I have been following several posts on Lord Keynes’ excellent blog on the legal issues related to the history of fractional reserve banking, arising out of Rothbard’s The Mystery of Banking. I wonder if he is a lawyer or has formal legal training. I do and that’s perhaps why I find these questions particularly interesting. Even if Lord Keynes doesn’t happen to have formal legal training, I can find no fault in what he has said in his various posts. Still, I think I might be able to make some useful contributions. Originally I meant to write a comment on his blog, but this has grown out of all proportion, so I decided to activate this blog and make this my first post, linking to it in a comment on Lord Keynes’ blog.

If there is one thing on which Austrian economists are to be commended, it is the fact that their writings are, in general, freely available on the Internet. Then again, one might be a little nasty and remark, not without justification, that that’s just as it should be, since no reasonable person should be expected to pay to read the nonsense they usually produce. I therefore had no difficulty in downloading and reading parts 1 and 2 of chapter VII of the 2008 edition of Rothbard’s book above. Rothbard certainly subscribes to the story according to which fractional reserve banking arose from the near-fraudulent activities of the goldsmith bankers. I seem to recall, although I do not have the book to hand right now, that Kindleberger, in his Financial History of Western Europe asserts that this story is a myth. I see that just about the only authority Rothbard cites to back-up his claims in this respect is J. Milnes Holden, The History of Negotiable Instruments in English Law (London:The Athlone Press, 1955). I have not seen this book (I’d much like to, but I can’t find it anywhere, online or offline) so I’ll refrain from passing judgement on that particular matter. Still, I have done extensive online research in old cases and I think I might have a few other useful things to say. In general, they complement and reinforce Lord Keynes’ own conclusions and, hopefully, show up Rothbard’s assertions as nonsense, at least as far as the law is concerned.

I think Rothbard’s mistake (one of several) starts at p. 87 of his book, where he adopts an economist’s, not a lawyer’s, definition of a loan. He writes: “In a loan, or a credit transaction, the creditor exchanges a present good—that is, a good available for use at any time in the present—for a future good, an IOU redeemable at some date in the future”. Starting from this definition, he asserts that the deposit of specie money at a warehouse is a bailment, not a loan. Lord Keynes has written extensively on the distinction between Mutuum and bailment but, and that’s the only reproach I might have for him, the authorities he cites concern Roman and what is called “Civil Law”. Whilst useful, in showing that the distinction is centuries- if not millennia-old, they do flash some warning lights for lawyers, since English or, more general, Anglo-Saxon law, often called Common Law (as opposed to, precisely, Civil Law), is not supposed to derive directly from Roman law. It is continental law (French, German etc), usually termed Civil law, as opposed to Common law, which is supposed to derive directly from Roman law. One might therefore wonder whether the distinction between mutuum and bailment is also valid in a Common Law context. This is what I set out to discover, after noticing that Lord Keynes’ references were mostly to Roman and Civil Law. I might say that that there is nothing in what I have found that would support Rothbard’s position and much to reinforce Lord Keynes’.


I think a better reference, in the context of English Common law, concerning the distinction between mutuum and bailment is to the famous 13th century English legal writer Bracton. In volume 2, p. 284 of his book De Legibus et Consuetudinibus Angliae lines 1-26 he states (in the English translation):


An obligation is contracted by a thing, as by the giving of a mutuum, a loan for consumption, which consists of things reckoned by weight, number or measure. By weight, as in things which are weighed, copper, silver or gold; in number, as in coined money; in measure, as in wine, oil or grain. Such things, [ascertained] by weighing, counting or measuring, are given so that they at once become the property of those who take them, for that is properly termed a mutuum which, being meum becomes tuum, and whenever not the very things but others of the same kind are returned to the creditor, or their value if they are consumed or lost by fire, earthquake or shipwreck, or stolen or carried away by thieves or enemies.He to whom a thing is given for use is also bound by the thing lent him. But there is a great difference between a loan for consumption and a loan for use, for he who has taken a loan for use is bound to restore the very thing, and, [though] he is not excused if he shows as much care in its safekeeping as he ordinarily bestows on his own goods if another could have safeguarded the thing with greater care, is not held liable for force majeure or accidents unless there has been culpa, as where he takes on a journey a thing lent him for use at home and loses it in an attack of enemies or thieves or by shipwreck; he is then clearly liable. A thing lent for use is said to be given ad commodum, as an accommodation, and is properly understood to be lent when it is given without recompense. For a loan for use ought to be gratuitous and if payment is involved the transaction ought rather to be called a letting and hiring than a loan.He with whom a thing is deposited is [also] bound re and held to the restoration of the very object he accepted, [or its value if it is lost and] he has committed some wrongful act. For culpa, that is, carelessness or negligence, he is not liable, for he who entrusts a thing to the care of a negligent friend can only blame himself and his own lack of caution.


The distinction between a loan and a bailment is, then, according to Bracton, that a loan concerns fungible assets (including, but not limited to, money), with an obligation to return to the lender a quantity of the goods (or money) loaned, but not the very objects that were loaned. For example, if I knock on your door to borrow a cup of sugar for the cake I’m baking, I undertake to return to you a cup of sugar but, obviously, not the sugar itself that was lent (which will end up in my cake!). If I also borrow the cup containing the sugar, this will be a bailment, provided I undertake to return to you the very same cup. Although Rothbard does talk about fungible goods (p. 89), he does not understand the legal significance of fungibility: it is the explanation of why a loan will show up in the borrower’s books as a debit and credit, whilst a deposit of, say, furniture in a warehouse (cf p. 87) will not. It also explains why money deposited with a goldsmith or banker becomes their property. Once one has understood this principle, it is obvious that a goldsmith who lends out the money deposited with him does not commit embezzlement: he is lending out his own money. Further, contrary to what Rothbard states (without citing any authority), the paper given to the customer of a goldsmith with whom money has been deposited should not be considered a warehouse receipt, but rather a promissory note or such analogous instrument (bond, bill, etc) since the goldsmith is under an obligation to pay a certain sum of money upon demand and not to return the very coins deposited with him. This is also why in later cases (such as Carr v. Carr) it was ruled that a bailment, or depositum, of money could only involve money in a sealed bag.


To sum up, then, Rothbard’s mistake seems to be the following: he completely ignores that, at least as far back as the 13th century, English law clearly distinguished between a loan of money or other fungible assets and a bailment of goods, with an obligation to return the very same goods deposited. It is also clear that anyone depositing money without taking appropriate steps to ensure the return of the very coins deposited (such as placing them in a sealed bag or box), should have been aware that they were transferring property of the money deposited to the banker or other person receiving the deposit, who might do with that money as they pleased, subject to their obligation to pay the depositor an equal sum of money (or a higher sum, if interest was added, subject to the upper limits provided for in the Statutes of Usury). This being the legal position, it is absurd to assert that goldsmiths or bankers who were treating the money deposited with them as their own, were embezzlers or were defrauding their customers. It is just as absurd to assert that they were printing counterfeit warehouse receipts. As explained above, they can’t have been handing out such receipts: what they were handing out were promissory notes, both to customers who deposited funds with them and to customers who they were making a loan to.


The recognition of the fungibility of money is also at the root of the adage that “money has no earmark”, which is also mentioned in Carr v Carr. Its origin is at least a century older and it is mostly used as justification for the impossibility of recovering specific monies that where misappropriated (or what has been purchased with them), in preference to other creditors. Thus, in Kirk v. Webb (1698) Prec Ch 84; 24 ER 41, a case decided in 1698, the Bishop of Coventry, as trustee for his niece, had received the rents and profits of her estate and applied them in a purchase of land. He then died without sufficient assets and it was ruled that the niece was not entitled to the land, in preference to the deceased’s creditors. This judgement was later (1699) confirmed upon appeal to Parliament. In a subsequent similar case, Kendar v. Milward, (1702) 2 Vern 440; 23 ER 882, the report expressly states that the case was “within the reason of the case of Kirk and Webb that money had no ear-mark, and could not be followed when invested in a purchase”. It must, however, be said that, in later cases, the harshness of these rulings has been somewhat relaxed, the courts finding ways to allow the owners of misappropriated funds to recover assets purchased with their funds.


Research in old cases also makes nonsense of Rothbard’s other assertion (p. 89) that “bailment law scarcely existed until the eighteenth century”. Coggs v. Bernard (1790) 2 Ld Raym 909; 92 ER 107 is a bailment case decided in 1704 (very early 18th century). It cites with approval part of the passage from Bracton also cited above and distinguishes between no less than six different types of bailment. The question at issue there was whether persons who, without payment, had undertaken to carry some casks of brandy from one warehouse to another and had broken one cask whilst setting them down, were liable to pay compensation for the damage. The problem was that no “consideration” had been given to the carriers for their promise to carry, but it was nevertheless held that they were liable. Coggs v. Bernard cites several other, necessarily older, cases on bailment, for instance Southcote’s case (1598) 4 Co Rep 83; 76 ER 1061, decided in 1601 (very early 17th century). In Southcote’s case the question was whether a person entrusted with the care of goods could be excused from delivering them up if they had been stolen from him (it was held that he could not). Contrary, then, to what Rothbard seems to be asserting, it is not that bailment law was not sufficiently developed to catch “deposit bankers”; quite simply, in the eyes of the legal system, deposits of money were not bailments, and bailment law, quite sufficiently developed, did not apply to them. They were considered debts and dealt with accordingly.


It should also be noted that Rothbard cites no case-law or other authority in support of his assertion (p. 89) that “It was only by the twentieth century that the courts finally decided that the grain warehouseman was truly a bailee and not simply a debtor”. Given the absence of any citation, one may legitimately doubt if indeed anything of the sort has ever been decided. The matter warrants further investigation which, however, seems to me outside the scope of this, already fairly long, post.


Why, then, do the first cases cited in support of the proposition that a banker’s obligation to his customer is a debt (and has nothing to do with bailment) all date from the 19th century? A first, simple, explanation is that it is very difficult to find case law in support of a proposition that is universally accepted and regarded as self-evident. If that is the case, no-one bothers to challenge it and the courts have no occasion to rule on it. I would contend that this was the case of the proposition according to which a banker’s obligation to his customer is a debt and does not result from a bailment. Indeed, a closer analysis of the three 19th century cases that seem to have definitely resolved this matter reveals that, at best, there was only an indirect link between that proposition and what was at issue in those cases. This might explain why Lord Keynes seems to be grappling somewhat with the two cases (Carr v. Carr and Foley v. Hill and others) on which he has posted so far (he has not yet posted on the third case mentioned by Rothbard, Devaynes v. Noble, and I think I know why; see further below). Upon reading the reports of those cases one may well be forgiven for thinking that they concern entirely different matters.


But I think there is more to those three cases. Ironically, it seems to be the exact opposite of what Rothbard suggests in his book: at p. 91, he asserts that “banking law was in even worse shape than overall warehouse law and moved in the opposite direction to declare money deposits not a bailment but a debt”. In reality, as I hope to show below, these cases can best be seen as (failed) attempts to move banking law away from the basic proposition that it had always maintained, namely that a customer’s deposit of money with a banker gives rise to a debt of the latter to the former, and no more. As I hope to show below, the arguments put forward against this proposition were not wholly unreasonable and would find much sympathy with modern bank customers. Nowadays, where the vast majority of the money we all possess and deal with is nothing more than (electronic) entries in a bank account, we would find much sympathy with the proposition that such entries are quite distinct from ordinary debts owed to us by individuals or non-banking firms. Still, cogent as they were, those arguments failed to convince early 19th century English judges who, and that is important, stayed quite put and maintained the orthodoxy such as they knew it. In other words, far from banking law moving in an unfortunate direction which gave rise to modern fractional reserve banking, as Rothbard asserts, quite the opposite happened: it remained where it had always been and, by so doing, failed to acknowledge the special nature of banks (and thus, perhaps, curb some of the worst abuses of fractional reserve banking).


Let’s take a closer look at the three cases in question. Lord Keynes has done an extensive post on Carr v. Carr (1811) 1 Mer 625; 35 ER 799, citing most of the report of the case verbatim, so I need not say very much. But what, I think, is important is the fact that even counsels for the defendant (who was asserting that a cash balance with a banker was to be considered as ‘money’ bequeathed to him, as opposed to a ‘debt’, bequeathed to the plaintiff) admitted that the cash balance was “strictly speaking, a debt”. Their argument, not wholly unreasonable, was that in the testator’s mind it was considered as money and that “in construction of Wills, words ought not to be taken according to their strict legal meaning, but according to the intention of the Testator; and that, in this case, the Testator could not have conceived it to be a debt”. The judge refused to follow this reasoning and ruled that the cash balance passed to the plaintiff, as a debt.


I need to say a lot more on the second case mentioned by Rothbard (p. 92), Devaynes v. Noble. Lord Keynes has not (yet) commented on that one and I bet this is due to the fact that he has failed to find the report. Indeed, Rothbard himself does not seem to know quite what he is talking about and, in all probability, had not read the report but relied on J. Milnes Holden, The Law and Practice of Banking, vol. I, Banker and Customer (London: Pitman Publishing, 1970, p. 31), which he cites at footnote 7.


Devaynes v. Noble is a very convoluted case that arose from the bankruptcy of the banking house of Devaynes, Dawes, Noble and Co. The bankruptcy occurred after the death of one of the partners, William Devaynes. His son and heir refused to take his father’s place in the bank, but the firm continued for almost a year after W. Devaynes’ death, when it was declared bankrupt. The main question was whether, and to what extent, the creditors of the banking partnership could claim against the estate of William Devaynes. Understandably, it was a monster of a case concerning all the bank’s customers, and a number of typical “test cases” were selected to be decided, in order to enable the assistant judge (the master) to resolve matters with regard to all the customers. The various cases are known by, and reported under, the corresponding customer’s name, which could cause confusion and make it very difficult for a non-specialist to identify the case Rothbard is talking about. The most important of those test cases for general purposes (and the most famous) is the second, Clayton’s case, but it is irrelevant for our purposes. The one that interests us (and is the one Rothbard refers to, although he probably did not know it) is the first, Sleech’s case (1816) 1 Mer 539; 35 ER 771. One of the matter’s at issue there was whether Miss Sleech was barred from recovering against the estate of Devayne, on the equitable defence of Laches, on account of the fact she had not withdrawn all her money from the bank immediately upon hearing of Devayne’s death but continued dealing with the surviving partners. Ordinarily, a delay of some eight months in acting to pursue a debt could not give rise to a defence of laches but an attempt was made, on behalf of Devayne, to argue that this was no ordinary debt. Again, on purely economical terms, especially from today’s perspective, it is not wholly unreasonable to say that a customer who has continued dealing with a bank for a whole eight months after an event that might be seen to affect the bank’s solvency, was not worried about the latter. It is in this context that counsel for the plaintiff (Devayne’s estate) made the assertion cited by Rothbard (p. 92) according to which “a banker is rather a bailee of his customer’s funds than his debtor” [p. 771 of the report; it is notable that, after citing this argument, the report contains an endnote numbered (2) which appears at p. 799 and is, in fact, the report of Carr v. Carr, cited above, which says the exact opposite of counsel’s assertion!]. However, the same barrister, admitted later, in reply (p. 777 of the report) that “In one sense, undoubtedly, the banker is a debtor to his customer”. He qualified this admission by adding: “but it is a species of debt contracted on the ground of mutual faith and confidence, that the entirety of the money shall be ready at a certain place, and at a certain moment; and it is on the ground of this peculiar confidence, that the customer who so deposits his money, does it without reservation of interest for the use of it”. To my mind, what this lawyer was trying to do was not deny that a deposit with a banker gives rise to a debt, much less plead in support of some fictitious ancient rule according to which it was a bailment, but argue, not wholly unreasonably, especially from today’s perspective, that it was a special kind of debt which should not be subject to the ordinary rules pertaining to debts. As Rothbard also admits, this attempt failed to cut any ice with the judge, who confirmed the previous position, that a deposit with banker gave rise to a debt, no less and no more.


Finally, there is the case of Foley v Hill and Others (1844) 1 Ph 399; 41 ER 683. Lord Keynes has cited almost the whole of the report so, again, I need not say too much. Basically, this case concerns what, today, might have been termed a “dormant” account in a bank. No transaction had been recorded in the plaintiff’s account for over six years, which meant that the bank was entitled to set up the Statute of Limitations against a claim from the customer. If I understand the report correctly, in particular the reference to an action for “money had and received” towards the end, the plaintiff was not so much in danger of losing the principal on the account, but also wanted to recover the interest, @ 3% per annum promised to him by the bank but not actually entered onto the account for over six years. Again, from a modern perspective, a bank customer might be shocked to discover that he was not entitled to claim interest against his bank merely because his account had remained dormant and the bank had neglected to record the interest on the account-sheet or passbook or the like. The plaintiff’s lawyers tried to argue that “the relation between a banker and his customer was not simply that of a debtor and creditor, but a confidential relation, which imposed on the former an obligation to keep the account according to the agreement made between the parties when the account was opened”. I have underlined the word “simply” because, to my mind, it indicates that counsel for the plaintiff is not trying to deny that there is a debt from the banker to the customer, but trying to argue that it is a special kind of debt. In any case, the attempt failed and the significance of the case seems to lie in the fact that the judgements in Carr v. Carr and Sleech’s case [and also in Sims and Another v. Bond and Another (1833) 5 B & Ad 389; 110 ER 834] was upheld by the highest court in England, the House of Lords, in the person of no less an authority than the Lord Chancellor of England himself.


Alright then, before finishing, a short summary of what all the above indicates: In spite of Rothbard’s arguments, there is nothing in old English law to suggest that, legally, the relationship between a banker, even a “deposit banker” in Rothbard’s terminology, was ever that of bailment. On the contrary, on the basis of what Bracton had written already in the 13th century, it is clear that the relationship in question was that of debtor and creditor. The cases in the 19th century that expressly confirmed this proposition arose out of attempts to assert (not unreasonably, in view of the increasing significance of banking) that there was something special about the relationship between a banker and his customer. These attempts were resisted by the courts and ultimately failed, with the old orthodoxy prevailing: a bank deposit gave rise to a debt, no more and no less.




Bracton, De Legibus et Consuetudinibus Angliae, both in the original Latin and in English translation, here.


All cases cited above can be found in the English Reports, via this search form.