From the Mises blog; archived comments below.
Someone asked me the proper way to view deposit contracts, in the context of a discussion about fractional-reserve banking (FRB). He noted that in my A Libertarian Theory of Contract I state that contractual obligations can either be “to do” or “to give”; and that “to do” contracts are generally not enforceable due to specific performance, but can only result in damages if non performance actually occurs. This implies that the only real enforceable obligations are “to give” something. My correspondent asked me if this means that a deposit contract is not really a contract–i.e., based on the idea that a deposit is a “to do” contract (i.e., safekeep this deposit), while a loan would be a “to give” contract (i.e., return the money at the end of the term). An edited version of my reply follows.
First, I don’t view a loan as an obligation to return “the” money. It’s just a transfer of title to a certain sum of future money (IF you own it at the time). See my Contracts paper, pp. 32 et seq.
Second: in my article I wrote: “Contractual obligations may be classified as obligations to do or to give. An obligation to give may be viewed as a transfer of title to property, as it is an obligation to give ownership of the thing to another. An obligation to do is an obligation to perform a specific action, such as an obligation to sing at a wedding or paint someone’s house. ” The purpose of noting the positive law’s obligation to do/to give classification was to show that even in the current law, it’s really all about title transfer–to set the stage for Rothbard/Evers’s title-transfer theory of contract. Actually, I don’t view contracts as really being obligations at all; they are just title transfers. If I sell you my apple for $1, then the title to my apple transfers to you, even if/while I still (temporarily) possess it. Now, I am holding your apple, and I now have an obligation to let you have it, when you demand it, but not because of a contract or obligation to do, but just because of property rights, which the contract has rearranged. Contracts change, or rearrange property titles. Once this happens, people can then have different obligations–obligations to respect the property rights of the owners. If you find yourself in possession of property belonging to another person because a contract’s title-transfer provision has been triggered, even though it was your money a minute before, you now have to respect the wishes of the (new) owner.
I then explained that in the positive law, “to do” contracts are generally not enforceable due to specific performance, but can only result in damages if non performance actually occurs. My inquisitor says this implies that the only real enforceable obligations are “to give” something. And this is correct. And even in the case where they are enforced, such as a contract for a piece of real property, even here it’s not against the person but just effectuates the property title. I.e., you don’t need to call it an obligation. Again, my point was to show that the practical results of the current law can be achieved by the title-transfer conception of contract.
Now, does this imply that a deposit contract is not really a contract, on the grounds that a deposit is a “to do” contract (i.e., safekeep this deposit), while a loan would be a “to give” contract (i.e., return the money at the end of the term)? I don’t think so. I think that there are no obligations to do something–to perform; but you can arrange a contract so that there are consequences (the payment of damages, say–a title transfer to money) for failure to do something–the failure to do serves as a condition or trigger of a title transfer. Again, this is what the result is, in effect, in positive law, because obligations to do are not usually enforceable by “specific performance,” but by an award of monetary damages for failure to perform–for a “breach.” In the title-transfer theory of contract, failure to perform is not a breach of an obligation; it is just a previously-agreed upon trigger of a payment of damages. (In this way it is similar to the law and economics idea of “efficient breach of contract“–but arrives at this naturally, as a result of a principled respect for property rights–without any wealth-maximization contortions.)
In other words, the law views contracts as enforceable promises, and has various theories to justify this. Rothbard and Evers (and I) view this as confused. Rahter a contract should be seen as one or more title transfers. The ability to transfer title to property simply flows from the property rights of the owner–his right to exclude people, to permit them to use it, or to transfer all or some of the title to others, for charitable purposes, or in exchange for something else (some performance, or some other title transfer).
Now, how does all this apply to deposit “contracts”? Well, first, I assume we are talking here about a genuine deposit, as opposed to a FRB “deposit.” In a FRB it is a loan by the customer, not a deposit, since the customer loses title to his money–he has to in order for the “bank” to have the right to loan it to some borrower (who needs to have title to the money in order to use it for the purposes for which he borrowed it). On this see Huerta de Soto, Money, Bank Credit, and Economic Cycles, pp. 3 et seq., where he explains quite correctly that a standard loan is a mutuum (sometimes called a “loan for consumption”–see my A Civil Law to Common Law Dictionary).
By contrast, in a deposit (see Huerta de Soto, pp. 4-5) the depositary or custodian does not acquire title to the property. In a regular or specific deposit, such as a warehousing arrangement or safe deposit box, the depositor retains title to his property. For example, in a deposit arrangement, you give your money to the bank to warehouse it for you. In this case you don’t actually transfer title to your gold–you put it in the bank’s safety deposit box, say. Now is this really a deposit “contract”? Well, this description is a bit misleading. Contracts transfer title; but the title to the gold has not been transferred. What of the depositary’s “obligation” to warehouse the gold? Well, note that property owners have the right to set rules for use of their property; they have the right to give others varying levels and types of permissions or consent as to what they can do with their property. I can invite you to a party to my house, and some of the rules, stated or implied, include: you may not fornicate on my kitchen floor; you may not slap my dog; etc. I can lend you my car (a commodatum, Huerta de Soto, p. 3–or “loan for use,” Civil Law Dictionary) to drive to the movie for an afternoon, but not to abscond to Canada for a month.
So in a typical deposit arrangement, the best way to describe the relationship and situation consonant with libertarian property principles and the title-transfer theory of contact is that the bank agrees (consents) to let you use their property (box) in certain ways; while you let them control your gold in limited ways (they can prevent even you from accessing it, if you don’t: show proper ID, come during business hours, pay any owed fees, etc.). You agree to pay them a fee every month–title transfer of money to them, in exchange for their letting you use their box. So there is no real title transfer of the gold. (Consider a guard you hire to guard your home while you are on vacation; you “deposit” your home into his care, and agree to pay him for watching it; you do not give him permission to hold parties at the house or to own it or sell it. The depositary is like the guard. Would you call the arrangement with the guard a “house deposit contract”? If he trashes your house has he breached his contractual obligations, or has he simply committed trespass?)
In any event, the regular deposit is expensive and inefficient for deposits of fungible goods, such as grain or money, so such items are usually intermixed with deposits of other depositors; this is an irregular deposit (Huerta de Soto, p. 4). The irregular deposit can be viewed in this way: say there are 100 depositors. These people all agree to transfer title to their gold, and in exchange they receive a fractional ownership of the fungible mass of gold in the bank, with the right to withdraw their share at any time. The bank again has no title to the gold. They are like the warehouser in the case of the regular deposit. The depositors jointly own all the gold, have specified among themselves how each can withdraw his portion and exit the relationship.
Now if the bank loans your gold to someone else, they are actually stealing it–converting it. Trespassing. It has little to do with contract, unless by “contract” you mean the rules you, as owner, lay down for the banks’ use of your property (gold)–but by this loose usage, if you fornicate in my house during a party, you are in breach of contract. Okay, fine, but this is a bit of a stretch. I’d say it was a type of trespass–a use of my property that i have not consented to. But you can view the permissions granted by the owner as a type of contract, since it’s like a temporary and limited grant of a certain specified right to use the property to another person, whereas a normal contractual title transfer is usually permanent and complete.
Incidentally, I think we can assume there are a bunch of subsidiary, implied (or written) accessory contractual title transfers: e.g., IF the banker doesn’t safeguard your gold, THEN he pays damages to you (which is a title transfer of his money), etc. So there is a contract there.
So I would say a “contract of deposit” refers to the various permissions with respect to property (the bank’s permission to limit access to your money, but not permission to loan it out; your permission to use the bank’s facilities to store your gold); and subsidiary and related title transfers (your payment of fees; the bank’s payment of damages if it breaches certain prohibitions).
Note, however, that if we are talking about a FRB “deposit,” which is not really a deposit, but rather a loan to the bank, the entire analysis changes. In this case, the FRB does acquire title to the customer’s property; in exchange the customer acquires a future, conditional title-transfer from the FRB: title to a certain sum of money in the future at a certain time (say, when the customer makes a “demand”), but of course, only if the bank owns at that time assets to which title can transfer to the customer. If the FRB is bankrupt, due to a run (as some of believe is inevitably the case), then when the customer demands money, the bank simply has no money. This situation is then analogous to that of the deadbeat debtor discussed on pp. 32-33 of A Libertarian Theory of Contract.
The former is correct, but it is more a practical matter than anything else. The latter is false, because courts often can (unless they are too lazy) recast matters in a negative form. E.g., when a house painter fails and refuses to paint your house after agreeing to, maybe even accepting a payment on account, courts can issue an injunction in a negative form “the defendant may not paint any more houses until after he successfully completes painting the respondent’s house to the court’s satisfaction” (a breach would result in contempt of court and imprisonment) – estoppel. Lawyers often put this sort of negative form in contracts, e.g. a penalty clause requiring the painter to pay all receipts to a named charity until the contracted work is complete; this prevents any gain from non-performance, and by the funds not going to the other party there is a low risk that a court would vary this when it deemed that acceptable damages had been received – it keeps the effective estoppel open ended.
So, no, as the practical constraint is neither inherent nor unavoidable, it does not cause all contracts to rest on “to give” – either in practice or at a philosophical level.
Published: August 13, 2009 3:05 AM
In Free Banking the distinction between a deposit (100% R deposit or 100% R note) and credit or “promises of payment” (
And if it is the case, the discussion should focus on how one type of tittle and other would interact: would a price discount appear in the market between one and another?
Published: August 13, 2009 6:14 AM
Published: August 13, 2009 6:22 AM
Published: August 13, 2009 7:27 AM
I must say I am not quite sure what you mean. Can you give a citation or authority for these contentions? I do not see the connection between the contract clause you suggest, and a judge ordering a negative injunction. In any event, it is possible for judges to do this, sure, but to my knowledge it is rare, and your assertion does not seem accurate to me. See “>here; Barnett here, p. 181 (“specific performance has traditionally been unavailable to enforce contracts for the provision of personal services, even if these services can be shown to be unique and damages therefore are in fact inadequate. … Courts have given two reasons for their reluctance to award specific performance. First, they have sought to avoid the task of administering specific performance decrees. 6 It is sometimes difficult for a court to monitor the quality of performance or to assess claims that performance is not being appropriately provided. 7 Second, specific performance decrees for breaches of personal services contracts have been thought to pose moral problems as well, “because they are perceived to be substantively unacceptable limitations on personal freedom.” 8 Forcing someone to perform a contract strikes courts as involuntary servitude and is said by judges to be against “public policy.”9”); La. Civ. Code Art. 1986 (“Upon an obligor’s failure to perform an obligation to deliver a thing, or not to do an act, or to execute an instrument, the court shall grant specific performance plus damages for delay if the obligee so demands. If specific performance is impracticable, the court may allow damages to the obligee. … Upon a failure to perform an obligation that has another object, such as an obligation to do, the granting of specific performance is at the discretion of the court.”).
So I think your contention is wrong: it is true that by and large, most obligations to perform service in the current law are not enforced by a court ordering specific performance; rather, the court orders a payment of money damages for “breach”. The point is that such a result could be achieved by a title transfer theory of contract.
I never argued it required it–I was illustrating how we could achieve results similar to the current regime with title trasnfer alone. I believe you are incorrect in your assertions about specific performance–but in any event, it is not relevance since the case for title transfer stands regardless.
Again, this is how the positive law classifies it. This distinction falls away in the title trasnfer theory. In the positive law, contracts are seen as enforceable obligations. You promise to give something, or to do some action; they are both enforceable. But in the former case the court will actually make you give the thing; but in the latter, they will just make you give some thing (money damages). So it is their own distinction that is confused.
Read my contract article that I linked here. Of course it’s allowed.
Published: August 13, 2009 8:02 AM
Kinsella correctly recognizes that money given to a FRB is not a deposit but a loan. This is exactly where those who see money created by FRB are confused. The bank has the right to take this loan and loan a portion of it once again because there has been a transfer of property ownership. Can the FRB loan any money that it does not receive in a loan? No, because it has no money unless someone loans it money. So does a FRB create new money? No, it simply loans out money that has been loaned to it. The idea of a multiplier effect is an illusion.
Now the fallacy of money creation is the assumption that the total amount recorded in the bank is actually still property of the “depositor,” and the portion loaned out is new money rather than just a portion of the money loaned. The truth is that the amount loaned is actually a portion of that loaned to the bank.
Consider this. Rather than money let’s say the FRB “deposit” is a car. Now let’s assume that the FRB takes off the tires and loans them to one person, then the seats are removed and loaned to another person, and on and on. Has the FRB created another car? No. Now let’s assume that someone enters the bank and demands back the car they loaned to the bank. Perhaps the bank still has the engine and the body but they then take tires and seats from another “depositor” and give their customer a car. Has a new car been created? Now when you are dealing with fungible assets like money the substitution is even easier, but no new money is created.
Once you correctly define the money “deposited” in a FRB you can then see that FRB actually does not create any new money. Only the FED can create money.
But where we run into a real problem is when the government guarantees FRB “deposits?” Because the money has been loaned to the bank, if there is a run on the bank or the bank otherwise goes out of business the “depositor” should suffer the loss just like any other entity would with a bad debt. But when the government replaces the money loaned with FDIC money received from the FED new money is created.
Published: August 13, 2009 12:41 PM
“This is an excellent piece and it demonstrates why fractional reserve banking does not create money.”
Thanks Dick. I think this piece helps to demonstrate that, as I explained in my recent talk on IP and Libertarianism, this is an example in my view of how a sound, coherent view of libertarian and economic principles can help one apply it to a murky area to see the proper way to view it.
I think the FRB advocates (of which I am not one) could grant it’s a loan but argue that the FRB notes–which are really then a type of credit, a promissory note, basically–could still serve as money substitutes. That is where the economic argument is, it seems to me. Personally, I see no reason in principle why this could not be the case; and it is certainly not inherently fraudulent, if enough disclosure is given (thought it is fraught with the potential for fraud). So just as “It is, for example, not strictly speaking impossible to introduce money through collective deliberation at a council” (Hülsmann), it is not impossible for people to trade credit notes as money, or for them to trade banana pies or toenail clippings. It’s just that we have good reason to think people would not trade these credit notes as money so long as they are not defrauded as to their nature.
Well, there is a multiplier effect now, given the way banks under the Federal reserve system operate. In a free market, there would not technically be a multiplier since each customer loses title to the dollar loaned to another person. But there is no legal prohibition on such customers using their credit notes in exchanges–or for them being used as media of exchange… or for them becoming general media of exchange. This is not legally prohibited, nor, in my view, it is logically impossible. However, what does block this from happening is just the operation of the laws of economics.
Published: August 13, 2009 1:03 PM
I agree with you that when the depositor uses his credit note in exchange it can create money (though not always), but it is interesting that the bank in this case is not creating new money, the “depositor” is. If we correctly define the “depost” as a transfer of title then there is no money creation. The problem is with our legal system allowing one asset to be “sold” twice. This is in fact fraud, but once again it is interesting to note that the fraud is not committed by the bank but by the “depositor.”
Published: August 13, 2009 2:03 PM
Let say I deposit a bill of 100 $ in a demand deposit account, and the bank then lend 90 $ of this 100 $ to somebody else. In exchange of my 100 $ in the form of paper money, the bank gives me 100 $ in the form of ‘account money’.
While the borrower is spending his 90 $ in market transactions, I can also, in the same time, spend my 100 $ of ‘account money’ in market transactions.
Thus there is now 190 $ in money supply in the market instead of 100 $.
This too is a creation of money.
Published: August 13, 2009 8:15 PM
– to decrease the amount of total demand deposits (“loans” to the bank)
– But it would decrease in what client/clients? In those that would agree to transform “demand deposits” in “time deposits” (a formal loan to the bank). So this would have to be previous to a new loan.
If you think about it:
– the most effective way this happens today in the market is through “Mutual Funds”, where the “Fund” actually only lends (buying securities, making time deposits, etc) money that entered the “Mutual Fund”.
-The Bank can now incur in a time lag risk between the portfolio of loans and the “time deposits”, so, the banks would have to have reserves (money) against the total amount of time deposits because in theory and practice could face a bank Run of “time deposits”.
So, in a 100%RB, a Bank must kept a reserve equal to 100% demand deposits plus the reserves against the risk of mismatch between its loan and “time deposits”, which means, that total reserves of a 100%RB is much higher than 100% (to the total of demand deposits).
This is the reason why i think Murray N. Rothbard talks about a 100% Reserve System would be “much more liquid” than the present one.
True “demand deposits” (100%R) should not be registered in the Balances of the Bank like
Asset: money deposit
Liability: demand deposit
must as off-balance items. Only time deposits should
For a FRB although, every deposits turns in to a loan to the bank and a “promise of payment” for the client.
And so we are back to the question:
Would 100% true deposits (100%) be exchanged with “promises of payment” at par value?
Published: August 14, 2009 3:02 AM
Subsequent drawing down of the loan balance by the borrower, with possible associated interbank reserve transfers, shouldn’t obfuscate the actual process of loan creation, which does not require any reserve base whatsoever and only depends on bank capitalisation being sufficient.
The money multipler is backwards. Loans create deposits, not the reverse.
Published: August 14, 2009 4:15 AM
When you deposit money into a FRB you join with many others. Now it is a truism to say that if you all attempt to withdraw you money an once the bank will become insolvent and you and the other “depositors” would lose money (assuming no FDIC). You would not be able to withdraw more than the total deposited plus that loaned out. So while it appears that the $90 was from your “deposit” if you draw your $100 the $90 loan actually comes from other “depositors” not you so no money is created.
Now if you deposit $100 and the bank allows you to draw $100 then your balance is zero and no money is created. If you are the only “depositor” and you deposit $100 and the bank lends $90. If you attempt to draw your $100 the bank will become insolvent and you will lose money unless it can call the loan.
It is important to understand when you deposit money into a bank and the bank holds a portion in reserve the result is actually the reverse of a multiplier. It is a negative multiplier because that portion in reserve is not in the active economy, only the amount loaned or drawn is active.
Let me repeat again, the bank cannot loan out or allow you to draw any money it has not received (as a loan) from a “depositor.” Any withdrawal in excess of “deposits” causes the bank to become insolvent.
Published: August 14, 2009 2:31 PM
There is money in the form of bank notes and coins, and money in the form of ‘account money’.
For the bank to become insolvent when every depositor attempt to withdraw their money, there must be more money in the form of ‘account money’ then money in the bank in the form of paper dollars and coins. There is less money in the reserve of the bank than money in the demand accounts. This surplus is there because the money supply has increased by way of fractional reserve banking, or in another words, money have been created.
Published: August 14, 2009 5:43 PM
(N.B: In my example, the 100 $ is not drawed but spent in the form of ‘account money’, and after the loan of 90 $ has been made.)
Nevertheless, that the 90 $ comes from my deposit of 100 $, or from someone else deposit of 100 $, it doesn’t change nothing. The result is the same: there is now 90 $ more in the money supply.
In exchange of my, or somone else, 100 $ in the form of paper money, the bank gives me, or somone else, 100 $ in the form of ‘account money’. This ‘account money’ is part of the money supply, say M1, but the 100 $ in the reserve of the bank is no more part of this money supply. Thus at this stage, the money supply is not changed.
When on reception of this 100 $, and base on this new money in its reserve, the bank decide to loan 90 $ to some borrower in the form of ‘account money’, as they usually do, then this new ‘account money’ becomes part of the money supply too, thus adding 90 $ to this money supply. There is now $190 ready to be spent in market transactions, while there was only $100 before this process.
Once the loan of 90 $ has been made, that I withdraw 100 $ from my bank account or let it in my account, or spend it by writing a check to a merchant, there is still 90 $ more in the money supply.
Published: August 16, 2009 6:14 PM