This is just a preliminary post trying to explain the shape of the problem. I’m nowhere near a solution. If or when I have one I’ll probably publish it on Spirit.

When Capital was written and while Marx was alive, all money was conceive of as a commodity: chunks of gold or silver, or for the slightly more adventurous, claims on chunks of gold or silver. There are treatments of credit and paper money on Marx’s work, but it’s impossible to deny that the focus of his view of money is as a universal commodity, where it acts as a sort of abstract form of barter.

Now, it’s important to point out that on the first chapters of Capital v 1 (the boring bits) Marx draws important distinctions between pure barter and monetary exchange. That’s partly what all those rather obscure forms are about. In a monetary society, there’s a process of socialisation of production. The universal commodity becomes in some sense the stand-in for use value as such. If one exchanges gold for a loaf of bread, it is stamping socially recognised value on the bread. So unlike neoclassicism (where money is just a convenient intermediate means of exchange for barter) Marxist economics does consider exchange has a different character in societies where it is an incidental event, and those where production is governed by monetary gain.

Nonetheless, and in spite of the particularities of monetary societies, and issues of formal/real subsumption and hybrid or overlapping modes of production¹ etc, the story Marx presents on how money arises is very similar to what we can read from Adam Smith or other classical economists:

• First there is a simple society where everyone produces the things they need in a more or less undifferentiated way.
• Later, increased division of labour results in people having more of some things than they need and less of others, so they engage in barter.
• Barter operations become increasingly cumbersome as division of labour proliferates, and people need to keep stocks of different goods in order to successfully bargain for what they need.
• Some particular commodity (or small list of them) becomes imbued with the character of universal exchange, so people begin to realise exchange through its mediation.
• Money arises in its full form by displacing barter and becoming measure of value, store of value and means of exchange, from which contradictions many problems of capitalism result.

There’s a minor problem with this account of the rising of money: it is historically false. Whereas the idea of exchange becoming normative through repetition seems to hold some validity, the notion that pre-monetary societies engaged in generalised barter in order to fulfil their needs is today discounted by all the records, historical as well as anthropological. One of the first pieces to push back against this view, What is Money, presents an entirely different account from that of Smith and Marx, which turns out to be closer to the truth.

Graeber, on Debt, treats this topic in far more detail. In essence, there are two primary ways to look at the role of money. One of them starts from barter, with money serving as a means of exchange, and does not really hold. The other one starts from debt as a quantified social obligation, and results in conceiving of money as a unit of account. The primary problems of the money-as-universal-commodity view are the following:

• Barter is extremely rare within social groups, taking place only in situations of very low trust between societies, not within them, or collapsed monetary societies that have run out of viable money.
• Historically, the use of actual commodities (as opposed to notations on ledgers, or tokens) in order to settle transactions, is rare and uneven.
• The first appearances of money are linked to quantification of debt within temple bureaucracies, not settlement of private payments.
• Today’s societies run entirely without reference to a universal commodity.

Marx would have had the misfortune to be alive while the view of money as commodity was particularly hegemonic. Unfortunately some of this is material to Marx’s theory of exchange, in that he asserts without the existence of a universal commodity that can have some kind of commonality in exchange to the other commodities, exchanges of equal value would not hold, since this substance of value would require their expression in the two things being exchanged. Simplifying, if I say a loaf of bred is worth a gram of gold, it is because there is something which is socially stamped on both that is equal: for Marx, socially necessary abstract labour time. However, if I say a loaf of bread is worth a euro, I’m in trouble, because euros are nominal units that can be printed with almost no regard to labour time (their marginal cost is near zero) and certainly their capacity to confront commodities in the market seems extremely decoupled from the labour they embody.

This problem of paper money was something other Marxists were forced to confront. For example, Hilferding, on a chapter in Finance Capital called Money in the Circulation Process investigates the role of paper money as he knows of it. Paradoxically, and contrary to what the models were predicting, the use of labour money could produce a decrease in prices. Hilferding tries to think of this problem in terms of socially necessary exchange, so that the money in circulation at any given time would, ideally speaking, confront the sum total of commodities in the market and derive its value from its ability to be exchanged against them.

This solution is not entirely satisfactory. Kautsky, in a more orthodox work, Gold, paper currency and commodity, attempts to present some counterarguments to it, and concludes any long-term regime where paper money comes to substitute actual metallic money is unsustainable. The problem is facts are stubborn, and we’ve had a capitalist world system running for at least 50 years without reference to commodity money, and capitalism isn’t clearly the worse for it. What’s more: even when capitalism was nominally bound to commodity money, it is difficult to determine how much of this was a matter of legal fictions and tradition, and how much had an actual economic impact.

Modern monetary theory is an attempt to make sense of modern economies based on fiduciary money on the basis of the actual operations carried out by issuing banks, the banking system, and the rest of the economy as a whole. While it contains some prescriptive recommendations to money sovereigns (including a job guarantee!) it is, at bottom, a descriptive discipline. Considering neoclassicism doesn’t even want to think of money (don’t look at it, it’s just an intermediate representation of utility gradients), MMT is our best hope to work out the role of money as it has come to be.

Under MMT, however, money is a purely symbolic category. Money is, so to speak, an informational component: something which has value because the state has made solemn pronouncements that it will accept it as means to solve public and private debts. So money is, under this account, a creature of the state, and more specifically a complement of the power to tax.

This is entirely at odds with the description of money as a universal commodity that objectively embodies socially necessary labour. How many mines are required in order to publish a law in the state gazette? It leaves us without a good theory of exchange, and with all of the carefully worked out commodity forms at the beginning of Capital turning into nonsense.

So the question is if we can get out of this quandry. Is it possible to synthesize Marxism and MMT? Before going further I should point out there are other things that may be attempted instead:

• Some Marxists have tried to find an implicit basis in commodity money for today’s fiduciary money. They often refer to the oil-dollar link, for example. I find their explanations ultimately unsatisfying.
• One could also attempt to think of money in terms of guard labour. Money would contain value inasmuch as labour is used in order to persecute forging, making its reproduction more difficult. One could, in fact, add to the guard labour, the necessary labour in evading it.
• Another option is pretending Marx was being metaphorical, not historical, about the rising of money, and that we can just ignore all the references to gold. This doesn’t solve how we end with an objective theory of value in the end, however.

There’s some ongoing work on whether Marx’s use of commodities can be circumvented without bringing down with it the whole value theory. I’m doing some of this reading, while trying to work out in my mind how to reconcile these very different approaches. While I think socially necessary exchange can have the germ of the solution, it is unclear to me how this social necessity is determine. There have been cases of extreme dislocations in exchange, due to hyperinflation or deflation, that are not necessarily accountable to the money supply. In this regard, Marx was correct not to be a quantity theorist,² but socially necessary exchange does not answer what are the preconditions and causes for money to be accepted as bearing value, and how much value. Circulation is taken as an independent variable, and this looks unlikely to me.

Slight excursus: can bitcoin help our understanding? It is quite an interesting experiment, as much as it is economically, socially and technically flawed. It embodies a stamp of labour, although the price fluctuations of bitcoin seem to be relatively unrelated to the amount of labour required to produce it. It also is grounded on a quantitative view, so that the number of units produced is carefully bounded, and there is certainly no state solemn declaration that would make it fiduciary money. Bitcoin is a curious return to some form of money commodity (computation, or more indirectly, energy), although some people have tried to reject it on the grounds that it lacks intrinsic value, or what we might call use value. It is here where one might try to deploy an idea of socially necessary exchange: could we say bitcoin’s value is given by its community, inasmuch as it is determined by the amount of bitcoin in circulation as it stands against the commodities one may purchase with it? Still, this is unsatisfactory because it does not give us good objective bounds for such a social necessity.

So, that’s the presentation of the problem. If I get anywhere closer to a solution you’ll here more of me on this topic.