Full Reserve Banking

CHAPTER 4

The (Feketian) Banking-School
Approach Cannot Prevent Cycles

In this section I will explain why the variant of the banking school that is defended by Rallo will not prevent business cycles. For this endeavor I have to analyze the nature of real savings that are necessary to sustain productive processes. I must then analyze some basic features of capital theory. Furthermore, I will analyze how the market for real bills affects the whole structure of interest rates and long-term investments. Finally, I will look at the supposed limit for credit expansion in the needs of trade.

4.1 The Demand to Hold Money Is Not Saving

4.1.1 The Essence of Fiduciary Media: Present Goods or Future Goods

Probably the most important theoretical disagreement between Rallo and the Austrian school is about the role and nature of real savings and its relation to the demand to hold money. This question is connected to the nature of fiduciary media, which I will discuss in this section.

One of the contributions of Mises in his 1912 work The Theory of Money and Credit was to enhance Carl Menger’s analysis of money and provide a new subjectivist typology of money (Hülsmann 2012). According to Mises (1953) money in its broader sense may be categorized into money proper and money substitutes. In a gold standard, gold is money proper. Money substitutes, such as demand deposits or bank notes, can be exchanged against money proper (such as gold) on demand. Money substitutes are perfectly redeemable claims on money (proper). There are two types of money substitutes: money certificates and fiduciary media. Money certificates are money substitutes that are completely backed by the issuer by money proper. Fiduciary media are money substitutes that are not backed by money proper.

In other words, fiduciary media are money substitutes that are issued beyond the amount of money certificates. When someone deposits money proper at a bank, the bank issues a money certificate. Fiduciary media are created by the issuing bank when it grants a circulation credit. Whereas with a commodity credit someone gives up purchasing power to the benefit of someone else, with a circulation credit and the issue of fiduciary media it is different: “The gain of the party who receives before he pays is balanced by no sacrifice on the part of the other party” (Mises 1953, 264). As there is no sacrifice, fiduciary media are created out of thin air (1953, 306).

Mises, who was trained as a lawyer, was influenced in his analysis of claims and titles by his academic teacher Eugen von Böhm-Bawerk (1881, 120). As money substitutes are claims payable on demand, they are considered present goods. Money substitutes are regarded as certain claims. They are redeemed at par; they do not trade at a discount. Perfect money substitutes (i.e., demand deposits) and money proper fulfill the same services of reducing uncertainty and facilitating transactions.

When money substitutes are issued, there is no credit transaction, because the issue of a title (perfect money substitute) to a present good (money proper) is not a loan. Things are different for loans, which are credit transactions. In a credit transaction, there is a transfer of goods delivered in the future. As Böhm-Bawerk (1881, 5–11) clarifies, a credit transaction involves a title to a still nonexistent future good.

The distinction between present goods and future goods is important to Mises and essential for the debate on real bill fractional reserve banking. For Mises a demand deposit (fiduciary medium) is a title (or claim) to a present good, while a bond and a real bill are titles to future goods. Rallo, however, does not apply the Misesian distinction between present goods and future goods. Rather, Rallo distinguishes between real assets, which are those assets that do not represent a liability of someone else, and financial assets, which do represent a liability of someone else (96).

For Rallo both demand deposits and real bills are financial assets, while for Mises the former is a claim to present goods and the latter a claim to future goods. Rallo’s conceptualization of real assets and financial assets, which may be useful in other contexts, does not allow him to see the differences between a title to present goods and a title to future goods: his focus on the distinction between real assets and financial assets hides the problems of fractional reserve banking.

A more nuanced view holds that there are three kinds of property (present goods)—namely, consumer goods, producer goods, and money. Moreover, there are titles (claims) to future consumer goods, future producer goods, and future money (Hoppe, Hülsmann, and Block 1998, 24). For Rallo there are only real assets and financial assets. The important difference between titles to present goods and titles to future goods remains unnoticed in Rallo’s conceptualization. Is there no essential difference between a car valet ticket (title to a present good) and a bond (title to a future good)? For Rallo, both are financial assets. This lack of distinction is especially problematic when titles to present goods for practical purposes exercise the same function as the present goods themselves, as is the case for fiduciary media and money proper.

As a consequence, Rallo has a completely different assessment of fiduciary media, which is at the root of many discrepancies. For Rallo fiduciary media are not present goods and perfect money substitutes but just very liquid financial assets. From there it is only a short, fatal step to his claim that people holding fiduciary media are basically investing in a financial asset and, hence, saving. The issuance of additional fiduciary media against real bills, for Rallo, thereby does not pose a problem, as there is real savings on the part of those who hold the fiduciary media.

Rallo cannot accept the possibility of perfect money substitutes used as present goods, because then the question of their backing with real savings would come to the fore. Perfect money substitutes are defined away because otherwise his whole argument would collapse. As “perfect money substitutes” must disappear in Rallo’s world, they are categorized under a new name: “financial assets.”

Apart from definitions, there is another question about essence at play. What is decisive in the matter of the nature of fiduciary media is the subjective point of view. The concept of perfect money substitutes is based on subjective thoughts in the same way that the law of diminishing marginal utility is. When, from the point of view of the actor, individual units of a good are perfectly interchangeable, the law of diminishing marginal utility applies. The law states that when the number of units of a good considered as perfectly interchangeable and at the disposal of an actor increases, the marginal utility of each of these units individually considered falls. These units are not perfectly interchangeable from an objective point of view. What makes the law of diminishing marginal utility applicable is not objective characteristics of goods but the subjective consideration of the actor. The same applies to money substitutes.

But can units be perfectly interchangeable or can titles to goods be perfect money substitutes? Indeed, there is nothing perfect in this world from a metaphysical point of view. But no sophisms can change the fact that subjectively, units can be considered perfectly, or—if you prefer—completely, interchangeable. An individual may consider two $5 notes and one $10 note as equivalent and perfect substitutes. In the same way, money substitutes can be subjectively considered perfectly interchangeable with money proper, at least at a specific moment in time and at a specific place. A $10 note can be considered perfectly interchangeable with $10 in a demand-deposit bank account. Indeed, in their accounting practices, companies consider cash and money in bank accounts as one entry—perfect substitutes.

What is essential here is that the $10 note and the $10 in the bank account may form part of what the individual considers subjectively to be his cash balance (Hülsmann 2000, 107). When the individual makes a decision to buy or not buy, to sell or not sell, a good or service, he takes into account the cash balance he subjectively considers as available; and this cash balance includes both money proper and perfect money substitutes. Perfect money substitutes are (objectively) used as if they were money proper.

Therefore, there is an essential difference between creating some secondary medium of exchange—for instance, someone issues shares in the stock market, which are very liquid—and creating perfect money substitutes that are unbacked by money proper (i.e., fiduciary media). Both are financial assets from Rallo’s point of view. However, there are crucial differences. In the latter case, the money supply, comprising perfect money substitutes and money proper, increases. There is a double availability. The person who holds the money proper believes it to be available to him, and the person holding the perfect money substitutes considers money proper available to him as well. No such double availability occurs in the case of an issue of shares in the stock market.

Now, let us further examine whether fiduciary media are financial assets. A financial asset implies an exchange of present goods against future goods. In a bond, for instance, the lender gives up present goods (money available today) against future goods (money available in the future). Similarly, a stock is a financial asset. The purchaser of the stock gives up money and gets in return the right to dividends in the future. Money, however, is neither a bond nor a stock; it is not a financial asset. Money is a present good that gives services in the present—namely, facilitating exchange and reducing uncertainty.

Uncertainty reduction is the key for understanding money’s nature and function. Without knowledge of the timing and the magnitude of future expenses, individuals demand money in order to hedge against that uncertainty. Money proper and money substitutes differ from financial assets, as they represent on-demand availability at par value (Howden 2022). With financial assets such as stocks or bonds there remain residual risks. Thus, the demand for money is the demand for present goods to reduce uncertainty and satisfies a different need from the demand for financial assets, which is a demand for future goods expressed by giving up present goods.

When an individual demands money or money substitutes by selling some present good, there is no exchange of present goods against future goods; there is no credit transaction. Money must be available in the present to render its services. Now, fiduciary media are perfect money substitutes. They are accepted in exchange as if they were money proper. They exchange on par with money proper. And indeed, in common statistics, both cash and demand deposits are included in M1 because they perform the same economic function. In short: as money is not a financial asset, and fiduciary media are perfect money substitutes, fiduciary media are not financial assets.

People hold fiduciary media because they are perfect money substitutes and perform the same function as the present good called money.

As Jeffrey Herbener (2002) puts it:

People only demand money-substitutes, not fiduciary media, and their demand exists only when they have confidence in full redemption based on the issuers’ practice of full redemption. People could not demand fiduciary media because they cannot distinguish between a money-substitute that is a money-certificate and one that is a fiduciary medium. If they could make such a distinction, then fiduciary media would not be viable. (83)

While fully redeemable bank money is a perfect money substitute, a real bill is not, according to Mises (Rallo, 27). It may be a secondary medium of exchange. Indeed, holdings of real bills do not serve a bank in a bank run. While real bills are redeemable in 30 days (or 60 or 90), in a bank run depositors want their money right now (Humphrey and Timberlake 2019, 29).

Again, the question of the nature of fiduciary media is important because of the role of savings. If fiduciary media were financial assets, implying an exchange of present goods against future goods, the holders of fiduciary media would be saving, and credit expansion by real bill fractional reserve banks would be backed by real savings. If, however, as I have demonstrated, fiduciary media are perfect money substitutes and act as present goods, then no one is giving up present goods and saving by holding bank money. I will analyze this issue in more depth in section 4.1.3.

4.1.2 A Short Legal Excursion

Connected to the question about the essence of fiduciary media—whether they are financial assets or perfect money substitutes—is the legal issue of fractional reserve banking. The legal issue is only indirectly touched upon by Rallo. Nevertheless, it is quite important.

Rallo sees a loan contract not as an exchange of present goods against future goods but rather as a relationship between creditors and debtors. A credit is a collection right or collection claim (128). For Rallo, fiduciary media such as bank notes or bank accounts are “on-sight or on-demand credits.” Therefore, in real bill fractional reserve banking there is no double availability of money proper, because the fiduciary media constitute a collection right only at the moment that someone demands to redeem them into money proper—not before or afterward. Therefore, the bank only has to have the reserves at the moment the holder of the right asks for collection.

Neither in Human Action (1949) nor in The Theory of Money and Credit (1953) does Mises write about the legal issues surrounding fractional reserve banking. However, Mises (1953, 268) remarks, “The depositing of the money in no way means that he has renounced immediate disposal over the utility that it commands.” This remark indicates that Mises believes that there is a double availability involved in the issue of fiduciary media. The problem of this double availability is at the core of the legal problems of fractional reserve banking.

Even if we assume that banks do issue “on-sight credits” with collection rights and not demand deposits, as long as these “on-sight credits” are considered perfectly available and equivalent to demand deposits, all traditional legal principles that apply to demand deposits must also apply to “on-sight credits.” Changing the name of demand deposits to “on-sight credits” does not change their nature. Legal principles apply to contracts that are equivalent even though different names are used. For instance, a contract that states that if it rains 25 percent of days there will be a payment of X is equivalent to a contract that says that if it rains 1 out of 4 days there will be a payment of X. The same principles apply to both contracts.

An “on-demand credit” that gives the holder the right to collect on demand is equivalent to a demanddeposit contract. When the holder always has the right to collect or withdraw his money at any time, and believes he is able to do so, it forms part of his cash balance, not only at the moment he withdraws his money but always. As such, fiduciary media are money substitutes and used as means of payment. Therefore, there exists a double availability at the moment fiduciary media are created. The holder of the fiduciary media believes the money proper is available to him, as he has the right to collect, and the bank that uses the money proper believes it is available for its investment purposes—namely, to grant credit.

4.1.3 Holding Financial Assets and Savings

Holding a fiduciary-media cash balance is not like holding financial assets such as stocks or bonds. When someone buys stocks and bonds, he is giving up purchasing power. When someone holds fiduciary media, he is not giving up purchasing power but holds a present good. There is a double availability because both the holder of fiduciary media and the bank act as if they had the money available.

Compare the following two scenarios: First assume that person A saves from his income and purchases a newly issued stock (a financial asset). The company takes the money to invest in a 10-year project. The saver, then, after 1 year sells the stock, and another person becomes the saver, as he reduces consumption and purchases the stock. If no one saves to purchase the stock from A, then A cannot sell the stock and cannot consume as much as he hoped. In any case, consumergoods prices do not increase. The investment project is backed by real savings and sustainable.

Now assume that person B saves from his income and adds to his cash balance, which he holds in a demand deposit—a perfect money substitute and present good. The bank takes the money and grants a loan to a company to invest in a 10-year project. When after 1 year person B spends his money, consumer-goods prices rise relative to prices of capital goods, and a new project that was undertaken to last 10 years may have to be abandoned as shorter projects become relatively more profitable with the rise in consumer prices.

While financial assets arise in credit transactions and holding financial assets implies saving, holding a cash balance is not saving. This error can already be found in Keynes (1964), where the author writes:

It is supposed that a depositor and his bank can somehow contrive between them to perform an operation by which savings can disappear into the banking system so that they are lost to investment, or, contrariwise, that the banking system can make it possible for investment to occur, to which no saving corresponds. But no one can save without acquiring an asset, whether it be cash or a debt or capital-goods; and no one can acquire an asset which he did not previously possess, unless either an asset of equal value is newly produced or someone else parts with an asset of that value which he previously had. In the first alternative there is a corresponding new investment; in the second alternative someone else must be dissaving an equal sum. For his loss of wealth must be due to his consumption exceeding his income. (81–82)

Keynes, on the next page, argues that the person who holds the additional fiduciary media created by credit expansion is actually saving: “Moreover, the savings which result from this decision are just as genuine as any other savings. No one can be compelled to own the additional money corresponding to the new bank credit, unless he deliberately prefers to hold more money rather than some other form of wealth” (83). Benjamin Anderson (1979, 391) comments correctly that “one must here protest against the dangerous identification of bank expansion with savings, which is part of the Keynesian doctrine.”

An increase in the amount of money units (or fiduciary media) does not equal an increase in real savings. The fact that the new monetary units change hands and are always in the cash holding of someone does not mean that they are real savings. Even if the fiduciary media are not spent immediately, that does not mean that there are additional real savings (Huerta de Soto 2012, 697).

Huerta de Soto (2012, 694, 697, 700) mentions that George Selgin’s case for fractional reserve banking is based on Keynes’s argument. Rallo partially follows the argument. Rallo does not identify all bank expansion with savings but only credit expansion in which real bills are discounted. He believes that when fiduciary media are created by credit expansion based on real bills and then not presented for redemption at banks but held, there is an increase in real savings. However, the error of Keynes, Selgin, and Rallo is the same. Holding newly created fiduciary media does not equal new saving. Saving implies prior abstention from consumption and the liberation of consumer goods. It implies giving up present goods, something that does not occur when fractional reserve banks expand credit based on real bills.

As Mises (1953, 272) puts it, “A person who accepts and holds notes, grants no credit; he exchanges no present good for a future good. The immediately-convertible note of a solvent bank is employable everywhere as a fiduciary medium instead of money in commercial transactions, and nobody draws a distinction between the money and the notes which he holds as cash. The note is a present good just as much as the money.”

Huerta de Soto (2012) reminds us, “All saving requires discipline and sacrifice of the prior consumption of goods and services, not merely the renunciation of the potential consumption afforded by new monetary units created ex nihilo” (548).

There may be, of course, forced saving as a consequence of real bill credit expansion. When the additional fiduciary media bid up prices, some individuals will be forced to reduce their consumption. Moreover, after an artificial boom started by real bill credit expansion, many capital goods will lose in value, there will be a recession, and in the recession consumption will fall. Yet not ex post forced savings but prior savings are needed for investment projects to be sustainable.

Rallo might reply that the producer that issues the real bill could already consume the goods that are used to collateralize the bill and, thus, the producer would be saving. However, this argument is problematic and in conflict with subjective value theory for several reasons.

First, the producer cannot, normally, consume all products he produces. Indeed, in an economy based on the division of labor, producers produce a large quantity of consumer goods whose marginal utility is quite reduced and then these goods are exchanged for goods and services that other specialists produce in large quantities. Indeed, in a society based on the division of labor, normally the producer does not produce the consumer goods for himself but for others.

Second, a consumer good becomes a consumer good when it is assessed as such subjectively by a potential consumer. Beforehand the good has to be produced, transported, and marketed to the final consumer. As long as it still needs transportation and marketing, it is not a consumer good yet but still at an earlier stage of production (Mises 1953, 83). It is a capital good or intermediate good. Not consuming the capital goods is required to finish a particular investment project, but it does not constitute real savings available to finance another project.

Capitalists may sell their unfinished capital goods and increase consumption at any stage of the production process. Yet for the production processes to be completed successfully, someone has to keep saving until the very end of the production process. Selling the unfinished capital goods does not increase real savings from an aggregate point of view. Nor does the discounting of bills backed by these unfinished capital goods increase real savings. When unfinished capital goods (backing real bills) are used for credit expansion, new fiduciary media are created without an increase in real savings.

It is true that capitalists save and that these savings are transformed into intermediate goods and finally into consumer goods. Yet it would be double counting to say that the savings that are converted into consumer goods not only sustain the production process of this consumer good but constitute additional savings available to finance another project. Capital goods can be said to constitute savings, but they are not available to sustainably finance further projects via fractional reserve real bill discounting.

Third, Rallo is correct that if the holders of fiduciary media asked for redemption into money proper, then an immediate credit contraction would ensue. But it is a non sequitur to deduce from this that as long as there is no bank run, there is an increase in real savings. Depositors’ not redeeming their deposits into money proper is not equivalent to their “financing” the credit through real savings—i.e., sacrificing a larger part of consumption. Take the following example: If a bank buys government bonds, creating a fractional reserve deposit for the government, then the money supply increases. The government uses the funds to buy from another bank customer (instead of withdrawing money proper), so there is no credit contraction. But that does not mean that there is additional real savings just because the bank customer holds on to the newly created fiduciary media. The real amount of goods available to sustain production processes has not increased. It is simply that new means of payment have been created.

Banks issue perfect money substitutes that are accepted as money proper. The factor owners paid with newly created fiduciary media do not have to wait to spend. They can directly use these money substitutes. They do not have to redeem them into money proper before using them. There is no renunciation of consumption at the moment of the creation of the fiduciary media on the part of the bank.

Just because someone holds on to the fiduciary media, it does not mean that real savings in society has increased. What would Rallo think about a money counterfeiter? If a money counterfeiter produces counterfeit notes and the money circulates, then the holders of the counterfeit notes will, according to Rallo’s argument, also forgo the disposal of present goods. They could have bought consumer goods with these counterfeit notes. If the holders do not use the notes immediately, it does not mean there is more real savings. And if they spend them and other people start to hold them, there is not more savings either. It is absurd to claim that the counterfeit notes in the pockets of the people “finance” the investment of the counterfeiter because they could have tried to redeem them against money proper. There is no more real savings; there is just “forced savings,” as prices tend to rise. And there is redistribution.

1.1.4 Rallo’s Reply to Sean Corrigan

In a response to Sean Corrigan (2005), Rallo (2014) insists that real bill discounting by fractional reserve banks involves saving. His argument is that the goods backing the real bill are already produced and could be consumed. Modifying one of Rallo’s examples, let us assume that the owners of the factors of production are paid with a voucher that they can exchange at the end of the production process against consumer goods. In this case, the owners of the factors of production are saving (voluntarily) during the production process. Now, instead of paying the owners of the factors of production with a voucher (real bill), the real bill is discounted and exchanged against fiduciary media, which are paid to the factor owners. The factor owners can exchange these fiduciary media for the consumer goods produced at any time. If the factor owners do not buy the consumer goods but hold on to the fiduciary media, they are, according to Rallo, effectively saving in the same way that they save when they are paid with vouchers.

While this is the most favorable case for Rallo, in which real bills are issued against consumer goods and these fiduciary media are untouched until the end of the production process, problems still remain.

First, and as stated in section 4.1.3, the production process ends when the consumer good is sold. Retail shops form part of the production process; they are the last stage of production. This last stage involves time, and the product has to be transported physically and spiritually (by marketing) to the consumers. Hence, Rallo’s distinction between two types of consumer goods—namely, goods that have been produced but not sold yet and can be used for real bill discounting and (real) consumer goods—is artificial and does not make economic sense.

Savings must be available until the very end of the production process to sustain factor owners and not only to the very end minus 30 days. Handing out fiduciary media to factor owners 30 days before the end of the production process does not shorten or speed up the production process. It allows factor owners to spend these fiduciary media before the production process ends. Indeed, if no time were gained in accessing money, then real bill discounting would indeed be redundant. Thus, there is some time gain that is not sustained by real savings.

Second, even if the factor owners do not touch the fiduciary media, their cash balances increase and they will bid for goods and services based on this increase. Maybe they use other fiduciary media or money proper for additional purchases that they would not have acquired without the real bill credit expansion.

Third, the fact that fiduciary media have to be created to sell the consumer goods shows that their price would otherwise have had to fall, or else the seller would have had to wait. Creating new fiduciary media does not increase real savings; it may just eliminate a price fall or speed up a sale.

Fourth, let us imagine that additional consumer goods are produced that the entrepreneur expects to sell for $1,000. A bank discounts a real bill for $990. New fiduciary media are created and increase the income of the factor owners. Now, the $990 is insufficient to purchase the $1,000 worth of goods. Self-liquidation cannot occur. More importantly, if all $990 of additional factor owner income were consumed in purchasing the newly produced goods, this would imply a change in the proportion of consumer and investment spending. One hundred percent of the additional monetary income would be spent on consumer goods and 0 percent on investment. The proportion of income spent on consumer goods would increase (i.e., there would be an increase in time preference). The increase in time preference, however, would require an adjustment of the structure of production.

If time preference does not change, then at least a small part of the $990 additional factor owner income will not be spent on consumer goods but saved and invested. In this case, self-liquidation of the real bill credit expansion does not occur. The additional consumer goods will not be purchased for $1,000 but must fall in price. Prices must adjust due to the economic growth. However, this is not a problem but part and parcel of a market economy. The production of fiduciary media cannot change that and is the potential source of intertemporal distortions.

4.1.5 The Need for Real Savings

Capital theory is an essential and distinctive building block of Austrian economics. Essential differences between the Austrian school and other schools of economics can be derived from their different theories of capital (Huerta de Soto 2012).

The key Austrian insight on capital theory is that production is a process that takes time. Consequently, there is a need for real savings to sustain production processes. Someone has to give up consumption in order to sustain the owners of the factors of production engaged in the production of capital goods. Saving provides present goods to sustain investment projects.

In capital markets, present purchasing power is exchanged for titles to future goods. This need for real savings to sustain production processes is not only the key difference between the Austrian school and other schools in the field of “macroeconomics” but also key to understanding Austrian business cycle theory. According to this theory there is an intertemporal discoordination between savers and investors due to credit expansion that distorts the interest rate. Investors invest as if real savings had increased and engage in too-ambitious investment projects, while consumers do not increase their real savings. Induced by credit expansion and artificially low interest rates, investment projects appear to be profitable that would not be profitable with higher interest rates that represent the real amount of savings.

Entrepreneurs taking new loans from a fractional reserve banking system start new projects, even though savings have not increased. Sooner or later the artificial boom comes to an end, as there are not enough real savings to sustain the factors of production in the new projects. It becomes obvious that there have been malinvestments, and they are liquidated in a recession.

Let us take a look at the system that Rallo proposes. He proposes a real bill fractional reserve banking system where fiduciary media can be redeemed into money proper (such as a fractional reserve gold standard).

To understand the workings of such as system, let us imagine Robinson Crusoe on his island. He catches fish with his bare hands. He catches 3 fish in a whole day, and he consumes all 3 of them. The 3 fish symbolize the consumer goods that he needs to sustain himself comfortably. The investment project that Robinson creates in his mind is to build a net to catch fish more efficiently. He believes that with his net, which is a capital good, he will be able to catch 10 fish in only half a day, thereby increasing his output. He also believes that it will take him 30 days of full-time work to build such a net. If Robinson is not willing to sacrifice consumption and to save and build a 90-day hoard of fish (assuming that they are not perishable), he will not be able to engage in his project. Someone in society could, however, save for him. For instance, Friday could sacrifice consumption and lend him 90 fish for 30 days (or some more days to give him time to catch 90 fish).

Instead of lending Robinson all 90 fish at the beginning of the project, Friday could save and lend 3 fish a day, extending the loan for 30 days. Alternatively, other inhabitants of the island could lend to Robinson. For instance, Friday could lend 45 fish for 15 days and Gordon another 90 fish (45 to pay Friday back and 45 for consumption) for the second 15 days. Alternatively, Tom could lend 2 fish and Peter 1 fish the first day for 1 day, and John 7 fish and David 2 fish on days two and three (3 to pay Tom and Peter back and 6 to sustain Robinson on days two and three), and so on. For the project to be sustainable, it is essential that 3 fish of net real savings are transferred to Robinson every day to sustain him. The maturity (quality) of the loans is not what is essential but the availability of net real savings (the quantity of fish saved and transferred to Robinson).

So far, so good. If enough real savings are transferred to Robinson to sustain him for 30 days, then the project can be completed successfully. The net is built, and society is richer, as a new capital good that increases productivity has come into existence.

Now, imagine that no one increases savings, but everyone consumes all fish that are produced. There is no net real savings. Then, the project is not viable, because Robinson has nothing to consume during the 30 days. Imagine that Friday owns a bank with some gold reserves. Now, Friday gives a loan to Robinson (collateralized by the fish he wants to catch with the net) without anyone increasing real savings. All fish that are caught are still consumed. Friday issues a money substitute (fiduciary medium) to Robinson that states that it gives the right to claim 90 gold coins from Friday’s bank on demand. The money substitute is a perfect one: it trades at par with money proper. As the market price of fish in the past was 1 gold coin, Robinson believes that he will be able to purchase the fish he needs and starts his project. However, if no one increases real savings, Robinson will not be able to purchase the fish to sustain himself. They will just not be there. The creation of fiduciary media, even though “backed” by consumer goods brought to the market within 90 days, does not create fish today. As a consequence of Robinson’s bidding, fish prices may go up, sending him the signal that it was an error to engage in his investment project.

Robinson may actually issue a real bill, and sell it to Friday’s bank in exchange for 90 newly issued gold fiduciary media. If market participants accept them as perfect money substitutes (and this is the case that Mises [1953] discusses), then these fiduciary media circulate and someone will always hold them. Maybe the fiduciary media get to Tom, and later to Peter. Both could ask for redemption into gold. However, let us assume that they do not ask for redemption. Even though Tom and Peter do not redeem their fiduciary media, the circulation of them does not create any new real savings to sustain Robinson. It only drives up prices. So it is an error to argue, as Rallo does, that the holding and acceptance of the fiduciary media is saving. As we have seen, it just does not create more fish.

Let us modify the example to highlight another point. Imagine that Robinson receives 90 units of fiduciary media from Friday’s bank in exchange for a real bill. He buys 90 fish from a hoard that Tom has. Note that Tom is not lending for 30 days; he only sold his stock against a perfect money substitute. He did not pledge to restrict his consumption for 30 days (as he could have done with a loan contract). Only when Tom (or someone else) reduces his consumption by 90 fish for 30 days will the project be viable. In other words, if all newly created fiduciary media will be saved, the project is viable. Only if the fiduciary media created will not be used for additional spending will the project be viable. If Tom only held the fiduciary media and did not increase consumption, this would imply that savings rates had risen, that time preference had been reduced. Only in this rare case would the creation of fiduciary media, the discounting of the real bills, lower time preference.

As Huerta de Soto (2012, 549) points out, once the newly created funds are invested and get to the owners of the factors of production, the owners have to save the funds completely in order to make the investment sustainable. If they spend part of the fiduciary media on consumer goods, there will be a recession. As Huerta de Soto (2012, 552) puts it, “Economic agents will ultimately have to save absolutely all monetary income derived from the new investment” (emphasis original).

However, if time preference is constant, then economic actors will spend the additional funds. Tom will not restrict his fish consumption. Then the project is not viable. And again, holding additional money substitutes does not equal a decrease in time preference. If Tom uses the fiduciary media during the 30 days to buy a consumer good from David, and David sells the fiduciary media to John to buy another good, then consumer-goods prices tend to increase, indicating the lack of real savings available to finance all investment projects currently undertaken.

It is quite simple: the creation of fiduciary media does not increase voluntary real savings, which are necessary to finance additional projects. The creation of fiduciary media does not cause a reduction of consumption that frees goods to sustain Robinson during the production process, even though the fiduciary media are held by actors and circulate.

4.1.6 Cash Holdings versus Flow of Savings

The crucial error in Rallo’s theory is not taking into account that cash holdings are a stock variable and savings are a flow variable.

The flow of savings is the portion of the produced consumer goods that are saved (Huerta de Soto 2012, 697) and made available to the owners of the factors of production that are producing capital goods. Economic actors save part of their monetary income, thereby sacrificing enjoyment of consumer goods. In other terms, savings is the portion of income that is not used in current consumption (Machlup 1940, 212). This savings— i.e., unconsumed real goods—is always and everywhere a flow variable.

Cash holdings are, however, a stock variable. In an economy with a constant money supply, this stock variable obviously is constant as well (Bagus and Howden 2011b). When the money supply increases, cash holdings—that is, the stock variable—increase. Yet this increase in cash holdings does not say anything about the portion of produced goods that are saved and available to sustain production processes. Adding a zero to all bank notes and bank accounts increases the stock of cash holdings (“savings” in everyday language) but does not increase real savings.

In other words, an increase in the stock variable of cash holdings does not imply an increase in the flow variable of savings (real goods not consumed). Indeed, cash balances may be increased by reducing saving and investment spending. The flow variable of savings is determined by time preference. The lower the time preference, the higher the proportion of income saved and the higher the amount of goods available to sustain investments. The stock variable is determined by the demand for and supply of money.

In relation to the real bills doctrine, it is important to understand that it is the flow of savings that is used to sustain factors of production in the capital-goods sector. This flow of funds (sometimes also called the “subsistence fund”; see Strigl 2000) does not grow by discounting real bills. The discounting of real bills by fractional reserve banks only leads to redistribution. It changes the persons who get access to the subsistence funds. Through the creation of fiduciary media, the holders of fiduciary media get greater control of the subsistence funds than they otherwise would have.

In order to understand better the relation between cash holdings and the flow of savings, it is useful to emphasize that there are three ways in which individuals can spend their monetary income: they can spend it on consumption, they can spend it on investments, or they can add to their cash balances. Now, individuals can add to their cash balances (stock of “savings”) in three ways: they can disinvest, reduce consumption, or reduce both investment and consumption spending.

Imagine an economy where people hold cash balances of $100. They spend 80 percent of their $100 of income on consumption, and 20 percent they save to invest. Now, when the people decrease their cash holdings (stock of savings) to $90, they can use the $10 to increase consumption spending from $80 to $90, which leads to a relative increase in consumer-goods prices in comparison to capital-goods prices and a shortening of the structure of production. Alternatively, they can decrease their cash balances and increase their investment expenditures by $10, from $20 to $30. This leads to a relative increase in capital-goods prices and a lengthening of the structure of production. As you can see, the reduction in cash balances (stock of monetary “savings”) may go along with a shortening or a lengthening of the structure of production.

Similarly, people can increase their cash holdings. They have here again three possibilities: they can reduce either their consumption spending or their investment spending or a combination of both. When they reduce their consumption spending, consumer-goods prices fall relative to capital-goods prices, and the structure of production is lengthened. As Mises (1998, 518–19) remarks on this particular case:

If an individual employs a sum of money not for consumption but for the purchase of factors of production, saving is directly turned into capital accumulation. If the individual saver employs his additional savings for increasing his cash holding because this is in his eyes the most advantageous mode of using them, he brings about a tendency toward a fall in commodity prices and a rise in the monetary unit’s purchasing power. . . . Whenever an individual devotes a sum of money to saving instead of spending it for consumption, the process of saving agrees perfectly with the process of capital accumulation and investment. It does not matter whether the individual saver does or does not increase his cash holding. The act of saving always has its counterpart in a supply of goods produced and not consumed, of goods available for further production activities.

Thus, when people add to their cash balances by abstaining from consumption, this implies a fall in time preference and a corresponding tendency toward a more capital-intensive structure of production. If new fiduciary media are created and injected into the economy, and the holders do not spend them—i.e., they are completely saved—then new projects are sustainable because of the relative reduction of consumption spending.

The other alternative to increasing cash holdings is reducing investment spending. This would lead to a relative increase in consumer-goods prices and a shortening of the structure of production. For instance, people may divest from capital equipment by not reinvesting in depreciating machines and thereby increase their cash holdings. Alternatively, they can sell financial assets to increase their cash holdings. So cash holdings (of money proper and money substitutes) may increase while at the same time real savings are falling. This insight does not combine well with the real bills doctrine. This is because the real bills doctrine implies that increases in the stock variable (namely, cash holdings) indicate that the flow variable (namely, real savings) has increased and that real bill credit expansion may safely finance additional investments.

To come back to our initial example of an increase in cash balances, it is also possible, if time preference does not change, to increase cash holdings by reducing consumption and investment spending proportionately. For instance, people may reduce both types of spending by 10 percent by reducing consumer spending from $80 to $72 and investment spending from $20 to $18, thereby adding $10 to their cash balances. Then there is no effect on the structure of production. As we can see, there is no necessary relation between the demand for money and time preference. The level of cash balances may change without changes in time preference and without changes in real savings.

One must distinguish between the purchasing power of money and the supply of real savings. The supply of and demand for money determine money’s purchasing power. An increase in the level of cash balances, however, does not say anything about the change in the proportion of investment and consumption spending or about the flow of real savings for investment. It is the supply of and demand for future goods that determine the amount of gross savings available for investments. The supply of and demand for future goods themselves are expressions of time preference. Time preference expresses itself in the proportion of spending on consumer goods (present goods) and investment goods (which hopefully yield consumer goods in the future).

4.2 The Yield Curve and Arbitrage

The banking school maintains that if only the right kind of loans are created by a fractional reserve banking system, there will be no business cycle. The right kind of loans are short-term, self-liquidating loans backed by goods to be sold in the near future: real bills. The interest rate on the real bills is called the discount rate. The Feketian banking school distinguishes between the interest rate and the discount rate, as if these were two radically different things and established in hermetically sealed markets. This separation allows the banking school to justify its argument that one type of credit expansion is unproblematic and to maintain that with the right kind of credit expansion, long-term interest rates, which are important for capital investments, are not affected.

In other words, according to the banking school, credit expansion to finance, for instance, mortgages lowers long-term interest rates, leading to a business cycle. However, credit expansion to finance goods that will be quickly marketed only influences the discount rate, not long-term interest rates, and therefore does not initiate a business cycle.

Yet the claim of hermetically sealed credit markets is flawed. Credit markets are like communicating vessels. When short-term interest rates (discount rates) are lowered by credit expansion, this also lowers longterm interest rates for two main reasons.

First, there is the practice of maturity mismatching or maturity transformation—i.e., borrowing short at low interest rates to invest long at higher interest rates. Maturity mismatching raises short-term and lowers long-term rates. When the discount rate is lowered by real bill credit expansion, maturity mismatching becomes more attractive than otherwise.

Not only is maturity mismatching perfectly legitimate (Bagus and Howden 2009, 2012; Bagus, Howden, and Huerta de Soto 2018), but it can also be successful. Maturity mismatching has an important function in making short-term revolving savings available to finance long-term projects, as I will discuss in more detail in section 5.

Second, even if there is no maturity mismatching or maturity transformation, as seems to be the policy objective of the banking school, a decrease in short-term interest rates has an influence on long-term rates. Savers will invest their funds differently. When short-term rates (the discount rate) fall, investments in long-term vehicles become relatively more attractive. Investment behavior changes on the margin. More people will invest long term, where rates are still higher, and fewer people will invest their funds short term, where rates have been lowered by real bill discounting and the creation of fiduciary media. People will shift their savings from short-term to longterm investment vehicles. As a consequence, long-term rates follow short-term rates and fall without any individual actors borrowing short and lending long. In other words, even assuming maturity matching, credit expansion (based on real bills) lowers long-term rates.

In short, when there is credit expansion in the form of short-term loans, the whole yield curve will shift downward. Not only short-term rates but also long-term ones are reduced, even though there is no increase in real savings. There is an artificial reduction of interest rates, possibly leading to an artificial, unsustainable boom.

4.3 Credit Markets are Interconnected

As we have seen, credit markets are not hermetically sealed. Moreover, the fiduciary media created can be invested flexibly. Credit created through the purchase of real bills is fungible. When a company gets new fiduciary media that have been created through the discounting of a real bill, the company has additional funds to invest. Normally companies will invest the funds where the highest yield is expected, and this tends to be in long-term projects. They might not use the very same monetary units that have been created, but they can use other funds which now are “liberated,” since money is a fungible good.

Let us assume a company in a 100 percent reserve system finances the transportation of goods to the market through equity. Now, real bill fractional reserve banking is introduced. By discounting real bills, real bill fractional reserve banking allows the company to get hold of additional funds. This practice frees up equity that before was used to finance the transportation of goods. The company can use its own funds in another way. The free funds will then be invested where the highest yield is expected, and this is, as indicated above, usually in longer-term projects. Money (and fiduciary media) is like water. It will find its way through little openings and holes and run downward, where the most profitable projects can be found.

As Rothbard (2000, 77) puts it, “A firm may simply cease using its own funds for financing short-term inventory, and instead borrow the funds from the banks. The funds released by this borrowing can then be used to make long-term investments.”

Similarly, Fritz Machlup (1940, 192, 201) states that because of credit expansion (of which real bill credit expansion is one type) there is less pressure on credit markets. Most importantly, the final form of credit and its use are not determined by the initial lender (Machlup 1940, 251).Through several channels the funds may end up financing long-term projects (Machlup 1940, 257–58; see also Machlup 1932).

First, the entrepreneur who discounts real bills at a bank may lend money to another entrepreneur, who invests in a long-term project.

Second, the entrepreneur who discounted the real bills will be able to pay with cash instead of asking for loans. The company that receives cash may then invest in long-run projects.

Third, there is some easing in the general credit market, which allows some marginal borrowers to invest somewhere. As the company financed by created credit (real bill credit) needs less real savings, marginal borrowers who could not find financing on credit markets will now get financing and invest where they expect the highest yield.

Thus, because of real bill credit expansion, interest rates tend to fall and the price of capital goods tends to increase. The fall of the interest rate has a stronger effect on the profitability of longer-term projects than on that of short-term projects (Machlup 1940, 252– 53). This makes it more likely that the credit ends up in producer-goods industries.

Along the same lines, Benjamin Strong, the first governor of the Federal Reserve Bank of New York, argued that customers could borrow using real bills to finance speculation and that there is no guarantee that fiduciary media created by real bills will be used to finance these goods (Chandler 1958, 197–98). Strong argued that one cannot know where the money raised through real bills will be used (Humphrey and Timberlake 2019, 73–74). Similarly, Ahamed (2016, 359) shows that when the Federal Reserve managed to reduce bank loans to brokers in 1928 and 1929, other sources of credit substituted for speculation. For instance, U.S. companies lent to brokers their surplus liquidity (influenced by the discounting of real bills, of course). In short, when real bills are discounted, thus creating new fiduciary media, these funds may end up in another area of the economy—for instance, financing stock market speculation.

What is important is not so much what kind of loans are granted, as the banking school claims, but rather what quantity of money (fiduciary media) is created by credit expansion. As Rothbard (2000, 76–77) puts it:

The important aspect of bank credit expansion is the quantity of new money thrown into business lending, and not at all the type of business loans that are made. Short-term, “self-liquidating” loans are just as inflationary as long-term loans. . . . The crucial point is the injection of new money into business firms; regardless of the type of business loan made, this money will then seep into the economy, with the effects described in the Austrian analysis. The irrelevance of the type of loan may be seen from the fact that business firms, if they wish to finance longterm investment, can finance it indirectly from the banks just as effectively as from direct loans. . . . All credit is interrelated on the market, and there is no way that the various types of credit can be hermetically sealed from each other.

The creation of credit without prior savings is problematic, independent of the type of credit. The type of loan is irrelevant because the essential question is whether someone has given up purchasing power and control over consumer goods. As Machlup (1940, 171) puts it, “No matter whether they [the banks] do this [create credit without prior savings] by making advances or overdrafts, by discounting bills or by purchasing securities, they are providing purchasing power which has not been given up by anybody beforehand. Part of the supply of money capital thus frequently is ‘created’ credit.”

Similarly, Mises (1998, 790) writes about the idea of qualitative credit control:

The mode in which the additional amount of credit finds its way into the loan market is only of secondary importance. What matters is that there is an inflow of newly created credit. If the banks grant more credits to the farmers, the farmers are in a position to repay loans received from other sources and to pay cash for their purchases. If they grant more credits to business as circulating capital, they free funds which were previously tied up for this use. In any case they create an abundance of disposable money for which its owners try to find the most profitable investment.

Real bill credit expansion is as inflationary as any other credit expansion. However, there is one merit to Rallo’s approach: Restricting credit expansion to real bills is a step in the right direction, as it rules out other types of credit expansion and thereby the overall quantity of credit expansion (Rothbard 2000, 77). And the quantity of credit expansion influences the quality of credit. Restricting the quantity of credit expansion increases the quality of loans. In contrast, the greater the credit expansion, the stronger the need to lower credit requirements.

It is true that real bill credit expansion leads to shorter terms on outstanding loans. They are paid back earlier, reducing the money supply. Yet, in the same way that banks may renew and grant new mortgages (backed by the market value of the houses) when mortgages are paid back by borrowers, banks may renew or grant new short-term loans through discounting further real bills. As long as the short-term loans are renewed, they have the same effects on the money supply as longer-term loans.

In the same way that Peel’s Bank Act can be seen as a step in the right direction because it restricted the possibilities for credit expansion (by making it illegal to issue unbacked bank notes), the real bills doctrine also restricts the amount of credit expansion by only allowing “high-quality” loans. Yet, as long as credit expansion remains possible, it is an error to think that business cycles will be ruled out. The proponents of Peel’s Bank Act promised the end of business cycles, but they failed because they did not understand that demand deposits equal bank notes in their function as money substitutes. The proponents of the real bills doctrine commit a similar error because they do not understand that fiduciary media are perfect money substitutes. They believe that a banking system following the real bills doctrine would lead to a stable economy without booms and busts in the same way that the proponents of Peel’s Bank Act believed that their proposal would eliminate business cycles.

In defense of the proponents of the real bills doctrine, one could argue that cycles induced by real bill credit expansion produce only negligible or no harm, as they distort the structure of production only slightly— namely, by 3 months. With funds secured for 3 months only, no long-term investment project is started anyway, the argument goes.

Yet this argument fails for two reasons. First, as stated at the beginning of this subsection, money is fungible, and when funds are liberated they tend to be invested where the highest return is expected, which is usually in longer-term projects. Second, even if we do not consider this fungibility, the 3 months add up over all stages of production, as Huerta de Soto (2014, 230–31) illustrates with a useful example.

Huerta de Soto assumes an evenly rotating economy (ERE) with 5 stages in which goods of high quality are sold from one stage to the next. The production process of each stage takes 9 months, and capitalists have to wait 3 additional months for payment because they allow their customers to pay for the goods received within 3 months. The economy is working completely without banking. In the thought experiment, a fractional reserve banking system is added that offers to discount real bills collateralized by the goods of high quality. Thus, the capitalists do not have to wait 12 months to be paid but receive money substitutes after 9 months. If time preference does not fall and capitalists do not save the created funds for 3 months, they will invest them, purchasing capital goods from their suppliers. There is a lengthening of 3 months in all stages, which adds up to 15 months over all five stages. The artificial lengthening is then 1 year and 3 months, or 25 percent of the 5 years that the production process lasted before the fractional reserve banking system was introduced. Once the real bills are being repaid, banks roll over or renew them, and so on. Thus, there are now 6.25 stages instead of 5 stages. The distortion of the structure of production is much larger than 3 months, even if we do not consider the fungibility of money.

We can thus conclude that any type of credit expansion, independent of the quality or type of the loans granted, distorts the structure of production if unbacked by real savings. The real bills doctrine, if it allows for credit expansion without prior savings, will trigger Austrian business cycles. It is the quantity of new fiduciary media injected into the credit market that counts. Any injection of fiduciary media will exert expansionary effects.

As Mises (1998, 439n17) puts it, “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.”

4.4 The Demand for Money and the Interest Rate; or, The Needs of Trade as an Exogenous Limit on Credit Expansion

In this section I will deal with the argument offered by the banking school: that credit expansion is limited by the needs of trade. Is credit expansion limited by the amount of real bills, which itself it is limited by the amount of real goods produced? Is it the case that first a real bill comes into existence and then a loan? Or is it the case that first banks lower interest rates, which makes it more attractive to issue and discount real bills?

Rallo (2019) believes that when the need of trade and the demand for money increase, banks expand credit by discounting real bills. The increase in the money supply in the broader sense is a response to an increase in the demand for money backed by real bills and ultimately by real goods. Banks do not expand credit on their own, but only react to a higher demand for money.

What proponents of the real bills doctrine do not see is that the demand for fiduciary media is not independent of banks’ action—namely, the ease with which they grant loans. Banks, by reducing the credit criteria or lowering the interest rate on loans, can induce a higher demand for loans. Furthermore, they can maintain their fiduciary media in circulation by renewing their loans at lower rates. Borrowers compare the interest rate charged on loans with the expected profit rate they will get from investing these loans. The lower the interest rate charged, the more investments appear profitable and the higher the demand for loans ceteris paribus.

Without prior demand for fiduciary media, banks can induce this demand by reducing interest rates (Huerta de Soto 2012, 683). The lower the discount rate, the more real bills will be presented at banks and the higher will be the demand for fiduciary media. As Rothbard (2000, 77) states, “Banks do not simply passively await business firms demanding loans: these very demands vary inversely to the rate of interest that the banks charge.”

Mises (1998, 436–37) similarly states:

The Banking School failed entirely in dealing with these problems. It was confused by a spurious idea according to which the requirements of business rigidly limit the maximum amount of convertible banknotes that a bank can issue. They did not see that the demand of the public for credit is a magnitude dependent on the banks’ readiness to lend, and that banks which do not bother about their own solvency are in a position to expand circulation credit by lowering the rate of interest below the market rate.

Machlup (1940, 265) considers the whole argument that banks passively discount a given amount of real bills presented to them as based on mere superstition. Banks could react to an increased demand for money by restricting the supply if that is in their interest:

It was supposed for a long time that there was a strict dependence of the total volume of bills and requests for their discount on the “volume of trade.” It has frequently been pointed out that this is a mere superstition. The number of bills coming forward for discount is largely dependent on the credit policy of the banks. An increase in the demand for loans against bills can in principle just as well lead to a tightening of the discount market as to the granting of the loans. It all depends on the policy of the banks.

Indeed, when, in a recession, the demand for loans (real bill discounting) from troubled companies increases, banks may decrease their supply while raising rates.

And it is not true that an overexpansion leads directly and necessarily to a reflux, in which fiduciary media are represented for redemption. As Mises (1998, 424) points out, when prices increase, people often increase their real cash balances because they restrict purchases in the hope that prices will drop again. Thus, at least for a prolonged time, the demand for money may adjust to the supply of fiduciary media.

One should also not forget that the ability to issue fiduciary media is not restricted by people’s wish to hold higher cash balances. There is always demand to hold fiduciary media, as long as they are considered perfect money substitutes, because they give their holder an advantage over those who have lower cash balances. The demand for fiduciary media is, therefore, not restricted by the needs of trade.

Additional fiduciary media give entrepreneurs a competitive advantage over their rivals independent of the evolution of the “needs of trade.” This is because whenever there is an increase in the money supply in its broader sense, there is a redistribution (Cantillon 2017). The first receivers of the fiduciary media benefit, as they can still buy at the old, lower prices, while the last receivers lose. An entrepreneur who receives a loan from a bank in the form of fiduciary media enjoys an increase in his cash balance. He may purchase the newest technology at the old, lower prices, while his rival who did not indebt himself cannot. The second receiver of the fiduciary media also has an advantage because prices have not increased across the board yet. Thus, Cantillon effects are a reason why there is a demand for fiduciary media independent of the needs of trade.

It is not only that the demand for fiduciary media is not exogenous to the banking system. It can not only be stimulated by banking policy but also feed on itself (Bagus and Howden 2010b). When real bill credit expansion causes the money supply to increase, asset prices may increase as well. People may feel richer (wealth effect) and demand more loans from banks collateralized by rising asset prices. The demand to hold fiduciary media increases (Bagus 2008b; 2007).

If the needs of trade do not put a limit on credit expansion, what about bank competition?

Mises (1998, 440) argues that competition in a fractional reserve free banking system would restrict credit expansion, even though it would not bring it to a halt: “It [free banking] would, it is true, not hinder a slow credit expansion, kept within very narrow limits, on the part of cautious banks which provide the public with all information required about their financial status.” In his Theory of Money and Credit Mises (1953, 397) similarly states, “It is clear that banking freedom per se cannot be said to make a return to gross inflationary policy impossible.”

This is because fractional reserve banks can cooperate and expand in concert, in which case they do not lose reserves to their competitors. There is an incentive to expand credit in rhythm since this allows them to increase profits (Huerta de Soto 2012, 664–71). And those banks that do not discount real bills will lose market share. This is, basically, the argument of S. M. Longfield (1840) in his reply to Henry Parnell (1827), who thought that clearinghouses would restrict credit expansion. Thus, a coordinated credit expansion is possible (Bagus and Howden 2010b).

________________________

Mises (1953, 52) states that claims themselves are not goods but give the right to dispose of goods. What he seems to mean is that claims are not the goods that they can be exchanged for. However, when these claims are on demand, they act as if they were present goods: they are perfect substitutes.

On probabilities of debt contracts, see Böhm-Bawerk (1881, 86).

Rallo comes up with a new term for these financial assets: “promises to pay.” So “perfect money substitutes” become “promises to pay” with a certain default risk. This semantic change seems a stretch and inadequate since bank moneys are money substitutes that are completely available for purchases from a subjective point of view and are objectively used as if they were money proper. They act in practice as (perfect) money substitutes.

While there are objective differences between an accounting liability of a central bank (i.e., cash in circulation) and an accounting liability of a commercial bank (i.e., a demand deposit), what is essential is the subjective viewpoint of the actor, which can make the latter the perfect substitute for the former. On the subjective view of the money supply, see Rothbard (1978) and Salerno (1987).

A counterargument by Rallo could be that IOUs may circulate without a discount and be used as money substitutes. Indeed, when Elon Musk writes an IOU of $100 (redeemable on demand), it could be accepted without a discount. There is no problem. Musk must save—i.e., restrict his consumption by $100—because the IOU could be presented to him at any time. In this case there is no double availability. The holder of the IOU may consider it as of high liquidity and reduce his money holdings as a consequence. Yet such an IOU will not be considered a perfect money substitute for several reasons. First, it is not easily divisible into smaller amounts. Second, its circulation is restricted to professionals: those who know who Elon Musk is and what his signature looks like, and those who can judge the authenticity of the document. While individuals may regard cash in hand and money in their bank account as subjectively interchangeable, they will find differences between an IOU from Elon Musk and cash.

Thus, Elon Musk’s issuing an IOU that is redeemable on demand does not increase the supply of money in its broader sense, which comprises money and (perfect) money substitutes, because the IOU is not a perfect money substitute but indeed a financial asset. No money has been created without prior savings, and Elon Musk must maintain liquidity to pay his debts.

Banks operate fundamentally differently from Musk issuing an IOU. No one is depositing cash with Musk and receiving a receipt or certificate. Rather Musk owns cash and issues a claim against it. He issues a claim to the cash he actually owns. In contrast, fractional reserve banks do not have cash reserves to redeem all claims but create money substitutes without having cash (creating fiduciary media). They use deposited money to grant credit in a volume higher than these deposits of money proper.

Hülsmann (2000) argues in this respect that banks cannot create money but only titles to money. He argues that it should be prohibited for banks to issue titles to money, but it should be allowed for them to issue IOUs. These IOUs, like Elon Musk’s IOU, would experience a “fringe existence” in a free market (109). Thus, they would be different from fiduciary media. The question here remains up to which point these IOUs would be considered immediately available on demand and, therefore, equivalent to demand deposits. If these bank IOUs are considered equivalent, then the obligation of demand-deposit contracts would apply and banks should have the cash at hand at all times.

Hansen (2021) argues along these lines that the nature of fiduciary media is incompatible with the aim of people holding money (to reduce uncertainty) and an entrepreneurial error.

Fritz Machlup (1940, 220) shows that even with short-term fractional reserve loans, there is double availability, leading to an economic crisis:

If . . . the modern credit system supplies this entrepreneur with new “short-term funds” in order to finance his “merely temporary capital requirements” next time they arise, or lends “short-term funds” to the other entrepreneur so that he can repay his loan, the effect is that command over the same productive resources is given twice over. It is obvious that in the long run this situation cannot endure and that equilibrium can be re-established only after the expansion of the more roundabout production processes has been followed by contraction associated with the usual phenomena of the crisis.

Huerta de Soto’s (2012, 700) statement on Selgin is, at least partially, also applicable to Rallo:

It is not surprising that these doctrines [doctrines that maintain that an increase in fiduciary media equals saving] have been defended by authors like Keynes, Tobin, Pointdexter and, in general, all who have justified inflationism, credit expansion and the “euthanasia of the rentier” for the sake of aggressive economic policies geared to insure an “adequate” level of “aggregate demand.” What is surprising, however, is that authors like Selgin and Horwitz [and Rallo], who belong (or at least belonged) to the Austrian School and thus should be more aware of the dangers involved, have had no alternative but to resort to this sort of argument in order to justify their “fractionalreserve free-banking” system. (emphasis in original)

Another case is that of a change in the proportion in the spending of consumer goods and producer goods due to dishoarding. When dishoarding leads to a relative increase in spending on producer goods, this reflects a fall in time preference. The producers of producer goods will have more buying power, leading to a relative reduction in consumption. On this case, see also fn. 43.

If a producer produces for the market, he has to have savings until his product is sold to the final consumer. That is, he does not need savings only until his product is “finished” in an objective way; he needs them until the product is actually purchased by the consumer.

The traditional real bills doctrine considers real goods in the process of production as good collateral for bills—i.e., not only consumer goods but also fungible and highly demanded intermediate goods. Thus, Rallo defends a special and restrictive kind of real bills doctrine.

Another question is the following: Why is credit expansion based on real bills (i.e., loans for 90 days) sustainable, but loans with 91-day, 92-day, 93-day, etc., maturity dates are not?

Here we have the most favorable case for the real bills doctrine. We can also imagine a scenario of an existing production process that has been financed in its totality through equity; then the last phase of the production process starts to be financed by real bill credit expansion, freeing up equity. Apparently, there was no increase in real savings. Credit markets and capital markets are interconnected. When more credit is created by discounting real bills, savings are liberated for other investments whose interest rate tends to fall, generating systematic investment errors.

Another point must be clarified: In the example, I analyze the effect of additional real bill credit expansion backed by additional consumer-goods production, not the rolling over of old loans. The old credit expansion had already had its effects, and the economic system had become accustomed to a certain amount of fiduciary media. The example aims to analyze the effects of an additional dose of fiduciary media injected through real bill discounting.

This implies in banking-school jargon that the “needs of trade” increase. Another question concerning a real bill fractional reserve gold standard is the following: would banks discount more real bills only when the needs of trade increase? Imagine that new gold is mined on our island by Tom. At this point, the new mining has not increased the number of available fish (it may have increased slightly the amount of consumer and producer goods, as gold is not only money but also a factor of production and a consumer good); thus real savings have not increased. If the newly mined gold is then used to purchase a real bill from Robinson, which is later rediscounted by a fractional reserve bank, then fiduciary media have effectively been created to finance a company’s investments. There is no increase in real savings. A similar critique can be brought forward against fractional reserve banking à la Selgin. When gold is mined, it flows into banks, raising reserves. This would allow Selgin’s banks to expand credit without any prior increase in real savings. See, on this point, Huerta de Soto (2012, 687) and Bagus and Howden (2010b; 2011b).

When people dishoard, find money in their backyard, or economize on their money holdings through payment innovation, they will spend the additional money in the same proportion on consumption and investment goods if their time preference remains the same. If they spend 100 percent of their additional purchasing power on investment goods, then their time preference has changed, and more resources will be directed toward investment. Economic actors are saving more (in relative terms). In contrast, with an increase in fiduciary media, there may be the illusion (Hülsmann 1998) that more investment projects are viable without an increase in savings. With no decrease in time preference, the (full) investment of the fiduciary media to lengthen the structure of production will lead to an unsustainable boom. Once the fiduciary media reach the owners of the factors of production and they spend it on consumer goods, the relative increase in consumer-goods prices will lead to a recession. See also fn. 6.

The case of dishoarding may also be illustrated with our Robinson example. There may be dishoarding to purchase producer goods (such as fishing nets) while consumer-good (fish) spending remains the same. The relative increase in the price of fishing nets will enable the producer (Robinson) to purchase more fish to sustain production. More fish (real savings) are transferred to Robinson in line with the change in time preference. The structure of production becomes more capital intensive with a lower purchasing power of money in line with consumer preferences.

As Bagus and Howden (2010b, 43) put it in relation to fractional reserve credit expansion:

The confusion between increases in savings and cash holdings is a confusion between stock and flow variables. Saving is a flow variable—the part of income that is not consumed. Cash holdings (savings) represent a stock in existence. Cash holdings do not represent saving. One may actually increase cash holdings by saving less (and consuming more), for example, spending a smaller portion of the available income on investments (or selling investments in order to consume). Fractional reserve banking leads to a change in the stock variable (cash holdings) that may create the artificial perception of a change in the flow variable (saving).

See Rothbard (2001, 689–90) and Huerta de Soto (2012, 694). As Hoppe (1994, 72–73) points out:

First off, it is plainly false to say that the holding of money, i.e., the act of not spending it, is equivalent to saving. . . . In fact, saving is not-consuming, and the demand for money has nothing to do with saving or not-saving. The demand for money is the unwillingness to buy or rent non-money goods—and these include consumer goods (present goods) and capital goods (future goods). Not-spending money is to purchase neither consumer goods nor investment goods. Contrary to Selgin, then, matters are as follows: Individuals may employ their monetary assets in one of three ways. They can spend them on consumer goods; they can spend them on investment; or they can keep them in the form of cash. There are no other alternatives. . . . [U]nless time preference is assumed to have changed at the same time, real consumption and real investment will remain the same as before: the additional money demand is satisfied by reducing nominal consumption and investment spending in accordance with the same pre-existing consumption/investment proportion, driving the money prices of both consumer as well as producer goods down and leaving real consumption and investment at precisely their old levels.

On the process of cash building by abstaining from consumption that leads to economic growth, see Huerta de Soto (2012, 448–49) and Bagus (2015, 65–66); and for a comparison of the direct investment of savings and cash building through savings, see Bagus (2016). Hülsmann (2009) would not agree with our analysis. Hülsmann considers not three margins but only two: present goods and future goods. He rightly argues that money is a present good. For him, reducing consumption spending to increase one’s cash balance implies a change in the composition of present-goods spending but not a change in time preference.

I do not agree with Hülsmann. It is true that money is a present good. Yet it is decisively different from consumer goods, as consumer goods can be used to sustain factors of production during the production process. Therefore, the distinction between consumer goods, producer goods, and money remains important. Hülsmann argues that in a commodity money standard an increased demand for money would raise the pure rate of interest. He argues that as prices fall, the return on investment (ROI) in money production (e.g., mining) increases. Then, factors of production are shifted to mining, raising the ROI in all other industries and reducing it in mining until the pure rate of interest stabilizes at a higher level. Yet when the proportion of consumption and investment spending remains the same, time preference does not change. The prices of factors of production in mining are bid up until the ROI in mining corresponds with time preference. The fact that there is an increase in the demand for money shows that consumers value production in one sector in the economy relative to that in other sectors higher than before.

Similarly, consumers could increase their demand for bicycles by reducing their demand for cars. The ROI increases in the former sector and falls in the latter in the beginning until factors of production are shifted and ROI equalizes. However, there is no change in the pure rate of interest. The same applies for an increase in the demand for money without a change in time preference (and the proportion of consumption and investment spending).

See Huerta de Soto (2012, 697). Huerta de Soto (2014, 231) gives an example of an increase in cash holdings through selling investment assets. Thus, an increase in cash holdings is compatible with an increase in consumption and a decrease in investment spending. An individual can sell bonds, increase his cash balance, and increase consumption. Another individual then does the opposite. A society may also increase real cash balances and consumption by reducing savings. The members of society reduce their savings and invest less in the purchase of financial assets. Then, maturing bonds, for instance, are not renewed. Instead, the funds are used to increase cash balances and consumption.

A growing economy does not change our conclusions. In a growing economy, consumption increases with constant time preference, as does the amount of goods available to sustain production. However, if credit is expanded, then interest rates will be lower than otherwise, and more projects will be undertaken. There is a distortion while there is growth. In other words, even if new goods are coming to market and there is growth, with constant time preference, credit expansion leads to malinvestments.

There exists a vast empirical literature on the relation between shortterm and long-term interest rates. The effect of monetary policy, which is traditionally directed at short-term rates, on long-term rates has been especially researched. For a review article that finds that short-term rates had significant effects on long-term rates, see Akhtar (1995). Even a banking-school author such as Felix Somary (1930, 228–31) points out that the “money market” and the “bond market” are connected by arbitrage.

Hayek (2008, 290) argues that there are instruments to economize on money, such as book credit or bills of exchange, that liberate funds: “They give to somebody the means of purchasing goods without at the same time diminishing the money-spending power of somebody else.” The same applies to mergers that reduce the need to use money proper in the economy. While mergers and the use of secondary media of exchange liberate funds, this involves no new money creation. The increased economizing on money is a limited and slow process. Credit expansion, however, may boost the creation of secondary media of exchange. Credit expansion expands the issue of secondary media of exchange such as bills of exchange because these secondary media of exchange are bought with fiduciary media and become more negotiable. Without credit expansion the growth in use of secondary media of exchange is slower and their supply more stable.

Moreover, and as pointed out in fn. 6, the effect of increased economizing and mergers is countered by other influences, such as the creation or spinning off of new companies or the creation of new productive stages, which increase the demand for money for transactions.

The argument that the real bill credit will only be used to finance consumer goods (if the needs of trade require it) is problematic for other reasons. There are several stages of production, and the consumer-goods stage is only the last one, which depends on the inputs of the earlier stages. One cannot expand the consumer-goods stage sustainably without expanding earlier stages of production (Machlup 1940, 205). The idea that real bill credit can be restricted in its effect on the consumer stage is, therefore, mistaken.

Machlup (1940, 253) admits, though, that credit expansion that finances working capital is less harmful than credit expansion that finances fixed capital.

As Machlup (1932, 275) clarifies, “The entrepreneur who receives fresh advances for working capital, in fact, has part of his own working capital set free—which was previously needed for circulating capital; so that he can now invest his own funds in fixed capital. Thus, the actual concrete and visible use of a fresh credit does not correspond to the intended use.”

For an empirical approach see Cachanosky and Lewin (2016), who find that more roundabout—i.e., longer-term—projects are more sensitive to interest rate changes. And as Machlup (1932, 276–77) points out, when new projects are started, the prices of production goods and factors of production increase. The fall of the interest rate affects circulation capital less than fixed-capital investment because the interest rate is of greater relevance for the latter. In other words, for short-term projects the increase in factor prices tends to be more important than the fall of the interest rate, while for longer-term project the rate fall is more relevant. Therefore, marginal borrowers tend to invest in longerterm projects. Machlup (1932, 281) also states that his friend Hayek would probably agree with him that all new funds, including shortterm funds, are finally used to finance long-term projects.

Alluding to banking practices like the real bills doctrine, Machlup (1940, 193) writes, “Thus, it seems that the likelihood that a credit expansion will be crash-proof is not increased by the fact that loans are made to selected industries on the basis of certain rules about liquidity.” He continues (1940, 200), “The ‘proper’ limits of credit expansion are not affected by the nature and quality of the credit.” Machlup (1940, 261) concludes, “It is not the form the credit takes nor the exact place where it enters the system that makes it dangerous: it is, instead, its amount.” Machlup (1940, 292) sums it up: “Qualitative credit control is effective only if it involves quantitative control.”

Something similar occurs when we look at central banks’ government-bond purchases and lending activities collateralized by government bonds. When central banks continue to roll over loans to the banking system collateralized by government bonds, it has the same effect as a central bank’s purchase of government bonds. New reserves are created. In the case of the loans, reserves increase as long as they are not rolled over anymore. In the case of the purchases, reserves increase until the bonds are sold or the bonds are paid back (Bagus 2011).

As Mises (1953, 345) puts it regarding banking-school representative Fullarton:

The fatal error of Fullarton and his disciples was to have over-looked the fact that even convertible bank-notes remain permanently in circulation and can then bring about a glut of fiduciary media the consequences of which resemble those of an increase in the quantity of money in circulation. Even if it is true, as Fullarton insists, that bank-notes issued as loans automatically flow back to the bank after the term of the loan has passed, still this does not tell us anything about the question whether the bank is able to maintain them in circulation by repeated prolongation of the loan. The assertion that lies at the heart of the position taken up by the Banking School, viz., that it is impossible to set and permanently maintain in circulation more notes than will meet the public demand, is untenable; for the demand for credit is not a fixed quantity; it expands as the rate of interest falls, and contracts as the rate of interest rises. But since the rate of interest that is charged for loans made in fiduciary media created expressly for that purpose can be reduced by the banks in the first instance down to the limit set by the marginal utility of the capital used in the banking business, i.e. practically to zero, the whole edifice built up by Tooke’s school collapses.

See also Mises (1953, 312): “The quantity of fiduciary media in circulation has no natural limits. If for any reason it is desired that it should be limited, then it must be limited by some sort of deliberate human intervention—that is by banking policy.”