Why Fractional Reserve Banking Poses a Threat to Market StabilityTags Money and BanksMonetary TheoryMoney and Banking
On April 16, 2018 I participated in a debate with George Selgin, hosted by Gene Epstein’s Soho Forum. I was arguing in the affirmative for the resolution: “Fractional reserve banking poses a threat to the stability of market economies.” The full video is available here (and note that some of my remarks allude to the opening comedy from Dave Smith before the debate). Rather than summarize the live debate, in the present post I’ll lay out the position I took and how it was influenced by my study of Joe Salerno’s writings on the topic—in particular Chapter 5 of this collection.
Instability, not Fraud
In order to focus the debate on the issue of economic instability, I purposely avoided all talk of “fraud.” Although Block and Davidson (2011) for example think that the ethics of fractional reserve banking (FRB) were key in evaluating the merits of the practice, I thought most people have already made their minds up on this score.
In contrast, even many self-described Austrians do not realize that Ludwig von Mises himself, in his mature writing in Human Action (but also throughout his career, as Salerno documents in the chapter I linked above), quite explicitly declares:
The notion of “normal” credit expansion is absurd. Issuance of additional fiduciary media, no matter what its quantity may be, always sets in motion those changes in the price structure the description of which is the task of the theory of the trade cycle. Of course, if the additional amount issued is not large, neither are the inevitable effects of the expansion. [Mises 1949, fn 17, p. 439]
Thus we see that for Mises, what we now call the “Austrian theory of the business cycle” (and which he originally developed as the “circulation credit theory of the trade cycle”) was not a theory of how central banks cause unsustainable booms. Rather, it was a theory of how commercial banks cause unsustainable booms through the issuance of fiduciary media, i.e. claims to money that are not backed by the banks’ holdings of actual money in the vaults.
This is the critical issue, and there is nothing mysterious about it. In his 1912 book The Theory of Money and Credit—which Guido Hülsmann says would have been better translated as The Theory of Money and Fiduciary Media, see pp. 32-34 here—Mises develops the notion of a “money substitute.” This is a claim on (base) money that is immediately payable at par, and about which the community has no doubts. The significant fact about money substitutes is that they perform the same services as money, and as such can be held in people’s cash balances as if they were money proper. (Here are links to free versions of Mises’ classic work, and here is my Study Guide for it.)
For example, back when gold coins were the underlying base money, if someone held a paper note issued by a reputable bank promising the bearer one gold ounce upon demand, then that note was “as good as gold” economically because most people would accept it at par. Any individual holding that paper note could obtain goods and services in the market just as easily as he could with an actual gold coin.
Likewise, in our day if you have a debit card issued by Citibank that effectively tells merchants, “This person has $400 in his checking account with us,” then that is “as good as dollars” in most stores. When you buy groceries by swiping your debit card, you aren’t handing over actual money in the form of green pieces of currency, instead you are handing over claims on dollars that are issued by your bank.
Now within the class of money substitutes, Mises made a further distinction. Those money substitutes fully backed by money in the bank vault were called “money certificates.” They were simply claim tickets that took the place of the original base money. There was no economic impact of a person depositing his money with a banker, and carrying money certificates instead, except for the increased convenience (which is why the individual would do it).
On the other hand, those money substitutes that were issued above and beyond the reserves of money held by the bank were called “fiduciary media.” The issuance of fiduciary media would have an economic impact, because it effectively increased the total quantity of money held in cash balances by the community. Beyond this monetary inflation, the specific problem with newly issued fiduciary media is that it entered the economy through the loan market, meaning that the first prices it distorted were interest rates.
This is a crucial point so I’ll state it in slightly different words: Richard Cantillon famously explained that when new money enters the economy, it does not simply raise all prices instantly by the same percentage. On the contrary, “Cantillon effects” refer to the redistribution of wealth as new money cascades out into the economy, passing from sector to sector. For example, if a medieval king debased his currency and created new coins (with a lower gold content), then spent them on obtaining horses, the first prices to rise would be those of horses. Then the horse breeders would spend their new coins on (say) fine restaurants and silk clothes, raising prices in those areas. Then the waiters at the restaurant might receive bigger tips, and so on around the economy.
In this context, Mises pointed out that commercial banks engaged in fractional reserve banking effectively create money “in the broader sense” (i.e. including not just base money but also money substitutes) and that this money enters the economy through the issuance of new loans. Therefore, unlike (say) new gold flowing into the economy through the spending of mine owners which would be spread across multiple sectors, new money being created through fractional reserve banking is focused entirely on the loan market. Therefore it pushes down interest rates to artificially low levels; they are the first prices to be bombarded with the “credit expansion.”
And as Mises’ quotation from above indicates, he thought this phenomenon occurred with any issuance of fiduciary media. This makes sense, once one understands the nature of fiduciary media. Contrary to the claims of Selgin (which I can’t elaborate here), even in scenarios where the public wants to hold more banknotes, the creation of new money (in the broader sense) entering via the loan market will still cause the interest rate to deviate from its proper level.
But Wasn’t Mises a Free Banker?
My above writings may confuse some readers, who could understandably have thought that Mises was clearly in the camp of the “free bankers” like Selgin and Larry White. After all, Mises explicitly says in Human Action that only the policy of free banking can contain the boom-bust cycle.
The confusion here is due to a difference in opinion about what would happen under a regime of “free banking,” wherein the government gives no special privileges to the commercial banks. Today’s writers in the free banking tradition (such as Selgin and White) think that in equilibrium, the profit motive will lead commercial banks to adopt a reserve ratio that could be quite low. So long as this ratio reflected the desires of the public to hold “inside money” and the frequency with which bank-issued deposits were redeemed for base money, Selgin and White believe that fractional reserve banking is quite healthy for the community.
In complete contrast, Mises advocated free banking because he thought it would restrain the issuance of fiduciary media. For those readers who have never entertained such a view, I urge you to read Salerno’s Chapter 5 here. Once you realize there is a distinction between “favors free banking” and “thinks fractional reserve banking promotes stability,” all of Mises’ writings on this topic become crystal clear. Mises only seems to vacillate when the reader assumes that anybody in favor of free banking must necessarily endorse low reserve ratios.
In the grand scheme, the actual positions of myself versus Selgin are quite close. We both want to remove politics entirely from money and banking, which obviously means getting rid of the Fed but it also means (so long as we have State courts) no legislation restricting fractional reserve banking.
However, where we differ is in our predictions and evaluations of the results of such a “free banking” policy. Like Mises, I predict that absent government privileges—such as allowing the Scottish banks to renege on their contractual obligations for more than two decades (!)—banks would be kept on a very tight leash. In contrast, Selgin had no problem (as he admitted during the debate Q&A) with banks having reserve ratios as low as 3%.
In closing, let me offer an analogy. I think that government ownership of roads causes far too many traffic accidents. Under a regime of “free roading,” the number of traffic accidents would plummet, but it wouldn’t hit literally zero simply because the cost of wringing out the last few accidents would be higher than the benefits.
At the same time, Selgin surely thinks that government subsidies currently lead to too much corn being planted in the U.S., and that a regime of “free agriculture” would lead to a lower equilibrium amount of corn. Yet Selgin wouldn’t view the new, lower amount of corn as a necessary evil; he would think it was a boon to consumer welfare.
Thus, even though my views on traffic accidents and Selgin’s view on corn are superficially similar—in that we both think a move to the free market would reduce the amount of each—there is a sense in which my views on traffic accidents are far different from Selgin’s views on corn production.
This contrast is analogous to our views on the same issue, namely fractional reserve banking. Selgin and I both favor a free market in banking, and we both agree that FDIC and the Fed’s “lender of last resort” policy subsidizes credit expansion. Yet I would view any remaining fiduciary media in a genuinely free banking system as a regrettable evil (that caused an attenuated boom-bust if large enough), whereas Selgin would view it as a healthy boon to consumer welfare.