The bi-annual Post Keynesian Conference at UMKC hosted a session on the Circuitist School, with an overview paper by Guiseppe Fontana, and papers by myself, Jan Kregel and Claude Gnos. There were over 70 people in the audience, about 20 of whom actively participated in the very vigorous discussion.
My presentation tried to explain, in 30 minutes, the analysis presented in much more detail in my paper on the dynamics of money creation. The paper challenges several accepted propositions in the Circuitist tradition, though in general it is intended to show that the Circuit analysis is sounder than Circuitists themselves realize. I expected it to be controversial, and I wasn't disappointed. I'll try to clarify some of the issues that arose in the discussion here. There's a link at the bottom of this page if you wish to comment; I'll publish comments on this page as they are received.
Firstly, some preliminary points. As Warren Mosler pointed out in the discussion that followed, some of the debate was due to my use of terms that had different meanings to some members of the audience.
Chief amongst these was the meaning of the word "money". Warren suggested that I should have used the term "bank deposits" to describe what I called "money", and that if I had done so, there would have had far less disagreement with my analysis.
I think Warren is right (though others might disagree). So I accept his suggestion that, for those to whom money is an entity whose existence necessarily involves, for example, the existence of a tax-levying state, please substitute "bank deposits" wherever you read "money" in my presentation—and indeed, in the paper—but only up to a point, as I explain below.
In this sense too, the concept of "destruction of money" is sensible—since I quite happily accept that, if one defines money as "bank deposits", then "repaying debt destroys bank deposits" is a proposition I in fact make in my presentation and paper.
I was also not trying to make any contribution to the debate as to whether money came into existence because of the State's requirement that money be paid in taxes ("Chartalism"), or otherwise. Instead, the objective was to see whether Graziani's verbal model of a pure credit economy could be put into mathematical form, and then see what insights could be garnered from such a model.
Having outlined what I wasn't trying to do, I'll provide a more detailed reply to some of the objections raised to my model.
Paul Davidson asserted that my model of a pure credit economy was
disproved
by instances of other monetary systems (and perhaps he was also challenging the
Chartalist perspective on money creation too). He gave an entertaining example
of how tokens issued by the American military during the Korean War affected the
Korean monetary system.
This was interpreting my paper as arguing that "All economies are pure credit economies", rather like saying that "All swans are white". Therefore, a single example of a non-credit economy was sufficient to disprove my case, just as the discovery of Australia's Black Swan falsifies the statement that "All swans are white".
If I had been trying to argue that "All economies are pure credit economies", then Paul's "Korean Black Swan" would indeed have shot me down. But what I was actually trying to do was argue that "White swans can fly": that is, that the verbal model of a pure credit economy that Graziani put forward in 1989 made sense, and could be expressed mathematically.
This is an interesting issue, because the verbal and simultaneous equation modeling done by Graziani and others (Bellofiore et al., Fontana, Rochon, etc.) implies that "White swans can't fly". The Circuitist tradition has concluded that capitalists can't make money profits, that they can get back the amount they borrowed (and not a penny more) only if workers spend all their wages (i.e., have a marginal propensity to consume of one), and that continued economic activity required continuous injections of new money (see Trond Andresen's paper on this issue).
Circuitists have also shied away from modeling interest rates because they can't work out how capitalists can even pay interest on their debts: by the accepted analysis, even if workers do spend all their wages, capitalists would make a loss if the rate of interest was greater than zero. My model disputed all these conventional Circuitist wisdoms: capitalists can make money profits, that capitalist profits, let alone revenues, can greatly exceed the amount they initially borrowed,
Randy Wray argued that my model was not "interesting" because it wasn't a model of a monetary economy with banks, but if anything, a model of loansharks. He also objected to the proposition in the model that capitalists repaid their debts.
The former point is addressed below. I replied to Randy's latter point that this objection could be met simply by changing one parameter in the model—by setting the repayment rate to zero (I get the feeling that he didn't believe me...).
The figures below shows what happens in the model when Randy's objection is taken into account by setting the repayment rate to zero. With no debt repayment, debt of course remains at the initial level indefinitely.
The sum of the three income accounts (Capitalist Credit, Worker Income and Banker Income) remains at the same level indefinitely as well: the sum of Deposits always equals the sum of bank deposits. In this sense, "money" is equivalent to "bank deposits", as noted above.
One other notable change in behavior is that, with debt never repaid, the level of bank deposits remains constant and therefore the system continues indefinitely. With debt repayment but no relending, bank deposits tapered to zero over time and economic activity ceased (continuous economic activity required the relending of repaid principal).
Otherwise, the system behaves as I outlined: with production, capitalists can make a monetary profit if the flow of income from selling the surplus from production (modeled implicitly here) exceeds the interest payments on debt. With the parameters used in the paper, this leads to the capitalist account balance falling to 56.90, workers' account 41.06 and bankers income account 2.05.
The account that I called "Bankers Principal" in the paper and presentation, but which, for reasons outlined below, I am now renaming "Bankers Equity", remains at zero throughout with no debt repayment.
On
the incomes front, profit stabilizes at 12.07, wages at 39.83, and interest
income at 5.
To "put the Rumsfeld", does taking account of Randy's objection on this front improve the model? No. As I noted, debt repayment can be removed from the model simply by setting the repayment parameter to zero, so at that level it's trivial. However it robs the model of one important insight, which I must thank participants in the discussion for clarifying.
This is that endogenous money involves a reversal, not simply of the "deposits cause loans" perspective of exogenous money, but also of a much deeper belief: that "reserves cause loans". I'll elaborate on this after addressing Randy's first point—that this wasn't an "interesting" model because it was a model of a "loanshark" economy rather than a monetary one.
I presume Randy's charge was predicated on the Chartalist argument that money originated in a State's demand that taxes be paid by a State-sanctioned token rather than in kind with commodities. A monetary economy, on this view, must therefore be one with a tax-levying Government and a currency-issuing Central Bank.
I am rather agnostic on Chartalism: it may be historically correct that money originated that way, but an event which occurred in deep antiquity is less relevant today than the modern status and role of money. Human societies were slave-based when money originated; they were feudal when modern banking began. Surely the nature of money and banking have been altered from their ab initio states by the capitalist transformation of human society.
Arguing the point here might be fun over a beer, but I doubt that a conclusive resolution could be made. So I'll put the case another way: if the Chartalist perspective is correct, then it should not be possible to construct a functional model of a monetary economy without a Government, Central Bank, taxes, and State-issued money.
Prior to my model, the state of Circuitist modeling suggested strongly that Randy was correct. Papers such as Graziani's seminal 1989 paper, Bellofiore et al, Fontana,... all concluded that, in a pure credit economy, capitalists could only recoup their initial loan if (a) workers spent all their incomes and (b) banks charged a zero rate of interest. Otherwise, they would make a loss, necessitating further loans (at zero interest!) to continue operating. All authors reached conclusions similar to those of Bellofiore et al that a monetary profit could not be earned, that aggregate profit was zero, and so on. Clearly, these initial attempts to turn the Circuitist verbal insights into mathematical models generated dysfunctional systems that demonstrably were not models of actual capitalism.
This model is functional: capitalists can make borrow money and still make money profits, debt can be repaid (if capitalists so desire). It is at least a skeleton of the financial flows of an actual capitalist economy—though it deliberately is not a model of financial crises, etc.; that will come later when I add Minskian flesh to this Circuitist skeleton.
Studying skeletons alone can be useful, however. The model leads to a number of very deep insights about the nature of money, and the money creation process. These include that "loans cause reserves": a bank can start with zero equity, and generate reserves out of the loan repayment process. A credit money system can thus be even more endogenous than is currently accepted in the endogenous money camp. It will help the explanation if I first sketch out Graziani concepts that my paper modeled.
The Circuitists proceed from the simple but profound insight that a monetary economy cannot be one that uses a commodity for exchange, since such exchanges are qualitatively identical to barter (an economy that uses a designated commodity for exchange is just a variant of a barter economy). In barter, a sale of commodity X by A to B is simultaneously a sale of commodity Y by B to A. In a monetary exchange, a sale of commodity X by A to B is affected by a transfer of a "token" from B to A. Quoting Graziani:
The starting point of the theory of the circuit, is that a true monetary economy is inconsistent with the presence of a commodity money. A commodity money is by definition a kind of money that any producer can produce for himself. But an economy using as money a commodity coming out of a regular process of production, cannot be distinguished from a barter economy. A true monetary economy must therefore be using a token money, which is nowadays a paper currency. (Graziani 1989: 3)
The Chartalist position is apparently that monetary exchange is accepted because the monetary token is issued by and backed by the State, and its use is enforced by the State's insistence that taxes be paid using it.
The Circuitist position is simpler, and closer to the essence of endogenous money: the token is a "generic IOU" which is accepted as cancelling all "specific IOUs" between trading parties. This in turn requires two further conditions—quoting Graziani again:
b) money has to be accepted as a means of final settlement of the transaction (otherwise it would be credit and not money);
c) money must not grant privileges of seigniorage to any agent making a payment. (Graziani 1989: 3)
From these premises, Graziani concludes that:
The only way to satisfy those three conditions is to have payments made by means of promises of a third agent, the typical third agent being nowadays a bank... Once the payment is made, no debt and credit relationships are left between the two agents. But one of them is now a creditor of the bank, while the second is a debtor of the same bank. (Graziani 1989: 3; all emphases in original)
The Circuitist School thus argues that a ...
September 30th 8:50AM: To be continued...