All Commentary
Wednesday, September 4, 2013

Questioning Rothbard

Is fractional-reserve banking fraudulent?

Fractional-reserve banking is the practice whereby a bank retains funds equal to only a portion of the amount of its customers' deposits as readily available reserves (currency on hand at the bank plus deposit accounts for that bank at the central bank) from which to satisfy demands for payment. The remainder of customer-deposited funds is used to fund investments or loans that the bank makes to other customers. —

Would fractional reserve banking exist in a world without a central bank? Put another way: Is fractional reserve banking inherently fraudulent?

Murray Rothbard claimed, in several works, that it was indeed fraudulent. He argued that fractional reserve banking would have to be outlawed in a world without a central bank and a strict 100 percent reserve rule would have to be enforced on bank liabilities. But outlawing fractional-reserve banking, if it is not necessarily fraudulent, would actually decrease efficiency and consumer welfare.

Rothbard’s Case

Rothbard’s argument goes as follows: Money deposited in a checking account that can be withdrawn at any time represents a bailment or a safekeeping device. A redeemable IOU thus issued to the depositor represents a warehouse receipt or a property title. If the bank now lends out this money—either through issuing more IOUs or by lending out money proper—it is breaking the terms of the initial contract with the depositor. Doing so amounts to fraud or overissuing of property titles and results in the bank keeping fractional reserves.

I think Rothbard is essentially correct in the way he frames his argument. Lending out or issuing IOUs on money that has been placed with the banker with the explicit intention of safekeeping and nothing else (a storage contract) would indeed amount to fraud. But his conception of the types of possible banking contracts is too narrow. So, while his argument may apply to the kind of contract that he describes, it does not apply to other contracts that do not share the same characteristics, but nonetheless do lead to fractional reserves.

Fractional Reserve Reframed

There are contracts (and have been through history) that do not violate any of the above stipulations. For example, it is perfectly possible for clients to place money under deposit with a bank that fully discloses the fact that the money is going to be lent out further. In fact, the bank may even pay the depositor a rate of interest, while also promising redemption of money at any time. So far, there is no question of fraud involved. The bank has not hidden the fact that the clients’ money will be lent out. The client even earns a rate of return on this activity (albeit a lower rate than he would earn on a time deposit). The only difference compared to a time deposit so far is that this account comes with the added benefit of being redeemable at any time. 

The bank also issues redeemable IOUs to the depositors in the form of either banknotes or checking accounts. Since this is not an explicit safekeeping deposit or bailment, the issue of the IOU being a property title or a warehouse receipt does not exist in this case. The IOU is just that: a debt claim on the money with a special promise to pay the money back any time the depositor may want it. Hence, the question of “over-issuing of property titles” does not arise. 

Readers interested in such historical cases may refer to George Selgin’s paper, “Those Dishonest Goldsmiths,” published in The Financial History Review in 2012, as well as my paper, “Fractional Reserves and Demand Deposits: Evidence from an Unregulated System,” published in The Independent Review in 2013.

Two Places at Once

What about the argument that because depositors have full use of their money (via the banknote or checking accounts) at the same time as it is invested (lent out to borrowers, maybe via banknotes or checking account), this arrangement amounts to one thing being in two places (or having two owners) at the same time—and so amounts to fraud? 

This is not a case of fraud, either. A claim to money is not the same thing as the money itself. Now, it is true that a claim to money is capable of performing all the functions that money performs. This comes about because of the unique nature of money as a medium of exchange. Since money is demanded or held only to be given away in exchange again, it is possible that a claim to money performs the same function just as well. That is, a claim to money can be held and given away in exchange (it can circulate) while doing all the job of a medium of exchange. This is not true of consumer goods or producer goods. 

So while a claim to a car or a claim to a house is fundamentally incapable of performing the services rendered by the underlying good, a claim to money is capable of performing the function of the underlying good (money). This does not mean one thing is in two places at the same time, however. Two different things are in two places at the same time. It is merely a unique (perhaps irksome) characteristic of money as a medium of exchange—something that is to be taken as a fundamental fact or datum of the economic system, which alone cannot imply fraud. 

Free Money

If we accept, contra Rothbard, that there is a kind of fractional reserve banking that is not fraudulent, does this mean that banks operating on this basis face no limit as to how many IOUs they can print? On the contrary, banks face severe limits in a free market. So while depositors may indeed use banknotes or write checks on their accounts while their money is invested, these only act as “money substitutes” if others in the economy accept them or don’t ask for redemption immediately from the bank. Another way of saying the same thing is that the banknotes or checking accounts must first of all circulate as money. 

It is important to realize that the bank would have no direct control over this phenomenon. It would not be able simply to order others to accept its notes and hold onto them without asking for redemption. In a free market, the bank must rely on its reputation and good judgment of the underlying preferences of its customers for base money. The same holds true for debtors of the bank. If the bank proceeds to issue debt (grant loans) in its own redeemable IOUs, it must be wary of debtors asking for base money at any time or a bank run may ensue. This places a severe limit on the bank’s ability to flood the market with its own IOUs. This is not the case in our current banking system. The central bank, fiat money, federal deposit insurance, and other government guarantees ensure the widespread circulation of bank-created or inside money.

In conclusion, is fractional reserve banking fraudulent? It certainly can be if it arises from contracts meant only for safekeeping, as described by Rothbard. However, this does not imply that fractional reserve banking is always a fraudulent activity. In a free market for money, we can and must expect fractional reserve banks of the second kind described above to exist and to serve customers. One may also expect strictly 100-percent reserve banks that cater to those interested in only keeping their money safe at all times. To outlaw all fractional-reserve banking in a free society without a central bank would be a mistake.


  • Malavika Nair is an Assistant Professor of Economics in the Johnson Center for Political Economy at Troy University. She is also an associated scholar of the Ludwig von Mises Institute.