This archived content originally appeared at, the predecessor site to, and does not carry the sponsorship of the Cato Institute.

The Economist On Money and the State

I couldn't help being glad to see The Economist refer to Carl Menger's theory of the origins of money just as I was about to explain that theory to my undergraduate classes. Nor did I at all mind having Menger's ideas contrasted with those of another of my favorite economists, Charles Goodhart. I was, however, sorry to see that venerable publication, whose first two editors, James Wilson and Walter Bagehot, were among the more important 19th-century proponents of free banking, embrace Professor Goodhart's "Cartalist" theory of money, and do so on grounds that won't stand up to critical scrutiny.

Menger, of course, famously argued that commodity money, far from having to be deliberately designed, can evolve spontaneously or, as The Economist puts it, "is a market-led response to barter costs."

But while The Economist allows that Menger's account offers "a good description of how informal monies, such as those used by prisoners, originate," it claims that the theory comes up short when it comes to explaining the origins of the most convenient of all commodity monies: those consisting of precious metals. Why? Because metals only make for convenient money once they have been packaged into coins, and "history shows that minting developed not as a private-sector attempt to minimize the costs of trading, but as a government operation." Consequently, the article continues, "another theory is needed, in which the state plays a bigger role." Cartalism is one such theory, according to which money, and efficient money especially, is a creature of the state, which invented coins with the ulterior motive of enhancing its revenues by making taxes payable in them and by occasionally resorting to debasement.

But is it true that minting developed, and could only develop, as a government operation? Goodhart's article is supposed to offer proof that this was so, by pointing to two instances in which the collapse or withdrawal of government coinage gave way, not to private coinage but to barter (in the former Roman empire) or to the use of a non-metallic money (rice in 10th-century Japan).

One needn't be a logician to recognize the inherent shortcomings of such a "proof by example." That the withdrawal of state-run mints has sometimes failed to prompt the establishment of private ones hardly establishes that private mints have never existed, much less that they never could exist. One may, on the other hand, disprove the claim that private mints have never existed by means of a single, contradictory example.

As a matter of fact, there have been numerous episodes of private coinage, including some very successful ones. What's more, it is far from clear that the very first coins ever made–the famous electrum lumps of Lydia, in Asia Minor–were government products. On the contrary: according to Thomas Figueira, one of the leading experts on the subject today, “It is uncertain whether it was someone endowed with political authority acting on behalf of his community or an individual acting on his own behalf who conceived of the idea of coinage.” Indeed, no one is even sure what those early coins were for, or even whether they ought to be called "coins" at all, since they were uniform in weight alone, but varied considerably in their gold and silver blend.

Of course it's hardly likely than any Tom, Dick, or Harry would have been able to have his markings treated as credible certifications of weight or purity or anything else. Whoever made the first coins had to be some sort of big shot, or its corporate equivalent. Being a tyrant might suffice; but that hardly means that it was essential. Nor does the fact that almost every coin produced since the obscure beginnings of coinage has come from a government mint mean that only governments were fit to coin money. It could instead mean that governments found the fiscal gains to be had by monopolizing coinage too good to pass up. That governments frequently took advantage of their coinage monopolies, not to mint good coins, but to mint lousy ones that they then compelled their citizens to accept, would seem to favor the latter hypothesis.

It ought to go without saying that there is no technological reason why coins couldn't have been a private invention, or couldn't have been privately manufactured at any time to the same standards, if not better ones, than their government-made counterparts . "A mint, Sir" Edmund Burke once reminded Parliament, "is a manufacture, and it is nothing else." A factory, we would now say. And since when are governments very good at, let alone uniquely capable of, running factories? As for state-sponsored enterprises generally doing a better job of quality control than their private-sector counterparts, if you believe it I must see about coming up with a few tons of ca. 1958 Chinese steel to sell to you.

In fact, all of the most significant coinage-related inventions–the manual screw press and its steam-powered counterpart are only the most famous examples–came from the private sector, and most were taken up by governments only reluctantly and after (sometimes deadly) resistance from government mint authorities. How often, on the other hand, have government authorities themselves been responsible for technological breakthroughs? (No, Tang wasn't invented by NASA.) Were it to come to a wager, I for one would much sooner keep a grip on my money than stake it on Croesus or Pheidon or any other ancient king.

But why speculate? In fact, Goodhart's contrary suggestion notwithstanding, there have, as I've already noted, been occasions on which coinage was handled by the private sector, and in the best documented ones the coins that resulted were not only as good as but superior to those from the government's own mints. That was certainly the case in the U.S. after the Appalachian and Californian gold discoveries, when private mints sprung up to supply convenient coining services to miners who would otherwise have had to send their gold to Philadelphia or (after 1835) to either Philadelphia or Charlotte for coining. The high quality of the private gold coins produced during these episodes is attested to, both by the extant coins themselves, and by the fact that the U.S. Mint, having failed to put the West Coast mints out of business merely by finally opening a San Fransisco branch, relied instead on coercion to do the trick.

Another instance of successful private coinage is one with which Professor Goodhart is now familiar, though he didn't know of it in 1998, when he published the article to which The Economist refers. It is the episode of private token coinage in Great Britain that is the topic of my book, Good Money. Professor Goodhart did me the honor of writing that book's introduction. And although he makes clear there that the story I tell did not at all cause him to abandon Cartalism, I am sure that he would agree that it proves that, in at least one instance, the state's withdrawal from coining did in fact lead to private entrepreneurs rushing in to fill the void.

That private coiners didn't always or even often rush in whenever governments failed to supply coins is itself hardly surprising in light of the fact that unauthorized coining, even of coins not meant to imitate official ones, has almost always been illegal, and has more often than not been a capital crime. So the absence of acknowledged private mints following both the fall of Rome in the 5th century and the Japanese government's abandonment of copper coinage in the 10th might prove nothing more than that no one wanted, literally, to stick his neck out.

Might. Except for the fact that the claim that no private coinage took place in either of these instances doesn't seem to be true. I do not merely mean that there was surreptitious private coining, that is, counterfeiting: despite the death penalty Japanese copper coins were aggressively counterfeited. I mean that after the Japanese government got out of the business of making coins–after having, that is, debased its coins until they were more-or-less worthless–legitimate privately minted substitutes, manufactured by local clans and wealthy merchants, did in fact take their place, along with Chinese coins and (yes) rice. That, at least, is what it says on the website of the Bank of Japan's Currency Museum. Nor is it true that coinage simply gave way entirely to barter after Rome fell. No doubt it did so to some degree everywhere, and perhaps to a large degree in some places; but private coinage also took place, and did so notoriously in Merovingian Gaul, where thousands of local mints supplied coins modeled (sometimes crudely) on their Roman predecessors, and bearing the names of coiners very few of whom were known rulers. In short, it seems that, even as proofs by example go, those offered by The Economist are rather paltry.

Yet I suppose that we will never see the end of the myth that only governments are fit to coin money. Were bread a government monopoly long enough, Herbert Spencer once remarked, people would react with horror to the suggestion that it might instead be supplied, and supplied with better results, by the private sector. Spencer was probably right. I'm just glad I'll never live to see The Economist prove it.

  • Paul Marks

    James Wilson and Walter Bagehot were related (by Bagehot marrying the daughter of Wilson), but their opinions were rather different.

    Wilson was someone who worked for a smaller government, whereas Bagehot was (in "The English Constitution" and other so called "classic" works) was in favour of "conceding whatever it is safe to concede" to the forces of collectivism.

    It is, of course, "safe" to concede nothing – partly because this moves the discussion from REDCING the size of government to INCREASING, and partly because small increases in statism do not produce the results that are hoped from them (in fact they produce new problems), and the collectivists respond with demands for larger scale interventions (see Ludwig Von Mises on Interventionism – or "Destructionism" as he calls it in the last section of his work "Socialism").

    Specifically on free banking – Walter Bagehot did indeed muse about Free Banking in theory, but in practice he argued for bailoutism (AGAINST the then GOvernor of the Bank of England – who was opposed to bailing out banks).

    Of course Walter Bagehot was not like the modern Economist magazine – the bailoutism he supported was on a very small scale and under strict conditions (the Econoimist magazine supports de facto unlimited bailoutism, with no real conditions), but his work does show a decline on the Classical Liberalism of Wilson and it is possible to show a process of change by which the flaws in Bagehot develop into something as radically wrong as the modern Economist magazine.

    And it is radically wrong – and not just on monetary theory (or monetary history – ignoring the successuful private mints that dominated the American West till Congress banned them in the 1850s).

    It is even wrong on the much simpler matter of fiscal matters.

    For example, the Economist magazine regularly (almost every issue)
    refers to radical government spending cuts in Britain – British government spending is, in fact, going UP.

    The Economist magazine also refers (again in issue after issue) to the radical cuts in the Welfare State (I am using the German and British term – but American government programs have developed so radically since the 1960s that it is legitimate to refer to the United States as also a Welfare State) suggested by Paul Ryan – in reality Congressman Ryan has argued for reducing the RATE OF GROWTH of government Welfare State spending, NOT "cutting" government Welfare State spending (there are people who support real cuts in the Welfare State – for example Senator Rand Paul, but the magazine does not consider their proposals).

    To confuse reducing the rate of growth of government Welfare State (health, education, and income support) spending with "cutting" the entitlements and so on, is an error so basic it is hard to overstate it.

    A magazine that does not understand basic fiscal policy, can not reasonably be expected to understand monetary policy or monetary history. As monetary matters are far to understand (and to explain) than fiscal matters.

  • Peter Surda

    Some Austrians might argue that, contrary to the claim presented in the referenced The Economist article, good money should hinder, not help, the government in taxing its citizens (see e.g. Michael Suede, with whom you exchanged emails in the past, or Jon Matonis, who often writes for Forbes). At least with Bitcoin, we know for a fact that it was not the government that started the coinage thereof. That honour goes to Mike Caldwell AKA casascius.

    • George Selgin

      I think the distinction that most matters here is that between anonymous payments media, mainly consisting of paper currency (though Bitcoin also qualifies), and media that leave a "paper (or computer) trail" of transactions. Payments made and received in Federal Reserve notes are anonymous, and so impossible for the IRS to monitor, while those made and received using transferable private-bank deposits aren't.

  • Bill Stepp

    Contrary to Mr. Goodhart's claim that "it is incredibly hard to value raw materials," the market does this quite easily. Traders did so even before organized futures markets existed.
    And I don't see why the use of barter after the collapse of Rome and turmoil in Japan implies that money needs the State to come into existence. Just because the use of money ceases for a short time doesn't mean that only the State can create it.

  • BillWoolsey

    One problem with histories of coinage is that some historians only count the coins as money if they are somehow official legal tender. The use of privately minted coins _is_ barter.

    Perhaps I am wrong, but my understanding is that Babylonian silversmiths made ring money (which were like spirals) which had to be broken off and weighed, but then made money that looked like little loaves of bread which was marked for weight.

    And what about Phoenecia? My understanding is that their money was private enterprise and coined. But because it wasn't governmental, they supposedly used barter.

    It was greek goverments that made round flat coins. And in the greek version of monetary history, they invented coinage. Or rather, their neighbor, the King of Lydia. By the way, didn't become King by marrying the queen and then killing the previous King? And didn't he have magic powers?

    On the other hand, I am not sure that rice or some other staple grain is all that bad as money for the typical ancient subsistence economy. The initial shift to metals, and then the progressive shift to more and more valuable metals, ending with gold, is unlikely to have been driven by convenience for the common people. The ability of agents of the state to store wealth when there were very primative financial markets, and then the operations of gresham's law due to fixed exchange rates between coins made of different metals seems more plausible.

    • George Selgin

      I agree wholeheartedly with your first point, Bill, about historians and others often refusing to recognize private coins as money at all. As for the point about rice versus gold, you also have a point in saying that government's played a big part in the rise of the gold standard. Still, I think that market forces also tended to favor the adoption of more valuable money commodities as the scale of transactions grew over time.

    • David Stinson

      " some historians only count the coins as money if they are somehow official legal tender. The use of privately minted coins is barter."

      George/Bill: What was their rationale for taking that view?

      • George Selgin

        I suppose you could say that they are (implicit) "Chartalists," holding money to be only those things that are so designated officially. In my own research I found that the private coins issued during Great Britain's industrial revolution are more often referred to as "tokens" than as coins, with the latter term limited to products of the Royal Mint.

  • George,

    Again another excellent article!

    As this discussion is part of my lineage, my ancestor Rüdolf Räber was a private minter from Aarau, Switzerland in the 13th century and his father before him in the mid-13th century in Basel, Switzerland. The minting they did was for the Templars as well as for the German nobles.

    Michael Reber

  • George Selgin

    Thanks Michael for both the compliment and the fascinating (and supporting) empirical tidbit.

  • This might be connected to Bill's point above, but it seems to me that Cartalism is a tautology. Cartalism is false. The proof is by contradiction. For example, the WSJ recently reported that Germans are using deutsche marks again:

    "Germans have yet to give up on the euro. But as Europe's debt crisis rages on, many are indulging their nostalgia for the abandoned mark by shopping with it again—and retailers are happily going along."

  • I could see people involved in a trade for gold taking the gold to a trusted metal worker for verification and that leading him to make coins.

  • Roger Koppl

    George: The Rome example may not be so good for another reason as well. It is not true the barter replaced monetary exchange *after* the fall of the Roman empire. Monetary exchange pretty much fell apart in the 3rd century, and it do so before Diocletian came along. The army was supported with in-kind taxes. Diocletian regularized the ad hoc requisitions of his predecessors. He also put in a series of other reforms that created the medieval economic system of Europe. These reforms created a peasant class, basically by tying sharecroppers to the land. Bruce Barlett has a nice rundown on these facts:

    • George Selgin

      Thanks for this, Roger.

      My own understanding is that Diocletian succeeded to a considerable extent in reviving Roman coinage, in part by decentralizing coin production, and that, although silver coinage ceased after he abdicated, the minting of both gold and bronze coins went on until the 5th century, when official Roman coining came to a halt in the West. It was during that century that the Merovingians began to strike their own mostly gold and silver coins, modeled on late Roman and Byzantine ones.

  • George: thanks for a good piece. It is frustrating to watch the many hacks and shills always saying that we would lack for air to breath and our hearts would stop beating, but for the fortunate intervention of government!

    I have one more argument against the alleged necessity of government minting. Let's say there is a government-minted gold coin. It has two attributes: (1) it is government-minted, (2) it contains a known quantity of fine gold. I think it is a simple matter to show that its value comes from the latter attribute and not the former. Proof? In 1971, when Nixon "closed the gold window", some economists predicted gold would lose its monetary value and fall to $7 per ounce(!) The opposite occurred, and this is proof that the coin has value because of its gold content not because of its government stamp.

    Another way to look at this: what happens to the value of old paper money notes when the government is no longer in power? What happens to old gold coins when the government is no longer in power?

    To assert that the government stamp on the coin is what gives it its value is disingenuous, at best!

  • Mike Sproul

    Don't forget that wooden tallies, clay tablets, and various other forms of improvised bookkeeping have been used as money for at least as long as coins have been used. I'd guess that coins have normally been a fairly small part of the money supply, just as they are today.

  • I wouldn't dispute Selgin on the history of coinage, but shouldn't we attribute the Chartalist theory of money to Georg Knapp? Keynes attributes the theory to Knapp in The General Theory, published the year Goodhart was born.

    I believe that money evolved statelessly, and I can believe the coinage did too, but modern money clearly is not stateless. Knapp's "State Theory of Money" might rather be called "The Theory of State Money". Chartalism is clearly the correct theory of the money in my pocket. The question is: Can stateless money ever replace state money now?

    • George Selgin

      Yes, Knapp coined the term "Chartalism" (now also rendered as "Cartalism"). I don't think anyone would disagree that the money we use is state money. But the claim of cartalism is that money is essentially a creature of the state, which can confer monetary status on any sort of token or paper, without having to initially promise eventual redemption either in a commodity or in some pre-existing non-commodity money. I've questioned the theory in these two academic articles, but am now more inclined than I once was to allow that there are cases that seem to fit it.

      • David Stinson

        Hi George. Another excellent post.

        According to wikipedia, cartalism is closely related to the dreaded MMT. I am assuming that perhaps wikipedia overstates the extent of the connection? Has Goodhart ever addressed MMT?

        • George Selgin

          I am sure that the Wiki article was mainly written by MMTers.

        • What is the MMT precisely? How do you distinguish it from Cartalism? Modern money clearly is not a commodity with a value determined by its utility, other than its utility as the exclusive account of statutory obligation and as money itself, as well as scarcity. State spending and taxing (and selling entitlement to tax revenue) does affect its circulation very significantly. Considering how much of my own income I pay in taxes and other statutory rents, and how much of it state spending ultimately supplies, I can hardly doubt this theory, and this theory is Cartalism as I understand it.

          Of course, the fact that modern money is a chit commanded into circulation by a state does not imply that money has always been or could only be or ought to be such a chit. Even if Cartalism does explain the origins of money better than market-oriented theories in most historical contexts over recent millenia, I still favor a market-oriented approach. At least, I favor free experimentation in this direction. Chattel, hereditary slavery was common enough in the same historical contexts, and we now widely agree to leave it there. Advocates of a statutory enactment, no matter how precedented, always bear the burden of defending it in my way of thinking.

  • David Stinson

    According to a Bank of Canada history of the Canadian dollar:

    In the 19th century "brass and copper tokens circulated alongside legal tender coins and helped to offset a shortage of low-denomination coins, useful in small day-to-day transactions. With a face value of a half a penny or penny, tokens were widely distributed by banks, non-financial companies, and individuals." (see

    • George Selgin

      Thanks for this–I had not known of the episode, which affirms my assumption that were banks exist and have branches, they are natural emitters of token coins.

      • David Stinson

        "where banks exist and have branches, they are natural emitters of token coins."

        Hardware stores, too. 🙂

        • The Canadian Tire Company presumably doesn't qualify as any Tom, Dick, or Harry, but Satoshi Nakamoto seems to fit the bill.

  • I have speculated that at least some pre-Columbian art may have served as money due to manifold specimens of small similar statues.

  • John S

    I just watched your lovely lecture, "The Private Supply of Money" (Hayek Memorial Lecture) on Youtube. Have you contacted the History Channel, PBS, or the BBC about making a documentary on the private coinage episode? I think it would be a wonderful subject, esp. with the beautiful images.

    • He first needs to relate private coinage to Hitler or ancient astronauts.

  • Bill Stepp

    Scroll down a bit at the link below for William Luther's correction to The Economist's article.

    I'm not surprised that gun control advocates believe in the State theory of history.