from one to the other according to the nature of the subject, provided only the change be clearly announced. What most concerns us is that we should not give weak reasons for adopting any particular denotation of the term; that we should not misapprehend the nature or importance of the characteristics by which we distinguish money from what is not money. Thus if we restrict the term, in the old way, to coined metal, we must at least not do so because metallic money alone has "intrinsic value;" since it is not the difference in the source of the value of coin, confusedly expressed by the word "intrinsic," which is practically important, but the difference in its range and permanence. It is not because coin is made of a more expensive material that it is a better money than notes; but because it could be used as a medium of exchange over a wider area, and because its value is not liable to sudden destruction through the insolvency of the issuer, nor to sudden diminution in consequence of excessive issues. And it should be borne in mind that these distinctions are not absolute; there is no reason why we should not have an international circulation of bank-notes; and the progress of science and industry might so enlarge the supply of gold as to make it easy for a wise and stable government to devise a paper currency of more durable value than gold coin would then be, if still issued as at present. So again, I have no objection to define money by the characteristic of being "legal tender;" provided we do not imagine that it is this legal tender alone that becomes "scarce" or "abundant," and consequently cheap" or "dear," in what we shall still have to call the money market. Only in this case it will be well also to remember that the notes of the Bank of England, though in a certain sense "legal money," are not so in the sense most important to the political economist; since their legal currency would cease, if the Issue Department ceased to give gold for them, and therefore could hardly be effective in sustaining their value, if this ever came to be seriously doubted. No doubt the quality of these notes is unique; in the severest crisis they would be taken as readily as gold. But this is not due to the fact that they are legal tender, but to the special provision made for maintaining their convertibility; and perhaps even more to the general belief that the credit of the English Government is practically pledged to maintain it. And here again it must be observed that the unique position of the Bank of England has a practically equal effect in sustaining the currency of the liabilities of its banking department : in the worst of panics every one has considered "money deposited" with the Bank of England as safe as its bank-notes in his own strong chest.
Hence it seems to me that, in relation to English finance, the  definition of money that includes bank-notes generally, and excludes the rest of bankers' liabilities, is the least acceptable of all; since it ignores the profound distinction that separates the credit of the Bank of England from the credit of all other banks, while it unduly emphasizes the more superficial distinction between the liabilities of provincial banks that are transferred by notes and the liabilities of the London joint-stock banks that are transferred by cheques. No doubt there is actually an important difference between the working of the cheque-system and that of the note-system; since cheques do not circulate as notes do; the receiver of a cheque commonly pays it in without delay, and thus selects the banker whose liabilities he consents to take as money, whereas the receiver of a note usually exercises no such choice. But this is not the ground which I find most frequently given for drawing a broad line between notes and deposits; but rather such trivial reasons as that deposits cannot be "currency" because they do not pass "from hand to hand," as though the mere physical transmission were the important fact, and not the transfer of the ownership of bankers' liabilities.
Again, suppose we adopt the widest meaning of "money" which we have found to be current in Lombard Street, and include bankers' liabilities, whether represented by bank-notes or not, we shall still be excluding that other kind of "currency" of which the material is merchants' credit, represented by bills of exchange. We shall have, therefore, to bear in mind carefully that bills of exchange - so far as they still circulate among traders, and are not at once discounted - perform the main function of money, in being a medium for transferring commodities. And finally, for some purposes it would be convenient to extend the denotation of money still farther, and define it as any commonly accepted medium of exchange, so as to include bills of exchange, as the "paper money of commerce," as well as bullion, its metallic money. But if we do this we must not forget that what is most important to the buyer of commodities is, that the seller should take what he offers in exchange, and not that other people should take it. Now there are other valuable articles besides bills of exchange - as, for instance, Government bonds - which are more convenient for transmission than bullion; hence, if foreigners will take them in payment for the goods they have sold, as is now more and more the case, they will fulfil this important function of money better than the noblest metal. In short, whatever course we adopt, we shall find that definitions are not talismans for simplifying the complex relations of facts, but merely instruments by which, when we have thoroughly analysed any part of the complexity, we may fix in our thought some of the most important results of our analysis.
[Note 1 (p. 564)] : See Prof. F.A. Walker's book on "Money," c. xxi pp. 481-2.
[Note 2 (p. 564)] : See the Fortnightly Review for February, 1879.
[Note 3 (p. 564)] : I may refer the reader, for example, to Tate's "Cambist," Seyd's "Bullion and the Foreign Exchanges," Nicholson's "Science of Exchanges," &c.
[Note 4 (p. 568)] : A special exception has to be made in favour of Mr. McLeod, whose Theory of Banking contains, so far as I know, the first clear and full exposition of the nature and functions of bankers' deposits. In saying this, I must guard myself against being understood to approve of Mr. McLeod's general treatment of economics.
[Note 5 (p. 570)] : Polit.Econ.b.iii.c.viii.
[Note 6 (p. 572)] : In saying this, I do not mean to ignore the possibility that London might remain a banking centre, even after the greater part of its trade had passed to other cities. But this could only occur in a more highly developed state of international trade than the present, and such a development would not be possible, if London had not previously been a great centre of trade.