Saturday, May 4, 2013

Rochon and Rossi on the History of Endogenous Money

Rochon and Rossi (2013) is a new discussion of the history of endogenous money in the latest Review of Keynesian Economics.

Economists are divided into two groups on the role of money in economic life, as follows:
(1) neoclassicals who think money is just a “neutral veil” over the real exchanges and activity in an economy. Money, in this view, is neutral in long run (monetarism), or neutral in both the short and long run (New Classical economics), and influences only nominal magnitudes (e.g., inflation), not real variables; and

(2) those heterodox economists, such as Post Keynesians, Monetary Circuit theorists, Sraffians, Old American Institutionalists, Marxists and Austrians, who think money is never neutral, either in the short and long run, and that monetary analysis captures fundamental truths about capitalist economies.
On the Post Keynesian view, it is not possible to study economics, without understanding money and its relation to output, production, investment and income (Rochon and Rossi 2013: 211).

The highest form of money with the power to finally extinguish all debts and taxes is fiat money created by the state, such as
(1) coins and notes, and

(2) base money created by central banks, which usually functions as private bank reserves.
Both these constitute high powered money, the monetary base, base money or in many countries what is called M0.

But most money in a modern capitalist economy is not fiat base money. Most money is actually credit money created by private banks and financial institutions, specifically the “bank money” of demand deposits and demand deposit-like accounts, such as checking accounts, transactions accounts and even some (so-called) savings accounts. The broad money stock mostly consists of credit money of this type.

Every time (1) a demand deposit or demand deposit-like account is opened or (2) credit is granted by a financial institution and a corresponding demand deposit account is created, new money is created. This money is destroyed as it is drawn down by the account-holder, or what we conventionally think of as “withdrawing” money, though that term is misleading, because such a bank account is not a bailment (that is, not, in legal terms, a depositum regulare).

Thus money is endogenous in the sense that it is mostly created and destroyed by the private banking system and, above all, created in response to the demand for it.

That is to say, money creation is normally credit-driven.

By contrast, mainstream neoclassical theory – apart from some New Keynesian and New Consensus macroeconomics which does recognise a limited form of endogenous money – sees money as exogenous and broad money creation as governed by the money multiplier. The latter views are wrong.

We see proof of this in the fact that bout after bout of Quantitative Easing (QE) have failed to restore full employment since 2008 (Rochon and Rossi 2013: 211).

But Post Keynesianism holds that the private banks are not constrained by their prior holdings of reserves, and that they will extend credit to all the clients that they deem to be creditworthy, although their credit standards may change over time, in that they might be lax during booms and more stringent during recessions (Rochon and Rossi 2013: 212).

Both broad money and, given that central banks have to accommodate the demand for reserves, even base money can be seen as endogenous.

But has money always been endogenous?

Post Keynesians have been divided on this question, and two views exist:
(1) The “evolutionary” view holds that money only became endogenous quite recently as the consequence of “financial innovations and/or an accommodating central bank” (Rochon and Rossi 2013: 212). Whether money was endogenous or exogenous depends on the historical period involved. This view is associated with Chick (1986).

(2) the “revolutionary” Post Keynesian view argues that money “has always been endogenous, irrespective of the historical period or of specific institutional arrangements” (Rochon and Rossi 2013: 212). Lavoie (1996: 533) advocates this view.
So who is right? Rochon and Rossi set out to answer that question and conclude that the “revolutionary” Post Keynesian view is right.

Rochon and Rossi argue that a society does not need a central bank for money supply to be endogenous to some extent (Rochon and Rossi 2013: 221).

Rochon and Rossi point out that throughout history we can find many examples of monetary systems in which credit money was created in response to demand for it, and where such credit money was elastic, and acted as a medium of exchange and means of payment (Rochon and Rossi 2013: 219). This kind of money is just a social relation.

This “debt money” or “credit money” represents the manner by which debts and debt/credit relationships between a borrower and a lender create an obligation that can be monetised and transferred, so that they then function as money.

For example, elastic credit money systems existed in ancient Egypt, Greece and Rome. As the financial revolution unfolded in late medieval Italian city states, bank money that was used as a genuine medium of exchange became a fundamental part of the money supply (Rochon and Rossi 2013: 219).

Before the 17th century, however, it was not exchangeable paper banknotes or instruments that constituted this money, but accounting entries or “book-entry money” (Rochon and Rossi 2013: 219), which was revolutionised by the entry of double-entry bookkeeping in the 13th century (Rochon and Rossi 2013: 218).

Even when physical goldsmiths notes or banknotes became important, these were just a paper representation of a bank deposit or debt owed, and a way to indicate how debts had been transferred (Rochon and Rossi 2013: 222).

Rochon and Rossi conclude that
“contrary to the argument provided by Chick (1986), money did not become endogenous over time. In fact, money has always been endogenous because of the necessarily triangular relationship involving a payer, a payee and a record keeper, even in those ancient times when money’s functions were carried out using a precious metal or, more generally, a given commodity; and this with or without the existence of ‘banks’ as such. In modern times, the banking system cannot but always respond to the needs of the economy to produce and exchange real goods and services – within as well as across borders. This is so even under a gold-standard system, as Joan Robinson (1956) noticed cogently” (Rochon and Rossi 2013: 225).
I would agree that money supply, even before the age of modern fiat money (that began in the 1930s as the gold standard was effectively abolished), was to an important degree endogenous.

The part of the money supply that was endogenous was the broad money stock. This was expanded by the creation of monetised debts and then the emergence of modern banks and their creation of demand deposits.

The evidence for how even the gold standard period had a system of endogenous money can be seen in my post here:
“The Classical Gold Standard Era was a Myth,” March 18, 2013.
There is also some discussion of the Rochon and Rossi paper over at the “Case For Concerted Action” blog.

Chick, Victoria. 1986. “The Evolution of the Banking System and the Theory of Saving, Investment and Interest,” Économies et Sociétés no. 3: 111–126.

Chick, Victoria. 1992. “The Evolution of the Banking System and the Theory of Saving, Investment and Interest,” in Philip Arestis and Sheila Dow (eds.), On Money, Method and Keynes: Selected Essays. Macmillan, Basingstoke. 193–205. [Reprint of Chick 1986.]

Lavoie, M. 1996. “Monetary Policy in an Economy with Endogenous Credit Money,” in G. Deleplace and E. J. Nell (eds.), Money in Motion: the Post-Keynesian and Circulation Approaches. Macmillan and St. Martin’s Press, Basingstoke and New York. 532–545.

Louis-Philippe Rochon and Sergio Rossi, “Endogenous Money: the Evolutionary Versus Revolutionary Views,” Review of Keynesian Economics 1.2 (2013): 210–229.


  1. What would you say is a good starting point for learning about Endogenous Money Theory?


  2. Agreed. However, I see certain dangers in using the argument that money is credit-driven in economic models. There is a strong tendency among some Post-Keynesians right now to have "credit" or "debt" as a sort of catch-all explanation. So, credit drives the economy -- and so on.

    This is, I think, just monetarism in a new guise. For monetarism output/prices are driven by the money supply -- which becomes a sort of mysterious entity with some sort of agency (a sort of anthropomorphism). If we start saying that credit is driving output/prices/asset prices we fall into the same trap. Especially because, roughly speaking, PKs think that credit IS money. So, in a sense we're just saying that money drives output/prices/asset prices, which is identical to Friedman's arguments (minus all the natural rate stuff).

    This is a danger recognised by, for example, Moore who pointed out that PK theory was in danger of falling into the monetarist trap by creating a monetarist-style "black box". In his paper "Unpacking the Post Keynesian Black Box"** he tries to remedy this by tying debt growth to wages. This is probably not a very good approach nowadays given that he was writing in the early 80s before the bulk of financialisation kicked in. Nevertheless we should remain diligent on this front.

    Credit and money are endogenous -- that is, they are passive. That does not mean that they are neutral. But it does mean that they cannot be used to explain anything. They are residuals.