30 December 2009

Prof. Lawrence H. White responds to Jesus Huerta de Soto on Banking

Thanks to Prof. Steven Horwitz who emailed me a series of responses to Jesus Huerta de Soto's major work against fractional reserve banking, I am pleased to re-publish them. The original site these are supposed to be found on does not seem to be up anymore. Here is part 1 of 3.


HUERTA DE SOTO'S CASE AGAINST FRACTIONAL RESERVES

Part I.

Monday, January 08, 2007


Some defenders of honest money - most notably a group of economists affiliated with the Mises Institute - think that it is not enough to insist on a commodity standard with competitive private issue of coins and notes. They think that banks must be prevented from backing their checking accounts and redeemable paper notes with anything less than 100-percent reserves, even if bank customers want and agree to fractional reserves. The most elaborate exposition of this view is a book of more than 900 pages by a Spanish economist, Jesus Huerta de Soto. The book is now available in English: Money, Bank Credit, and Economic Cycles, translated by Melinda A. Stroup (Auburn, AL: Ludwig von Mises Institute, 2006). The entire text is >available online at the Mises Institute website.
Professor Huerta de Soto's argument against fractional-reserve banking, so far as I understand it, boils down the following syllogism:
1. Warehousing and fixed-term lending are legitimate.
2. Fractional reserve banking is neither warehousing nor fixed-term lending.
3. Therefore fractional reserve banking is not legitimate.
This is an example of what logicians call the fallacy of denying the antecedent. The two premises are correct, but the conclusion does not follow from them. Another example would be: (1) A dog has four legs. (2) A cat is not a dog. (3) Therefore a cat does not have four legs. In the first chapter of his book, Huerta de Soto spells out the features of warehousing contracts, and contrasts them with the features of loan contracts. He does not succeed in showing that a warehousing contract has any necessary relevance to banking, much less that it is the only allowable kind of banking contract. That fractional-reserve bank accounts are not warehousing contracts does not imply that they are contrary to nature or illegitimate.
Huerta de Soto begins by distinguishing a warehousing contract from a loan contract. In a loan contract, he notes, the borrower "is obliged to pay interest, as long as an agreement has been made to that effect" (p. 3). So it would seem that loan repayment terms may vary, and the terms in a particular case may be set by mutual agreement.
But for Huerta de Soto there is an exception to freedom of loan contract. The term to repayment of the loan apparently cannot be stipulated as "until one party chooses to terminate the loan", even if both parties agree up front to such an arrangement. The repayment date must be definitively fixed at the outset (pp. 3-4):
In addition, a fixed term is an essential element in the loan or mutuum contract, since it establishes the time period during which the availability and ownership of the good corresponds to the borrower, as well as the moment at which he is obliged to return the tatundem [i.e. to repay the amount owed]. Without the explicit or implicit establishment of a fixed term, the mutuum contract or loan cannot exist.
[Emphasis in the original]
Huerta de Soto's insistence on a "fixed term" implies that loans with prepayment options (where the borrower has an option to terminate the loan early) or call options (where the lender has an option to terminate the loan early) cannot exist. Prepayment options are standard on home mortgage loans and student loans. The common checking account is (at least to all appearances) a type of callable loan where the lender is the account-holder and the borrower is the bank. The claim that such loans cannot exist was not a slip of the pen. Huerta de Soto later declares (pp. 17-18) that "it is impossible to imagine a monetary loan contract without a fixed term". Loan contracts with prepayment or call options are apparently unknown to his imagination, or for some reason do not qualify as loan contracts.
Huerta de Soto denies that a "deposit" is a type of loan, but in his terminology this only means denying that a warehousing contract is a loan. Because a depositum (in Latin) was a custodial or safekeeping or storage contract, what he calls "deposits" are by definition warehousing or storage contracts "by which one person - the depositor - entrusts to another - the depositary - a movable good for that person to guard, protect, and return at any moment the depositor should ask for it" (p. 4). (Of course, "at any moment" is an overstatement: warehouses typically limit withdrawals to normal business hours.) To avoid ambiguity between Huerta de Soto's terminology and the ordinary modern meaning of "deposits", I will herein avoid the term "deposits" and speak instead of "warehousing contracts" on the one hand and "bank accounts" on the other.
The simplest ("regular") warehousing contract calls for the safekeeping and return-on-demand of a specific and identifiable item. Huerta de Soto gives the pertinent examples of "a painting, a piece of jewelry, or a sealed chest full of coins". A storage contract for fungible goods may be more complex, as he explains. The most economical arrangement for storing wheat may be a large silo where grain goes in at the top and out at the bottom. Olive oil may be most economically stored in a large tank with other olive oil. In such cases, by mutual agreement, "the goods deposited become indiscernibly mixed with others of the same type and quality" and consequently the storage contract does not call for the return of the specific items (grains of wheat or molecules of oil) deposited. Instead the contract calls for the return of equivalent stuff (wheat or oil of the same grade and in the same amount). Huerta de Soto calls this an "irregular-deposit" contract.
Given the logistics of storage in silos and tanks, mutual agreement to an "irregular-deposit" contract for wheat or olive oil is understandable. But Huerta de Soto fails to explain what compels people to store money in a vault on such terms. He lists the case of coins "in a banker's safe" as a case where indiscernible mixing also takes place, but doesn't adequately explain the advantage to the bank customer of allowing such mixing. Suppose for a moment (contrary to medieval experience) that silver coins are perfectly fungible. Why would the customer or the vault-owner want the vault-owner to take the coins out of the bags in which they came to the bank? What would be the advantage of mixing them? It is difficult to see how mixing the coins would lower the total cost of warehousing. Huerta de Soto asserts that it would be "very costly" to "place bills in an individually numbered, sealed envelope" (p. 6), but this is not very persuasive, given that warehousing of coins in sealed bags was practiced in the earliest days of banking. If a customer who brings in coins only wants them stored, he has no apparent reason to agree to mixing. Unmixed storage in sealed bags or envelopes avoids the need for counting the coins on deposit and withdrawal. My own view is that bankers historically mixed coins in the vault in order to economically provide services other than warehouse storage.
Huerta de Soto takes it for granted that coins are perfectly fungible, but silver coins in the Middle Ages were in fact commonly debased (by the sovereign mint) or underweight (due to deliberate shaving or "sweating", or honest wear-and-tear). The standard symbol of a medieval banker was a balance scale, because bankers spent so much of their time weighing coins. The image of a banker weighing coins is also familiar from medieval art. The non-uniformity of coins made weighing (rather than merely counting) necessary. It also made claims on a trusted bank (drafts or checks or banknotes), denominated in pure weight units (called "ghost monies" by historians because the units were not embodied in actual coins), and redeemable for the corresponding weight of actual coins, a more uniform medium of payment. Greater uniformity was a major advantage of bank claims over coins (the other major advantage being greater portability), and thus a major rationale for "depositors" to bring coins to various banks in exchange for claims on those banks. If Huerta de Soto were to recognize the facilitation of payments as part of the rationale for opening a bank account, however, he could no longer declare with certainty that "its fundamental purpose is the custody or safekeeping of the good" (p. 4). If we cannot presume that safe storage is a bank customer's "fundamental purpose," then we cannot presume that the customer would never knowingly agree to anything but a warehousing contract.
In a footnote (p. 6, n. 6) where he tries to explain the advantage of mixing coins in the vault, Huerta de Soto actually admits that the customer who brings coins to a bank may want something other than storage: "In the specific case of the monetary irregular deposit, the occasional use of cashier services offered by banks is an additional advantage." It is not clear what type of "cashier services" he has in mind, but to be relevant here the services must be linked to being an account-holder (which rules out simple coin-changing). Suppose he has in mind the sort of payment services provided by a checking account or banknote. If the use of payment services is one advantage, why can't it be the main advantage sought by a customer opening a bank account? If it is, then the bank need not act as a warehouse (keep 100-percent reserves) to fulfill his fundamental purpose. The customer may rationally agree to a non-warehousing contract in order to receive lower-cost payment services. Unlike a fractional-reserve bank, a money warehouse cannot offer a checking account without storage fees. It cannot offer an anonymously transferable bearer note (an ordinary banknote) because it needs to track its claim-holders in order to assess storage fees. (I make this point at greater length here.)
Huerta de Soto states (pp. 4-5) that "[t]he deposit of fungible goods is definitely also a deposit [as defined above, i.e. a warehousing contract], inasmuch as its main element is the complete availability of the deposited goods in favor of the depositor, as well as the obligation on the part of the depositary to conscientiously guard and protect the goods." (It would have been more precise to say "in favor of the depositors" - plural - here, because no particular item is completely available to any particular depositor. If I bring in two bushels of wheat and agree to get back two equivalent but different bushels of wheat, the goods that I brought in do not remain completely available to me. Someone else may withdraw them.) We can regard this statement as true by definition, insofar as it refers to warehousing contracts. But to regard it as true of bank accounts - as when Huerta de Soto writes (p. 18, n. 20) of "the essential, unmistakable purpose of the checking account or 'demand' deposit contract: to keep the tatundem continually available to the depositor" - is entirely to beg the question of the nature of the contracts between banks and their customers. We need to consider whether people who have opened checking accounts might have wanted and agreed to something other than a warehousing contract, such as a callable loan contract.
Huerta de Soto (pp. 15-16) points out quite rightly that in a warehousing contract, "interest agreements are contra naturam and absurd." A warehouse cannot pay interest. But this only means that if we observe interest being paid on a bank account, then that account represents something other than a warehousing contract. It does not mean that interest on bank accounts is contrary to nature and absurd. In a table (p. 19) Huerta de Soto contrasts the economic and legal features of a money warehousing contract with those of a fixed-term money loan. This is interesting, but it does nothing to establish that these two contracts exhaust all possibilities. It doesn't show there can't be such a thing as a callable loan to a bank, or that a checking account can't be such a loan.
Huerta de Soto cites Roman authorities on the legal nature of money "deposits". He quotes (p. 28) the following passage from a jurist named Paul in support of the view that a bank may only act as a money warehouse: "if a person deposits a certain amount of loose money, which he counts and does not hand over sealed or enclosed in something, then the only duty of the person receiving it is to return the same amount." I may be missing something, but I read "only duty" as denying rather than affirming that the recipient has an additional duty to hold 100-percent reserves.

4 comments:

Lee Kelly said...

Thank you for putting this online.

Anonymous said...

Any system that does not follow the laws of energy system or ecosystem - "the nature" is fraud and will fail. This is what we are seeing with modern fractional banking system, systemic failure.

What angers the people of occupy wall street movement is the way the banking system failure is handled by the government, instead of nationalizing the banks they bailed them out, continuing the same failed system it is like rewarding failure.

David Hillary said...

part 2 is here: http://www.lostsoulblog.com/2010/01/prof-lawrence-h-white-responds-to-jesus.html

David Hillary said...

part 2 is here: http://www.lostsoulblog.com/2010/01/prof-lawrence-h-white-responds-to-jesus.html