29 January 2010

Prof. Lawrence H. White responds to Jesus Huerta de Soto on Banking (part 3)

Following parts 1 and 2:
Prof. Lawrence H. White responds to Jesus Huerta de Soto on Banking
Prof. Lawrence H. White responds to Jesus Huerta de Soto on Banking (part 2)

and thanks again to Prof. Steven Horwitz who emailed me the series of responses to Jesus Huerta de Soto's major work against fractional reserve banking, I am pleased to re-publish part 3 of 3. The original site these are supposed to be found on does not seem to be up anymore.

Part III.

Monday, February 05, 2007

In the third chapter of Money, Bank Credit, and Economic Cycles, Jesús Huerta de Soto (p. 115) considers "the proposed arguments intended to legally support a monetary irregular deposit contract in which the depositary can make self-interested use of money on demand deposit." This is a rather loaded characterization, given that defenders of fractional-reserve banking in fact deny that a checking account is an "irregular deposit" contract in Huerta de Soto's sense (a warehousing contract). Huerta de Soto (p. 117) even more clearly constructs a straw man when he attributes to the defenders the view "that it is legitimate, from the standpoint of general legal principles, to misappropriate funds deposited for safekeeping and to issue deposit receipts for more money than is actually deposited," where "deposited" means "warehoused". Nobody - of course - advocates that it is legitimate to misappropriate funds. What the defenders consider legitimate is for a bank to use funds lent to it for its use, and (where allowed by its customers) for a bank to have demandable liabilities in excess of its cash reserves on hand, provided that it does not fail to meet any contractual obligation to satisfy redemption demands that are actually presented.

In my review of Chapter One I noted that Huerta de Soto seems to deny the possibility or legitimacy of loans with prepayment options or call options. In Chapter Three he explicitly denies the legitimacy of another common financial contract, the repurchase agreement. Repurchase agreements "entail an evasion of the law" (p. 118).

In Chapter One, Huerta de Soto enumerated the features of a term loan contract and the contrasting features of an "irregular deposit" (warehousing) contract. Here he re-emphasizes (pp. 119-20) that "the very essence of the irregular deposit contract demands that the purpose of safekeeping or custody predominate." [Emphasis in the original.] Indeed, a customer who takes his goods to a warehouse for safekeeping does not authorize the warehouse to use the goods being stored. He pays storage fees and does not receive interest. A customer who makes a term loan to a bank (e.g. buys a "certificate of deposit"), by contrast, authorizes the bank to put the funds to use. He does not pay storage fees and does receive interest. The contrast is very clear. How then do we regard the seemingly intermediate case of a demandable bank account or banknote? When a customer opens a demand account that pays interest or does not charge storage fees, which features require that the bank defrays its costs by putting some of the funds to use, it seems clear that the customer (as in the case of a term account) is authorizing the bank to put the funds to use. That is, the demand account is a loan, not a warehousing or "irregular deposit" contract.

Not so fast, says Huerta de Soto: the demand account or banknote contract calls for equivalent funds to be repaid to the customer whenever he may call for them. Since it shares this feature with a warehousing contract, Huerta de Soto concludes that it must be a warehousing contract, and the bank's use of any of the funds must be illegitimate. But this is a simple non sequitur. Demandability is feasible with less than 100% reserves. Demandability therefore does not entail a warehousing contract.

Here is Ludwig von Mises' clear explanation for the feasibility of fractional reserves in The Theory of Money and Credit:

A person who has a thousand loaves of bread at his immediate disposal will not dare to issue more than a thousand tickets each of which gives its holder the right to demand at any time the delivery of a loaf of bread. It is otherwise with money. Since nobody wants [coined] money except in order to get rid of it again, since it never finds a consumer except on ceasing to be a common medium of exchange, it is quite possible for claims to be employed in its stead, embodying a right to the receipt on demand of a certain sum of money and unimpugnable both as to their convertibility in general and as to whether they really would be converted on the demand of the holder; and it is quite possible for these claims to pass from hand to hand without any attempt being made to enforce the right that they embody. The obligee can expect that these claims will remain in circulation for so long as their holders do not lose confidence in their prompt convertibility or transfer them to persons who have not this confidence. He is therefore in a position to undertake greater obligations than he would ever be able to fulfill; it is enough if he takes sufficient precautions to ensure his ability to satisfy promptly that proportion of the claims that is actually enforced against him.

Huerta de Soto is aware of the argument that a prudent bank can in practice honor all the redemption demands made by holders of demand liabilities while holding fractional reserves. He rejects it, though he does confront Mises' argument directly. He believes that historical experience, before governments intervened to prop up banking, made it "clear that all banking systems which had been based on a fractional reserve had failed" (p. 126). All? Really? The curious reader wonders how Huerta de Soto would account for Scotland, Sweden, Canada, and so on, systems that other banking historians have considered successful fractional-reserve systems without central banks or deposit insurance. More generally, the reader wonders how a universally failing business plan survived for centuries.

Huerta de Soto presumes that a demandable bank liability must represent a warehousing contract. He cites examples (pp. 122-24) of specious arguments by violators of warehousing contracts (attempting to justify their own use of stored goods or securities), as though such cases showed the illegitimacy of bankers making use of funds provided to them under non-warehousing debt contracts. He declares "erroneous" Judge Lord Cottenham's decision in the 1848 case Foley v. Hill to the effect that, under the terms of his contract, a banker - unlike a warehouseman, who has a different contract - is authorized by his customers to put to use the funds placed in his custody, though of course he is obliged to repay them when demanded. Huerta de Soto views the judge as having ruled "that the monetary irregular-deposit contract was no different from the loan contract, and therefore that bankers making self-interested use of their depositors' money did not commit misappropriation." In fact, the ruling means that the bank account in question was not an "irregular-deposit contract" in Huerta de Soto's sense, i.e. was not a warehousing contract, but was a debt contract. To say that a particular bank contract was a debt (or loan) rather than a warehousing contract is not to say that a warehousing contract is the same as a loan contract. The judge did not say that he would refuse to enforce the terms of a warehousing contract; he said he did not find one.

In a section on Spanish law, Huerta de Soto (pp. 129-30) quotes Article 310 of the Spanish Commercial Code, which specifies that "deposits" in a warehouse or bank shall be "governed first by that company's statutes, then by the prescriptions of this code and last by common law rules applicable to all deposits". In other words, banks and their customers can contract around the code's default rules. Rather than applaud this freedom of contract, Huerta de Soto says that it "grants bankers a statutory privilege".

It would be odd to consider it an illegitimate privilege for a bank to put funds to use if one recognized that its customers authorize the bank to put those funds to use. But Huerta de Soto (p. 136) denies that customers would ever provide any such authorization:

If the depositor were informed that the contract he plans to sign is a loan contract by which he will grant a loan to the bank, and that therefore the money will no longer be available to him, he would certainly not go through with the contract as if it were a deposit [i.e. warehouse contract], and he very well might decide to keep his money.

But the customer is of course informed that the contract he plans to sign does not charge him storage fees and does pay him interest on his account balances. Surely that is tantamount to informing him that he is not making a warehousing contract. In a very practical sense, moreover, the funds will be available to him whenever he seeks a cash withdrawal or writes a check to transfer funds outside the bank. Though the bank admittedly could not satisfy all customers simultaneously should they all try to withdraw at once, the risk of that event (experience tells us) is small (though not zero) for customers of a prudent bank. Absent a run, the bank can satisfy all redemption demands that its customers actually do make. Runs have been rare for banks in competitive systems where government does not restrict capitalization, diversification, or branching.

Huerta de Soto rightly dismisses the view that the same account contract can mean one thing to the customer and another thing to the banker. But a demand account embodies no such contradiction. Huerta de Soto (pp. 136-7) considers that "depositors clearly turn over their money with the desire to retain full availability of the good turned over (monetary deposit 'on demand')". But this is far from clear for a customer who opens a typical bank checking account. In Huerta de Soto's terms, the customer can only "retain full availability" if the bank holds 100% reserves. This implies - for the bank to cover its costs and not incur losses - that the customer must agree to pay storage fees (and to receive no interest) on the account balance. A customer who opens a checking account that charges no storage fees (or pays interest) must therefore not intend to "retain full availability" of the money turned over. Huerta de Soto apparently thinks that customers do not notice that no storage fees are being assessed (or that interest is being paid), when he declares: "It is certain that when the overwhelming majority of depositors make a demand deposit, they are under the honest impression that they are in fact doing just that: entering into an irregular deposit contract, the essential purpose of which is to transfer the custody or safekeeping of their money to the banker." We are apparently supposed to believe that customers think they are making a warehousing contract, when nothing in the contract specifies warehousing, and the absence of storage fees (or payment of interest) logically entails that the bank will put the funds to work earning a return in which the customer will implicitly (or explicitly) share.

If warehousing is not the customer's purpose in opening a checking account, what is? I suggest: writing checks. The account gives the customer the ability to transfer and receive money conveniently. For that purpose, 100% reserves are an unnecessary cost. It is true, as just noted, that the probability of bank default under fractional reserves is not strictly zero. But it does not need to be zero, only sufficiently low, for the customer to find a fractional-reserve checking account (with no storage fees, or bearing interest) more attractive than a 100%-reserve account (storage fees, no interest).

Huerta de Soto (pp. 145-47) does entertain the possibility that the bank customer agrees to a debt rather than warehousing contract, but takes it for granted that in a debt contract the customer must "agree to the loss of availability of the good and its transfer to the banker for a set term in exchange for interest." But loss of availability for a set term, of course, is incompatible with writing checks on the account. In assuming that a debt contract must have a set term, Huerta de Soto again begs the question. He assumes what he wants to show, namely that there is no such thing as a demandable debt contract. In a table (p. 156) that summarizes his argument, he asserts that even if the terms of a demandable debt contract were considered internally compatible (case 4), "the contract is null and void because it is impossible to carry out (without a central bank)". "Impossible", I gather, means that the probability of a breach of contract is not zero, absent deposit insurance or a government lender of last resort. But why should a non-zero probability of breach render a contract null and void?

In his discussion of asset sales under repurchase agreements (pp. 118, 157), which contracts he rejects as "an evasion of the law", the author appears to be under the mistaken impression that the party agreeing to repurchase the asset at a set price is necessarily agreeing to repurchase it on demand or "whenever requested". Needless to say, that is far from typical. (In fact, I am not aware of any such contracts.) The typical agreement has a fixed date of repurchase. It is essentially a term loan in which ownership of the collateral changes hands for the duration.


Musgrave said...

I’ve read do Soto’s book, and went to a lecture by him at the London School of Economics a few months ago. I agree that his points about warehousing contracts etc. are a waste of ink and paper.

His best argument against fractional reserve (FR) is in his Ch 4 which you don’t seem to have reached yet. This is that FR enables private banks to create money “ex nihilo”, which they do big time during booms or periods of “irrational exuberance”. And that just stokes the boom.

Then during the inevitable bust, banks do the opposite, that is destroy the money they have created or “deleverage”. And that exacerbates the recession. This is exactly what happened in the last two or three years.

For that reason, I agree with do Soto that FR should be banned, or heavily curtailed.

David Hillary said...

de Soto's characterisation of FR banks creating money out of thin air is common from anti-FR bankers but the argument fails.

Firstly, it is clear that banks create money out of non-money assets rather than out of nothing. For a bank's demand liabilities to function as money, the following inputs are required:
1. The bank's credit standing must be recognised and accepted as good by at least a substantial number of people.
2. The bank must have the facilities and resources to provide for the redemption of its demand liabilities through the channels liability holders wish to use.
3. The bank's terms of its notes/current accounts must be competitive with other banks.

The nature of the bank's business as a financial intermediary that borrow and lends money on its own account means that any bank notes and current account balances it issues are used primarily to fund interest bearing investments. This demonstrates that the bank creates money out of its productive investment assets, using the technology of its credit standing and management systems etc. An analogy is securitisation: issuers and securitisation vehicles create securities not out of thin air or out of nothing but out of other assets combined with the technology of the securitisation vehicle/contract structure and governance.

For the anti FR banking argument about supposed economic impacts of bank issued money to work, you have to prove the quantity theory of money. I don't find it at all plausible.