In 1892, Carl Menger wrote On the Origins of Money—the original basis of Austrian monetary theory. This essential work lays out a theory as to how monies emerged through voluntary human action and exchanges on a free market. Menger also critiqued other monetary theories—money from social compact or civil edict—as unhistorical. He wrote,
Nevertheless it is clear that the choice of the precious metals by law and convention, even if made in consequence of their peculiar adaptability for monetary purposes, presupposes the pragmatic origin of money, and selection of those metals, and that presupposition is unhistorical….
…no historical monument gives us trustworthy tidings of any transactions either conferring distinct recognition on media of exchange already in use, or referring to their adoption by peoples of comparatively recent culture, much less testifying to an initiation of the earliest ages of economic civilization in the use of money.
Here it might appear, however, that Menger has—through his specific critique—left his own monetary theory open to a similar critique: the regression theorem too is not based on empirical historical evidence, but rather applying economic principles in the present to determine the origin of money in the past. In other words, Menger could be accused of making assumptions about what happened in history based on theory without historical evidence, just like the theories he criticized. Is this critique valid?
A Review of Austrian Monetary Theory
Following Menger’s work, Mises solved the circularity problem of money’s value by arguing that the original price of a money was established by its previous exchange ratios to other commodity goods in a barter economy. This gave it purchasing power as a medium of exchange. In discussing the validity of Mises’s regression theorem of money, Rothbard spoke of “the last day of barter.” By this he meant that, theoretically, before the point in time when gold or other commodities were used as media of exchange or money, rather than just valued commodities, they were goods in a barter economy—directly exchanged for one another. Before they were used as money, their prices were determined by supply and demand and could be expressed in an array of other goods or fractions of goods for which they could exchange. For example, the price of an ounce of gold prior to money in a barter economy is whatever full or partial goods or services for which it will exchange at that moment.
Due to the inherent limitations of a barter economy, valued goods with particular characteristics—scarcity, divisibility, portability, durability, recognizability, fungibility, high value per weight, etc.—began to be used for indirect exchange. (Note: goods without those characteristics could also be used for indirect exchange, but it would be unlikely that they would become generally-accepted media of exchange). Direct exchange is the voluntary and mutually-beneficial action where individuals give up what they want less to get what they want more from another willing actor in an act of trade (without coercion or fraud). Free exchange depends on subjective valuation and disagreement over value between the participants—each one values what the other has more than what is given up in exchange. Mises called this “reciprocal surrender.” While this is mutually beneficial, it is inherently limited because each side of the trade must want what the other has more than what is given up and possess the goods or services another wants more than what he possesses. Additionally, they must know about each other. Under direct exchange, only certain exchanges can take place and there can be no economic calculation.
Indirect exchange occurs when people recognize that there is general demand for certain goods that had the qualities described above (i.e., scarcity, divisibility, etc.), therefore, instead of direct exchange, people exchange their goods and services for those goods—not to consume them—but to trade those goods again to obtain a greater variety of other goods. For example, if someone wanted apples, a chair, shoes, etc. and had a horse to exchange, that person could exchange his horse indirectly for gold (or another valued commodity with similar characteristics) in order to then use the gold to obtain the other goods in the present or future. This would be done because the individual, while he may not have a personal use for gold, knows other people want it and will be willing to trade other goods for it. As gold, in this example, becomes demanded—not just as a consumer good, but to be used indirectly in further exchanges for other goods and services—it can become a generally-accepted medium of exchange or money. As more people see the value in using gold for indirect exchange, as well as direct exchange, the recognizability of gold as money spreads in a network effect. The original purchasing power of an ounce of gold as money depended on the array of goods and services (or fractions thereof) for which gold could have been exchanged in barter in the immediate past (e.g., the day before), thus a logically-complete explanation for the origin of money’s purchasing power.
Previously, monetary theory ran into a problem of circularity or infinite regress when it came to the question of the origin of money’s purchasing power—money has value because it’s accepted in exchange and it’s only accepted in exchange because it has value; money has value now because it had value before. Mises—extrapolating and expanding on Menger—solved this through his regression theorem, that is, that money today has value because—if we regress back through time and apply the principles of catallactics—then people must have started using money-quality commodities for indirect exchange and its original value came from its purchasing power as a barter commodity.
This is what Rothbard is getting at when he describes “the last day of barter.” Where did money get its original value the first day it was used as money? The answer was to look to all the goods and services for which that good exchanged as a commodity the day before on “the last day of barter.” Rothbard wrote,
…when gold first began to be used as a medium of exchange, its marginal utility for use in that capacity depended on the existing previous array of gold prices established through barter. But if we regress one day further to the last day of barter; the gold prices of various goods on that day, like all other prices, had no time components. They were determined, as were all other barter prices, solely by the marginal utility of gold and of the other goods on that day,…
The determination of money prices (gold prices) is therefore completely explained, with no circularity and no infinite regression. The demand for gold enters into every gold price, and today’s demand for gold, in so far as it is for use as a medium of exchange, has a time component, being based on yesterday’s array of gold prices. This time component regresses until the last day of barter, the day before gold began to be used as a medium of exchange. (italics in original)
The Critique against Menger’s Method
Even some Austrians have raised the point that Menger’s monetary theory may be overly speculative, even if it is most likely that money did emerge as Menger, Mises, and Rothbard posit. Gary North argued that—while he did believe the regression theorem was most probably the way money came to be—Hobbes, Locke, and Rousseau also imagined a “state of nature” in which they speculated about the origins of government without historical evidence. In their conjectural histories, they imagined histories with no documentary evidence and came to different conclusions on what must have happened. Gary North observes,
It is not that Menger and Mises laid aside the facts. It is that there were no facts to lay aside. The explanation which they gave for the rise of the money economy is reasonable, in the sense that it is difficult for an Austrian School economist to reject. He does not believe that a fiat declaration by a political sovereign would have been capable of changing the self-interested behavior of individuals in a barter-based economy.
It is important to note that North was not critiquing the validity of the regression theorem, but the necessity of it. The argumentation is valid. Mises basically argued a sort of transcendental argument: p is a necessary pre-condition for the existence of q; since q, therefore, p. This is an argument from the impossibility of the contrary, which is valid, but there must be a true connection between the necessity of p and the existence of q, with no alternatives. The question is whether money necessarily had to develop according to the way Menger and Mises described. North concluded, “The Menger-Mises regression theorem is not historical. It is developmental. It is an example of conjectural history. It seems more consistent with known human behavior than the theory of state-created money.”
Clarifying and Defending Menger’s Argument
What has to be established here is the precise nature of Menger’s argument and whether or not he engages in an unreasonable double standard—holding other theories to the standard of available historical-empirical evidence while admitting his theory lacked historical evidence. We also have to determine whether we should expect the same level of historical verification from different claims of different monetary theories. Reading more carefully, Menger wrote,
The idea which lay first to hand for an explanation of the specific function of money as a universal current medium of exchange, was to refer it to a general convention, or a legal dispensation….
Tested more closely, the assumption underlying this theory gave room to grave doubts. An event of such high and universal significance and of notoriety so inevitable, as the establishment by law or convention of a universal medium of exchange, would certainly have been retained in the memory of man, the more certainly inasmuch as it would have had to be performed in a great number of places. Yet no historical monument gives us trustworthy tidings of any transactions either conferring distinct recognition on media of exchange already in use, or referring to their adoption by peoples of comparatively recent culture, much less testifying to an initiation of the earliest ages of economic civilization in the use of money.
What we see from Menger is not an unfair double standard, but a careful distinction between what should be expected if certain monetary theories were true. If it was true that money came into existence via social compact or government fiat, those types of events—in every place that began to use money at one point in history—we would expect that to be historically verifiable as public and society-wide. On the other hand, the Menger-Mises theory posits that, through human action and exchange—catallactics—which we would expect to be historically mundane, subjective, individualized, and either not considered historically noteworthy or even historically traceable (at least down to the first moment).
Kristoffer Hansen, responding to North, observes that the absence of direct empirical evidence does not weigh equally against all theories of money. In Menger’s theory, lack of historical evidence is unsurprising while it is surprising in the theory of chartalism. Menger’s deductive-conjectural monetary theory—developed further by Mises and Rothbard through the regression theorem—reasoned from general economic laws of exchange and presupposes that the earliest uses of goods in indirect exchange would have been subjective, decentralized, private, and historically mundane. On such a view, it is unsurprising that the earliest transitions from barter to media of exchange left little documentary evidence. However, the same cannot be said for chartalism.
If money originates through state action—specifically through the issuance of a fiat-token and the imposition of tax obligations payable in that token—then this process should be historically salient. Hansen writes, “On the other hand, Menger was right to cite the absence of evidence against the state theory of money, since it is the type of historical conjecture that needs historical evidence to appear plausible.”
State origination of money, according to chartalism, is not a mundane or decentralized process; it is an institutional and political event, especially in the more recent and more literate societies of colonial America. Ironically, it is arguable that the historical evidence we do possess weighs heavily in Menger’s favor and that critiques of barter and money’s emergence from barter actually critique a neo-classical oversimplification, not true barter.
If chartalism describes the origin of money, then it must be possible to identify historical moments at which money did not yet exist and subsequently came into existence through token issuance and acceptance of taxes in said token.
