Blog

1. Introduction

I am very grateful for the invitation to contribute to this blog dedicated to the late Professor Augusto Graziani.¹ Earlier this year (in May 2024) a memorial conference for Graziani was held at the Università degli Studi del Sannio di Benevento, Italia, to mark the 10th anniversary of his death. I did not attend the conference, but I was interviewed at the time by Giuseppe Fontana and Rosa Canelli, who asked me such questions as when I first met Professor Graziani or came across his writings. Also, what has been the impact of Graziani’s work on my own contributions and thinking about the functioning of modern economies. I believe that a clip, or clips, of that interview were eventually played at the conference.

Augusto Graziani was an international scholar of the first rank, the highest rank, and a personal inspiration to all those who knew him well. It was an honour for me to be able to contribute those remarks for his memorial. What I would like to do in this blog post is to repeat some of the same remarks, and take up an issue that I, for one, would like to know more about in relation to Graziani, namely the connection between his work and that of Karl Marx.

2. Augusto Graziani and the Theory of the Monetary Circuit

I must have first met Augusto Graziani at a conference in England in the early 1990s. I can’t remember exactly when or where, probably in London or at Cambridge University. What I do remember is being impressed by the paper he gave, and then having the good fortune to be seated next to him at the conference dinner. I learned a good deal that evening. It was around that same time that I started to be in relatively close touch with Giuseppe Fontana, who had been Graziani’s student, and from Giuseppe I was able to gain a somewhat more detailed understanding. As for Graziani’s writings, very important to me was a pamphlet published in English around 1989, which was later reprinted as a journal article (Graziani 1990). This was called ‘The theory of the monetary circuit’. This is where Marx comes in, as the phrase ‘the monetary circuit’ was initially due to Marx. The original German was ‘Der kreislauf des Geldkapitals’, where Geldkapitals means something like ‘money capital’. A later contribution was the publication of Graziani’s Frederico Caffe Lectures in book form, The Monetary Theory of Production (Graziani 2003). The reference here, evidently, is to Keynes. In that same vein, another publication that should be mentioned is an article in the journal Economic Notes (an Italian journal published in English) which was entitled ‘The debate on Keynes’s finance motive’ (Graziani 1984).

From my point of view this idea of the monetary circuit, and the notion of endogenous money, is fundamental for a correct understanding of what might be called (very broadly speaking) the ‘Keynesian’ and/or ‘Post Keynesian’ approaches to monetary macroeconomics. It is certainly a crucial component of my own macroeconomic system – however one might like to refer to that. I am referring particularly to the role of credit creation (and hence money creation) that is necessary for the generation of the monetary profits (my emphasis) on which the whole system depends.

By the time I met Graziani, I was already familiar with the various French versions of circuit theory. Omar Hamouda and I had invited Bernard Schmitt to the Keynes and Public Policy conference in Toronto in 1986. Shortly afterwards, I got to know the late Alain Parguez (and I think that the 1986 conference was the first time I heard his name). Alain was a frequent visitor to Canada, and Ottawa in particular. Therefore, when I finally encountered Graziani, that was a somewhat different approach to me. I was particularly struck by the clarity of his exposition, which helped me to further develop and solidify my own ideas on the topic.

3. The Ontological Status of Various Terms in the Circuit

In my conversation with Rosa and Giuseppe, I suggested that one of the most important questions to be resolved is what are the main are the differences between Graziani’s approach and that of others such as Alain Parguez. One way of putting it is that if we think of the monetary circuit in terms of Marx’s:

(1)                                           M – C … C’ – M’,

what is the ontological status of the M terms (standing for money) compared to that of the C terms (standing for commodities)? I am using the English symbols, of course. I remember that I once had a 2-hour conversation with a francophone person that was very much at cross purposes. As I remember, M stood for commodities, ‘marchandise’ or something like that, and A ‘argent’ stood for money. It was all very confusing.

In any event, the way in which I would interpret equation (1) is that entrepreneurs start the production process with a sum of money, M, on hand. With this they then acquire some commodities (including labour power), C. Next, they engage in production, illustrated by the sequence of dots (…) in equation (1). They use the initial C to make more (i.e., more valuable) commodities C’. The term (C’ – C) thus corresponds to the real value-added in the economy. Next, the entrepreneurs sell the enhanced commodities, C’, for more money M’.

Therefore C’ minus C represents what modern economists call real GDP and Marxists would call surplus value. Meanwhile, M’ minus M is the total of aggregate money profit. There are several important questions that can be asked about the relationships between these magnitudes. Most notably, how it is even possible for M’ to be greater than M, and hence for monetary profits to exist? This clearly cannot happen at the aggregate level if the money supply is fixed. And yet this is often precisely the assumption that is made in the macroeconomic textbooks. I have argued that the logical answer can only be credit creation, and thereby money creation, by the banking system (Smithin 2016). Marx himself did not seem to really focus on (what I think) is this all this important question and was possibly hampered from doing so because of the underlying assumption that money is itself a commodity. If C is a commodity and M is a commodity, then what is the ontological difference between them?

Meanwhile Keynes (1933, 81), writing in 1933 as he was preparing his own major work, The General Theory of Employment Interest and Money (Keynes 1936), for publication, had this to say about Marx’s monetary circuit:

            The distinction between a co-operative economy and an entrepreneur economy bears some relation to a pregnant observation by Marx – though the subsequent use to which he put it was highly illogical. He pointed out that the nature of production in the actual world is not, as economists seem often to suppose, a case of C – M – C’ … of exchanging a commodity (or effort) to obtain another commodity (or effort). That may be the standpoint of the private consumer. But it is not the attitude of business, which is a case of M – C – M’ … parting  with money for a commodity (or effort) in order obtain more money.

Keynes is by no means endorsing Marx or Marxism.  He says that the use to which Marx himself put this otherwise useful idea is ‘illogical’. Nonetheless, it is evident that Keynes thought that M – C – M’ (continuing to use Keynes’s truncated version) is a crucially important insight. It is fascinating to speculate on just what it was that Keynes’s thought was illogical about Marx’s treatment. Perhaps it was this same issue about whether money is also a commodity or not? The first three chapters of Keynes’s Treatise of Money (Keynes 1930, 3-43), after all, had put forward a different ontology of money, focusing on the money of account, which runs very much along what we might now call ‘creditist’ or ‘chartalist’ lines.

More recently the Cambridge University economic sociologist and political economist Geoffrey Ingham (Ingham 1996, 2000, 2004) has argued that money is a ‘social relation’ in its own right, specifically a social relation of indebtedness, with its own mode of social valuation within ‘a separate and relatively autonomous monetary sphere’ (Ingham 2003, 61). In his last book, the posthumously published A Market Theory of Money, Sir John Hicks (1989, 42-3) similarly reduces (what he argues) are the principal functions of money down to just two, from the usual three or four. These are, namely, (1) the unit of account (standard of value in Hicks’s own terminology), and (2) the means of payment of debt (‘medium’ of payment in Hicks’s own terminology).² These perspectives are also those I have adopted in my own work. In the most general terms, I would say that this sort of attitude is ultimately based on a ‘Weberian’ rather than a Marxian approach to economic sociology. Philosophically speaking it is based on metaphysical realism rather than ‘critical’ realism.

4. Marx and Money

Where did Augusto Graziani stand on these issues? On the one hand, he was one of the co-founders of the modern theory of the monetary circuit, which I and others have adapted and used profitably for our own purposes. On the other, he was the author of an important paper on ‘The Marxist theory of money’ (Graziani 1997). To what extent were the modern and traditional approaches inter-related in his work?

On this topic, I would recommend consulting several key papers by Riccardo Bellofiore (e.g., Bellofiore, 2019, 2020, 2024). Bellofiore was also a student of Graziani, and as he himself remarks (Bellofiore 2019, 1) very much an ‘insider’ in the development of early Italian circuit theory.² According to Bellofiore the answer is clear, if subtle. This is that, throughout his work, Graziani always had the intention of remaining in contact with the underlying structure of Marx’s argument, particularly as regards the role of the labour theory of value. The following quote, from another article published in the next year (Bellofiore 2020, 63), is characteristic of the argument:

            … an attentive look at the formation of Augusto Graziani’s circuitist perspective confirms how relevant was the influence of Marx – though it was hidden from view most of the time.

Earlier, I briefly referred to linguistic problems in discussing the various terms in M – C – C’ – M’ across different languages. These do, in fact, play a key role in the way in which Bellofiore interprets Graziani. Bellofiore (2020, 19-21), following Graziani, has stressed that in several European languages, including Marx’s original German, there are usually twowords standing for ‘money’. For example, Geld and Munze in German, argent and monnaie in French, denaro and moneta in Italian. The first of each of these couplets refers to monetary valuation in the sense of ‘abstract wealth’ or ‘wealth in general’, and the second to something like ‘change’, ‘currency’, or ‘cash-in-hand’ (with reference to the so-called ‘medium of exchange’ function). The significance of this is that it is then possible to argue that argent (say) may well be continuously increasing (that financial accumulation, in some sense, is ongoing) even while the amount of monnaie in circulation remains the same at the end of the circuit as it was at the beginning. By paying attention to these distinctions, therefore, it is possible for the analyst to reconcile the idea of the circuit with that of the labour theory of value.

What is also striking about this, at least to this speaker of English, is that each of the terms for the supposedly ‘abstract’ calculation of wealth retains an etymology derived from physical commodities, gold (Geld) and silver (argent). The word denaro is itself derived from the Roman denarius, a ‘silver penny’, or 1/240 of a physical pound (libra) of silver. One would have thought, logically speaking, that the linguistic nuances would have been the other way round. In English, the word for abstract wealth (money itself) does not have any such connotation. The only (shadowy) physical reference remaining is the name of the unit of account in the currency of the United Kingdom (UK) – i.e., the pound. Does this, perhaps, have something to do with the fact that England, and then the UK, was after all the original home of what Geoffrey Ingham (2004, 107) calls ‘capitalist credit money’?

The substantive point that needs to be made here is that the alternative approach to the circuit, as sketched out in section 3 above, does not actually make use of any of these historical/linguistic distinctions. It is based rather on a realist social ontology. As already mentioned, the notion of a medium of exchange disappears altogether, and ‘money’ is understood to be simply an asset which combines the functions of a unit of account and means of payment (of debt). Hicks (1989, 103-4) puts the point in the following way:

            We are well on the way to a credit economy in which any money that does not bear interest has become no more than small change or petty cash. It is surely at least a tolerable simplification as others to which an economic theorist is accustomed to take it that this has already happened.

Recall, also, that these words were originally written well over 35 years ago, when Hicks was already an octogenarian (Hicks 1989, 24).⁴

An ontology of the social system that is under consideration here would include just four categories, (1) goods and services – including machines, buildings, raw materials, etc. (2) money – where this specifically refers to a subset of the deposits of ‘all deposit-taking financial institutions’⁵, (3) financial instruments – all of which are ultimately claims (of one sort or another) to money, and (4) human effort – labour power in Marxist terms. It will always be possible to keep track of the notional accumulation of wealth by assigning ‘money’ values to the various components of that wealth in terms of the unit of account. (This might be called ‘capital’, in some quarters.) However, the difficulty with this procedure is in making those valuations ‘effective’ (which was precisely Keynes’s point, of course, cf. the notion of ‘effective demand’). For example, I can claim that my house is worth $2,000,000, but this is not ultimately going to be effective unless I can persuade somebody to give me $2,000,000 for it.

5. (Wages), Interest and Profit

In the recent paper already cited, Bellofiore (2020b, 57) has succeeded in bringing these issues very much to a head when he asks two specific questions about the way in which the monetary circuit is supposed to work. These involve (1) ‘the monetary payment of interest by firms to banks’ and (2) the ‘monetary realisation of profit by firms’. It will be convenient to take these two questions in reverse order.

The basic starting point of circuit theory is that at the macroeconomic level firms become indebted to the banks to be able to pay the aggregate wage bill (Bellofiore 2020, 62, Graziani 2003, 68-69). If this is exactly so, and we also suppose that the system is starting ab novo, the total amount of money in existence at the start will just be equal to the nominal wage bill, WN, where W is the nominal wage rate and N is the amount of employment. But, if this is all the money that there is, when the firms come to sell the output there is no question of them being able to realize any monetary profit in the aggregate, let alone being able to repay principal plus interest to the banks. This is so even if the workers do no saving at all and are willing to spend all their earnings.⁶ If, however, money is defined as in the last paragraph of section 4, the solution to the problem seems more-or-less obvious. The money supply is endogenous, and for there to be more money creation there will have to be more lending/borrowing over and above the initial loans that financed the original wage bill. In the real world this could come from a variety different sources, including borrowing by other entrepreneurs to start off a new round of production (Keynes’s ‘animal spirits’), borrowing by consumers, borrowing by the government (deficit financing), and borrowing by the foreign sector. In the first instance, this is just a question of ‘kicking the can down the road’, or so to speak. The next problem will then be to find some way of paying off the loans made by the second mover, those of third mover, and so on. The theoretical problem in macroeconomics becomes one of specifying the conditions under which such a profit equilibrium may be achieved – one that might also persist over time. Numerical examples have presented in such sources as Smithin (2018, 74-80), and Smithin (2022, 49-53), in which it is made clear that this notion of profit is essentially monetary.

The point about Bellofiore’s treatment, on the other hand, is to show how this same issue can be otherwise resolved while remaining securely within a Marxian framework. Bellofiore explicitly argues that Graziani, for example in Graziani (2003, 98-100), was able to by-pass, or as Bellofiore puts it, ‘dissol(ve)’ the problem of the realization of monetary profit simply by sticking closely to Marx’s original approach. The distinction between Munze and Geld, explained above, is again relevant. The argument is that valorization does not require an increase in Munze itself. In effect, Marxian surplus value can be realized in real terms (Belliofiore 2020, 65). Such a solution would not be available looking at the problem through the lens of the alternative social ontology described above. However, from the point of view of the main question posed in this paper, it demonstrates Graziani’s underlying Marxian premises.

The question of interest payments from the firms to the banks can be addressed in a somewhat similar fashion. In the approach set out in section 3, the solution of the problem of the payment of interest to the banks is already solved – at more-or-less the same time as the realization of monetary profit. For the first mover to be able repay principal plus interest, and make a profit, there always must be a second round of credit creation (and a third, and a fourth). This can be taken as reinforcing Keynes’s point that interest is a purely monetary phenomenon.

In Graziani’s Marx-influenced approach, however, the solution is to suggest a payment of interest ‘in kind’ (Bellofiore 2020a, 58) – that is, in terms of surplus value – from firms to banks. In essence, the two factions within the capitalist class (entrepreneurs and rentiers) simply share the net product. This is a characteristically Marxian point of view, as opposed to thinking of a three-way ‘class struggle’ between the workers and each of two (separate) classes of rentiers and entrepreneurs. Bellofiore (2020, 52) admits that in the context of a purely theoretical exposition, at least, the Marxian solution does require a certain amount of ingenuity to make it work. Some sort of ‘technical stratagem’ needs to be devised. This would be seen as a drawback from the point of view of a strictly realist social ontology. Nonetheless, the argument succeeds in retaining contact with Marx’s original theoretical structure.⁷

Conclusion

I hope that this short paper has been successful, firstly in indicating the importance of Graziani’s work. Secondly, following Bellofiore, how Graziani in his work on the monetary circuit was indeed able to retain the connection with Marx’s original formulations, particularly as regards the labour theory of value. Thirdly, how that differs from the use made of the concept in approaches based on a different social ontology. That seems to me to be essentially a difference between a (clearly original) theory, supported by scholarship in the history of thought and careful textual exegesis, on the one hand, and another original monetary theory based on an alternative social ontology (an alternative philosophy) on the other. Ultimately, therefore, on different metaphysical and epistemological principles. This does not mean however, that there cannot always be fruitful and productive interaction.