“The fundamental error of our financial policy lies in the attempt to create wealth by creating currency: it is putting the servant before the master -- the wrong power, in advance. We can create wealth only by producing commodities.” The frequency with which monetary crank schemes are proposed indicates that this great truth written by Charles Holt Carroll (148) in 1859 has not yet been learned.

Case in point is Nelson Hultberg and Antal Fekete’s call for a revival of the Real Bills Doctrine (RBD). The Real Bills system is a form of monetary crankism: at its core is the fallacy that paper can create wealth. Carroll, one of the most astute critics of paper money, had it right when he wrote that the RBD is “the most remarkable and the most mischievous heresy that ever found an advocate in any science.”(267)

Limiting the danger of inflation is most prominent reason for using gold as money. While the supply of gold can at best grow slowly, the quantity of paper can be multiplied without limit. The resulting inflation erodes the purchasing power of wages and savings. The Real Bills Doctrine -- a theory advocating the creation more paper money substitutes -- cannot be exempt from this evil.

Yet RBD theorists hold that the discounting of bills that they propose is non-inflationary. They believe that they have discovered Inflation Lite: a miraculous form of fiduciary media that facilitates more trade but does not increase prices. Implementation of the RBD would be a path to inflation; non-inflationary monetary expansion is a mythical beast.

The doctrine states that banks should be allowed to monetize short-term business loans. Part 1 presents an explanation of the monetization bills of exchange, explain the difference between transfer credit and credit expansion, and exposes the fallacy of a credit shortage. In addition, part 1 shows that only savings can fund production, and describes how forced savings occurs in response to credit expansion.

The proposal advanced by Fekete is a non-solution to a non-problem. Because it provides no benefit, there is no point in adopting it. In the best case, something that has no benefit would be harmless. However, the RBD is far from being harmless. The current article will show that adoption of the RBD would inevitably be inflationary without any limit The current article will also emphasize some subtleties of the Austrian critique of monetary expansion.

The discounting of bills as per the doctrine would introduce fiduciary media into circulation. The creation of fiduciary media is always inflationary because the paper notes have equivalent purchasing power to money itself and therefore affect prices in the same way. Carroll sees the point clearly: “Nothing is created by this operation but debt -- no capital, value, or wealth whatever -- but price is added to commodities thereby as effectually as if so much gold had been produced or earned by labor and added to the currency.” (137)

A market price is created any time money or a paper claim functioning as money is spent. As Mises explains here, money prices are formed through the use all real money and fiduciary media in circulation:

If notes are issued by the banks, or if bank deposits subject to check or other claim are opened, in excess of the amount of money kept in the vaults as cover, the effect on prices is similar to that obtained by an increase in the quantity of money. Since these fiduciary media, as notes and bank deposits not backed by metal are called, render the service of money as safe and generally accepted, payable on demand monetary claims, they may be used as money in all transactions. On that account, they are genuine money substitutes. Since they are in excess of the given total quantity of money in the narrower sense, they represent an increase in the quantity of money in the broader sense.

When money is loaned in a credit transaction, the increase in the purchasing power of the borrower is offset by the saver’s withdrawal of purchasing power. When a saver loans to a borrower, different prices will be created than if the saver had kept the money instead of loaning it. This is so because the money loaned will be put to different uses by the borrower than it would have by the saver. The borrower might use the money to rent office space, while the saver might have used it to purchase a car. But there will be no general tendency toward higher prices in an economy based on transfer credit even when credit transactions are common.

On the other hand, an increase in the quantity of fiduciary media necessarily results in a higher market price for some good because when they are issued, there is no offsetting savings that withdraws demand elsewhere. When a business sells its bills to a bank for unbacked paper claims, the firm might use their phony paper money to pay wages to employees, rent office space, or purchase machinery. Whatever it is, it will sell at a higher price than would be the case in the absence of the fiduciary media. As Hülsmann explains:

Suppose I get an additional fiduciary banknote of one ounce of silver sterling from my banker. This banknote permits me to satisfy wants that hitherto were not sufficiently important to be considered (they were submarginal). If I pay for a meal in a restaurant with this banknote then, without any doubt, I have affected market prices. In fact, by my very purchase I have formed market prices. These prices would have never come into being without the additional issue of a banknote. Selling the meal to other persons would have required a price reduction to attract submarginal consumers.

Suppose that a merchant is short of cash but he possesses a bill. He tries to sell his bill for cash. There are two possibilities: under a system of transfer credit, he sells the bill by obtaining credit (the transfer of someone else’s savings). Or, if the RBD were adopted, a bank would expand credit and pay the merchant with fiduciary media. The first potion, borrowing savings, is more costly to the merchant because the he must offer the saver a sufficient rate of interest to make them willing to part with their money. The alternative, fiduciary media, can be printed at nearly zero cost. The bank that prints money instead of borrowing savings can therefore offer a lower rate of interest. Credit expansion allows the merchant to pay less interest – which means to receive more cash – for his bill.

Consider the situation of a merchant who needs some quantity of cash to pay current expenses, and who owns a bill of exchange. Suppose that a bank operating according to the RBD is willing to offer him exactly as much cash as he needed for his bill. Then, under a system of strict transfer credit, the bank would offer him less than that amount because of the higher cost of borrowing savings compared to creating fiduciary media. Starting from the same initial conditions in a transfer credit system the merchant would not be able to sell his bill for enough cash to pay his obligations.

There are two possibilities here. One is that he might be bankrupt. As I explained in part 1, this is not a bad thing; it is part of the market’s process of allocating resources. The assets to do not go away, or even necessarily cease to be productive. The firm’s creditors would take over the ownership, the assets would be revalued at lower prices, and in more solvent hands might be put to more better use.

The other possibility is that the merchant can reduce his costs. Suppose that he is able to do so either by negotiating with his suppliers for lower prices or with his employees for lower wages or by purchasing fewer inputs. Then transactions would occur but at lower prices than under a system of credit expansion. This example illustrates how, under a flexible price system, prices change to facilitate transactions. There is no need for an “elastic” monetary system when prices can move.

There is a difference between the prices of a bill under transfer credit and under the RBD. The difference consists of purchasing power shifted from one person to another by the monetary expansion that occurs when fiduciary media are issued. The additional purchasing power of the merchant only comes into existence at the expense of all other money holders elsewhere, by diluting the value of their monetary units. The issuance of fiduciary media, then, enables the seller of the bill to obtain something that they could not afford in economic terms, at the expense of the rest of the population who find their own money to be worth less as a consequence.

It defies logic to say that thing is true and that it is not true. For the system of monetizing bills of exchange to be non-inflationary would mean that it did not result in higher prices. Yet the entire motivation for the system is to enable business transactions that could otherwise only take place at lower prices, or not at all.

It is claimed that paper money printing if done according to the rules of the RBD is not inflationary because the paper finances the production of particular goods and then goes away. There are two problems with this. In the first place, is a serious misunderstanding of the nature of productive activity. The paper does not fund production. There is no way that paper by itself can fund production, only the goods purchased with the paper fund production. The holder of the phony paper notes is only able to buy existing goods because he can outbid others who were not so lucky as to be sitting next to the printing press. The only way to provide goods more cheaply is to produce more of them through savings, work, and investment.

Secondly, this line of thinking rests on the idea that the certain money somehow corresponds to specific goods. Under a commodity money system, money is a good in the economy that functions as the medium of exchange. Money prices are the exchange ratios between money and goods. Money prices are formed through the interaction of all money holders and all goods owners in the economy. There is no identification between specific money units and any particular goods. In the process of price formation, money is acquired to be spent, and then acquired again and spent again, forming price at each exchange along the way. The explication of this point by the French economist J.B. Say could not be improved upon:

The silver coin you will have received on the sale of your own products, and given in the purchase of those of other people, will the next moment execute the same office between other contracting parties, and so from one to another to infinity; just as a public vehicle successively transports objects one after another.

There do exist instruments that are collateralized by particular goods. These are known as bonds, equity shares, etc. But these instruments are not money. The evaluation and pricing of these instruments is complex as they are heterogeneous and carry different degrees of credit risk. Even collateralized bills of exchange are subject to market risk. Firms can produce inventory and then find themselves unable to sell it.

There are additional fallacies in the association of monetized bills with particular goods. Fiduciary media are created at a distinct point when they are loaned into existence. This is called the “point of injection”. When a bank expands credit by monetizing a bill, the point of injection is the credit market. However, the point where this new paper enters the spending stream does not limit its effect on prices to that point. As Mises explained, “variations in the value of money always start from a given point and gradually spread out from this point through the whole community”.

Even for short-term loans, there is nothing about spending of new money that limits its purchasing power to the production of those particular goods in process that were the collateral for the monetized bill. In the short term as in the long term people receive wages, buy groceries, pay rent, go on vacation, and fill up their gas tank.

A major point in Fekete’s writings is that bills are liquidated within 91 days or less. The fiduciary media are withdrawn from circulation after a short time, so they can’t do much harm, or so we are told. The defense of the doctrine is the theory rests heavily on the belief that credit extended for 91-day-or-shorter periods is economically fundamentally different than credit for longer periods.

Numerology notwithstanding, there is nothing special about the number 91. There is no economic distinction between loans shorter than or longer than some number of days. It makes as much sense to say that purchases of bananas should be paid in gold coin while strawberry consumption should be funded with bank credit expansion. All stages of production – including shipping partially finished goods in process -- consume real resources that have alternative uses. All credit must be borrowed (whether for a short or a long time) from the same potential pool of savings, namely present goods. Present goods are scarce in the present. There exists nothing with which to fund investment other than present goods that have been saved. The choice is only whether the savings are voluntary (as they would be if they were offered on true credit) or forced (as would be the case when fiduciary media are issued).

The focus on the life cycle of a particular bill is misplaced. It is the total volume of credit expansion and contraction in the banking system is responsible for inflation and deflation within an economy. The life expectancy of any particular bill of exchange does not provide a measure of the credit expansion that would occur if the RBD were implemented.

If, as Mises wrote on this subject, “When the loan is paid back at maturity, the banknotes return to the bank and thus disappear from the market,” then there would be only a small amount of credit expansion, followed by an equal-sized credit contraction. But, he continues, “this happens only if the bank restricts the amount of credits granted.…The regular course of affairs is that the bank replaces the bills expired and paid back by discounting new bills of exchange. Then to the amount of banknotes withdrawn from the market by the repayment of the earlier loan there corresponds an amount of newly issued banknotes.”

Mises reminds us that

The fatal error of Fullarton [a member of the Banking School] and his disciples was to have overlooked the fact that even convertible banknotes 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 banknotes 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.

During the historical debates between the Currency School and the Banking School, the latter made a similar argument. They maintained that banks, by expanding credit, were only accommodating the “needs of trade”. They argued that the issuance of unbacked paper notes was a market mechanism that arose simply to fill a need for a certain quantity of credit.

This argument runs aground on the following problem: the demand for credit is not independent of the volume of bills issued. To say otherwise ignores the impact of money supply on money demand.

Unlike for other goods, money demand depends in part on money supply. To understand this, first consider why it is non-money goods do not behave this way. It is quite reasonable to suppose that an increase in the supply of lawnmowers will more fully meet existing demand. For every new lawnmower that is produced, a previously sub-marginal purchaser will be supplied. But there is no reason to think that an increase in the supply of mowers will change the existing level of demand because the value of lawnmower to one home owner does not depend for the most part on how many other people have them.

But money is different. Unlike other goods, the supply of money influences the demand for money. The reason for this is that the services provided by money depend on the purchasing power of a single unit, while the purchasing power of each unit depends on the total supply. This will be explained as follows.

People hold cash in order to have a certain amount of real purchasing power, not any fixed number of money units. The number of money units required to provide the desired amount of purchasing power depends on the purchasing power of a single unit. But the purchasing power of each money unit depends in part on the total quantity of money circulating. If the quantity increases through inflation, prices increase causing the purchasing power of each unit to decrease. As the unit purchasing power decreases, people will need more units of it to carry out transactions at the same real prices, so money demand will rise.

Imagine that you were in another Italy before the transition from the Italian Lira to the Euro. You are used to carrying around some quantity of Lira in your wallet. Now you must determine how many Euros to carry around for the same purpose. It is impossible to answer this question unless you know the prices, in Euros, of various goods that you might wish to buy. The purchasing power of the Euro is nothing other than the inverse of the prices in Euros of goods. If a newspaper cost €1, you might carry around €10 in your pocket, while if the same paper were priced at €1000, you might need to hold €10,000 to get through your day.

If credit expansion is taking place then fiduciary media will be issued. For the same reason that money demand increases when money supply increases, money demand will increase as credit expands. But in this case, the fiduciary media will satisfy some of the demand for money. This is precisely what would happen if the RBD were adopted. As more bills were discounted and more fiduciary media would enter the system, prices in general would increase. At a higher level of prices, more credit would be needed to finance the same investments as before. The demand for money and credit to complete the same volume of transactions would increase. A self-reinforcing spiral of increasing credit supply, increase prices, and increasing demand that in the end would be limited only by the solvency of the banking system.

Here again we see the error in the idea that particular fiduciary media are “backed” by specific goods and therefore non-inflationary. The money prices of goods are formed by the interaction of everyone who has a money balance and everyone who has something to sell in exchange for money. This means that the goods in process, in the case of a non-monetized bill, have already been priced given the existing supply of money. When the bill becomes a fiduciary medium, new prices are formed, through the interaction of all money and fiduciary media in relation to the same set of goods. This will result in higher prices for the goods in relation to the new total supply of money and fiduciary media.

We turn to Mises for a restatement of this argument:

It is not true that the maximum amount which a bank can lend if it limits its lending to discounting short-term bills of exchange resulting from the sale and purchase of raw materials and half-manufactured goods, is a quantity uniquely determined by the state of business and independent of the bank's policies. This quantity expands or shrinks with the lowering or raising of the rate of discount. Lowering the rate of interest is tantamount to increasing the quantity of what is mistakenly considered as the fair and normal requirements of business.

Carroll provides a historical example:

Adam Smith supposed that an excess of convertible paper currency could not be circulated, because the excess would at once return upon its issuers for redemption. This is one of his errors, and the more surprising because of the experience of France with Law's banking sixty years before the "Wealth of Nations" was written. For four years the inflation continued there, until general prices advanced fourfold, indicating a fourfold expansion of the currency, and yet the currency did not return upon the bank for redemption to any inconvenient extent until a few weeks before its doors were closed in hopeless insolvency, although money was rushing out of the country all the time. It is a question of confidence on the part of the people; if they prefer the paper to money, and do not call upon the bank for payment, there is no difference in effect between and inconvertible and a so-called convertible currency, and, as we see in the example of France, it is easily possible to press upon a credulous community as much convertible as an intelligent people will bear of an inconvertible currency.

Inflation of consumer prices is harmful to employees and business firms for many reasons: people get inflated into higher tax brackets, retired people living on fixed incomes are impoverished, the purchasing power of wages does not keep up with prices, and others.

It is too simple to say, as Hultberg does, that Rothbard and other Austrians have rejected the RBD because it is inflationary, if they mean that a demonstration of the stability of the CPI under the RBD would rebut Rothbard’s critique. Austrian economists see inflation as more than changes in final goods prices. A further clarification of the Austrian critique of credit expansion will help distinguish the Austrian view from the RBD and show that this criticism is groundless.

While economists of the Austrian school would deplore these evils, they are have done heroic work in drawing attention to an even bigger problem. If banks can lower the rate of interest by expanding credit, one might ask, why not encourage this to enjoy the benefits of a lower interest rate all the time? Mises and later Hayek investigated the relationship between the organization of an economic system and bank credit expansion. What they found was that the below-market interest rate brought about by credit expansion is a temporary phenomenon. The below-market rate of interest distorts the productive structure of the economy, resulting in a wasteful boom-and-bust cycle. During the transition from boom to bust, the interest rate will rise to or above its market rate.

Austrian economists have been critical of inflation not only for its effects the purchasing power of money, but also because the credit cycles waste scarce accumulated savings. All credit expansion causes a credit cycle to some extent, whether or not ordinary consumer price inflation shows up. When an Austrian economist says that a monetary system, such as the RBD, would be “inflationary”, they do not necessarily mean that would result in an increase in end goods prices. Nor would it be sufficient to say in response to the Austrian that “if end goods prices did not rise under that system, then everything is fine.”

Credit expansion can coexist with stable or even declining prices as measured by inflation indexes, as they did for example in the 1920s and the 1990s. During a period of credit expansion alongside rapid investment in new technology resulting in high productivity growth, prices will not fall as fast (or not rise as fast) as they otherwise would have in absence of credit expansion. And this credit expansion will drive a boom and bust cycle.

Selgin’s Less than Zero: The Case for a Falling Price Level in a Growing Economy is an economic a history of periods during which overall prices fell due to productivity growth in excess of the growth in the supply of money. While wages did fall in nominal terms, nominal prices fell faster. Far from being anti-labor, these were periods of rising real wages. Selgin explains that prices in England fell so rapidly during 1873-1896 that economic historians who believe that falling prices must indicate a depression cannot explain the general prosperity of this period. The standard of living of laborers improved because their lower nominal wages were able to purchase more goods at even lower nominal prices.

© 2005 Robert Blumen
Editorial Archive