By Eduardo Garzón, economist. Originally published on 22 March 2019 in Spanish here.
Translation by Carlos García Hernández
A few days ago, the Marxist economist Michael Roberts, taking the opportunity that in the United States there has been a lot of talk about Modern Monetary Theory lately, wrote an article critical of this approach, which was published here.
Specifically, he focused on the macroeconomic model used by the proponents of MMT, questioning at the root some of its underlying assumptions. I am writing this article to respond to that criticism: on the one hand I explain why I consider that one of the theoretical approaches of MMT questioned (the causal relationship established between profits and investment) is more realistic than the one used by the Marxist view on the same issue, and on the other hand, I correct what is undoubtedly a misinterpretation of MMT by the author (specifically, on how to finance the public deficit).
Roberts begins by quoting James Tobin to point out that accounting identities are very useful but dangerous in that they say nothing about the causality between their variables. That’s right: knowing with certainty that investment, for example, equals profits is not enough to identify whether the former causes the latter or whether it is the other way around. This is something widely recognized by the proponents of MMT, accustomed to using the accounting identities of Wynne Godley and Francis Cripps (1974), who always warned that causality was something that was not built into these identities. That is why MMT advocates developed a theoretical framework that makes causal sense of such identities. Randall Wray (2015) explains in his MMT handbook quite precisely why causality runs from expenditure to revenue, from investment to profits, and from deficit to surplus.
Let us recall that explanation starting from the accounting identity:
Spending = Income
Every sale necessarily involves two parties: the seller and the buyer. The buyer spends and the seller earns. The buyer’s spending is always necessarily equal to the seller’s income; it cannot be otherwise. If I buy an apple from the greengrocer for 1 euro, I will be spending 1 euro and the greengrocer will be earning 1 euro. What one party spends, the other party earns, because the money does not disappear and its amount is not altered in the transaction. There is no other possibility. This explanation of the accounting identity has no comment on causality. We will turn to that now.
Even if two parties are needed to carry out a transaction, one of the two must initiate it and allow the buying and selling to take place. And that party is always the spender, and not the earner. The party that earns cannot make money on its own from a sale because it needs someone else to initiate the process. On the other hand, the spending party can decide on its own whether or not to spend money on the purchase, because even if it does not have enough money, it can go into debt (or create money, which is a type of debt) and then buy the product.
The indebtedness, although strictly speaking it occurs with respect to another economic agent who must accept that commitment, will normally be possible because it will mean future income for the creditor (a business) and only in exceptional cases (when the debtor does not have the credibility that he will be able to keep his word) will the indebtedness not take place. Therefore, if the buyer does not want to spend, he will not spend; and if he wants to spend, he will spend (except in the aforementioned exception). In other words, the buyer can normally choose on his own whether he will spend or not. However, the same is not true for the seller: if the seller does not want to earn money, he will not do so; but if he wants to earn money, he will need a buyer who wants to spend. The seller cannot decide on his own whether he is going to earn money or not. In other words, the one who spends is the one who holds the key to the sale and purchase.
This reasoning can be extrapolated to the case of profits and investment, and to the case of balances. If we suppose two economic agents, the surplus of one of them is equal to the deficit of the other. No one can save if there is no one to “dissave” on the other side. On the other hand, in order to “dissave” it is not necessary to have someone wanting to save, it is enough – in the worst case – to get into debt or create money, which is always possible under normal circumstances. Consequently, the one who “dissaves” is the one who holds the key to the balance. Therefore, when we apply the identity of balances to the public and private sectors, we discover that the private sector cannot “dissave” indefinitely because it uses a currency it does not create (and if it did use the one it creates, it would have no way of imposing its use); on the other hand, the public sector can do so because it issues the currency it uses and also imposes its use by force. This is what MMT tells us: deficits produce surpluses, but – as Hyman Minsky (1986) showed – the former cannot be eternal in the case of the private sector; however, they can be eternal in the case of the public sector, which not only issues the currency that is generally used, but also has the authority to make it be used (mainly through taxes). Hence, the “engine” of economic activity is the public deficit, not the private deficit.
Why does Michael Roberts understand causality in reverse? Because he uses a commodity-money concept (like most Marxists) rather than a debt-money approach. He believes that money cannot appear before production, but somehow represents the quantity of goods and services that have been created (namely, by labour). According to this view, no one could be using money unless it had originated beforehand thanks to a productive process. Consequently, the key to economic activity would lie in supply, not demand. In this paragraph he explains it clearly:
“Contrary to the Keynesian/post-Keynesian/MMT view, the Marxist view is that “effective demand” (including government deficits) cannot precede production. There is always demand in society for human needs. But it can only be satisfied when human beings do work to produce things and services out of nature. Production precedes demand in that sense and labour time determines the value of that production. Profits are created by the exploitation of labour and then those profits are either invested or consumed by capitalists. Thus, demand is only ‘effective’ because of the income that has been created, not vice versa.”
But it is enough to conceive of money as a debt and not as a commodity to realize that the above is not true. Money is a commitment to future payment and, as such, can be created ‘out of thin air’; you don’t have to wait for someone to produce something for that money to originate. If you go into debt -and the counterparty trusts you to keep your word – you can purchase goods and services without having to have earned income beforehand. Producers make things when they see that they can sell them, and selling them does not require the buyer to have earned money in the past. Effective demand, therefore, is not born out of supply, but is born out of nothing, and it is supply that tries to adapt to it. Effective demand is not limited by any amount of money, because money can be unlimited (because it is a debt, not a commodity).
Roberts writes at the end of his article:
“The more important question, however, is what drives a capitalist economy. It is the profitability of capitalist investment that drives growth and employment, not the size of a government deficit.”
I could subscribe to that phrase. Capitalist accumulation requires private investment; if capitalists do not invest it is impossible to talk about capitalist economic growth. We could talk about another type of growth, but certainly not a capitalist one. But all this cannot be achieved without public deficits, which is the only sustainable source of money in an economy. If people did not have in their hands the money created by the State (through public deficits), with what would they buy the products sold by capitalists? Someone could answer this by saying that they could do it with private money, bank money for example.
But just a little above I have exposed the explanation of why the deficit of any private economic agent is not sustainable: it does not have the authority to force people to use the money it issues. To create money is to go into debt, and the private sector cannot go into unlimited debt because at some point people would realize that someone is promising a future payment that never comes, so at some point they would stop using that money out of fear and distrust. This finding is none other than the credit theory of Mitchell Innes (1913). On the other hand, the public sector can borrow unlimitedly because it imposes by force (with taxes) the use of its money. No private economic agent has the power that the public sector has to force the use of its money. This is the chartalist theory of Georg Friedrich Knapp (1924).
Finally, a brief note on what is a misinterpretation by Roberts. In his article one can read the following:
“Borrowing could be done by issuing government bonds (orthodox Keynesian) or by ‘printing money’ i.e., increasing cash reserves in banks (MMT). Issuing bonds may reduce Private Investment to boost Government investment, but the credit created would stimulate overall Investment. Printing Money (MMT) would raise Investment without reducing Private Investment (magic!).”
Leaving aside the lack of rigour and imprecision (because I hope it is not malicious intent) in using the word “printing” to talk about money creation when even central bankers themselves acknowledge that money is overwhelmingly created through computer keystrokes, it is completely false that MMT makes any kind of distinction between financing the government deficit with bonds or with new money, much less distinguishing the effects of such forms of financing on private investment.
MMT’s approach in this respect is, by the way, identical to that of those ascribed to the monetary circuit view, with Augusto Graziani (1990) at the head, who already quite some time ago put on the table an absolutely different understanding of the financing of the public deficit (among other things). To begin with, the public deficit is not financed; indeed, it cannot be financed. This entry is nothing more than the result between a) the amount of money that the State injects into the economy through spending, and b) the amount of money that it withdraws from the economy through taxes. In order to spend, it does not need to dispose of anything beforehand; it is enough for it to credit the bank accounts of the recipients of that spending. In order to receive, all it has to do is to debit the bank accounts of the taxpayers. Since what the State debits is money that it injected before through spending, the public deficit is nothing more than the amount of money that the State does not withdraw from the economy through taxes, ergo it makes no sense whatsoever to speak of financing spending. How can you finance something that is accounted for after the expenditure has been made?
So, what’s the point of issuing government bonds? Well, somehow it is necessary to prevent interest rates from sinking. Each new euro of public deficit is a new euro of bank reserves. Banks try to place them in financial investments to give them profitability, but when they have a lot of them, they have no choice but to agree to lower interest rates (when there are many sellers the only way to sell is to lower the price). It is by the simple law of supply and demand that the public deficit tends to push interest rates down. To prevent that from happening, states and their central banks can either offer more return on bank reserves (this way, banks will not try to get rid of them at a low interest rate) or they can offer banks a financial asset in which to place bank reserves at a decent return. That financial asset is public debt. States with monetary sovereignty do not need to sell public bonds to spend; they do so to remove bank reserves from circulation and thus ensure that banks do not try to place them at all costs, with the collapse in the interest rate that this would entail. Public debt is a tool to control interest rates, not to finance spending.
That is why the impact on investment – or on any other economic variable – is independent of whether the State issues public debt bonds or raises the profitability of bank reserves. Incidentally, if the state and its central bank did not make these moves, interest rates would fall, making private investment cheaper. Therefore, public deficits in this case would not harm private investment, but would, if anything, facilitate it. Roberts seems to remain anchored in the orthodox and erroneous conception of the impact of public deficits on private investment. The only case in which the public deficit could harm private investment – and this is independent of how the public deficit is “financed” – would be when the public sector has bought so much on the market that there is nothing left for private entrepreneurs. This, which would be possible, could even be positive if public investment were made according to social and ecological criteria, since it would be displacing a private investment that is always subject to private profitability and only tangentially and occasionally to the aforementioned criteria.
References
Godley, W. & Cripps, F.T. (1974). “Demand, inflation and economic policy”. London and Cambridge Economic Bulletin 84(1): 22–23.
Graziani, A. (1990) “The Theory of the Monetary Circuit’”, Economies et Sociétés 24, 7:7–36.
Innes, A. Mitchell. ([1913] 2004). “What is money.” Credit and State Theories of Money, edited by L. Randall Wray 14-49. Cheltenham: Edward Elgar.
Knapp, Georg F. (1924). The State Theory of Money. London: MacMillan & Company Limited.
Minsky, H. P. (1986). Stabilizing An Unstable Economy, New Haven, Yale University Press
Wray, L. Randall (2015). Modern Money Theory: A Primer on Macroeconomics for Sovereign Monetary Systems. Palgrave Macmillan.