Wednesday, August 7, 2013

The Quantity Theory of Money is Wrong

The quantity theory of money is still at the heart of mainstream analysis of price movements and inflation.

In essence, there are two versions of theory:
(1) The Equation of Exchange: MV = PT,
M = quantity of money;
V = velocity of circulation;
P = general price level, and
T = total number of transactions.

(2) the Cambridge Cash Balance equation: M = kPY,
M = quantity of money;
k = demand to hold money/the amount of money held on hand;
P = general price level, and
Y = volume of all transactions in the value of national income.
Irving Fisher’s equation of exchange is actually not the basis of neoclassical monetary theory. Rather, the Cambridge cash balance equation is the more influential version of the theory (Flynn 1984). The Cambridge cash balance equation also replaces the velocity of circulation concept with the idea of the demand to hold money.

The worth of these equations and the quantity theory as a general theory of inflation are doubtful. (Of course, “inflation” must be understood in what follows as a general and sustained increase in prices as measured by a price index).

First, the contention that money stock increases induce direct and proportional changes in the price level is empirically questionable (De Grauwe and Polan 2005).

Secondly, there is the direction of causation. The quantity theory assumes the direction of causation runs from money supply increase to price rises.

At most, the quantity theory captures a basic truth that a sustained general increase in prices requires a growing money stock.

But, while a money supply increase is a precondition for this, it is also an intermediate factor, and not generally the cause of price inflation.

The fundamental causes of a general price inflation are still supply side factors (rises in wages or prices of factor input costs) or demand side ones (high demand causing price increases in flexprice markets).

The direction of causation in an endogenous money world is not, generally, from money supply increases to price increases, but from credit demand and price increases to money supply increases (King 2002: 166; Robinson 1970; Davidson and Weintraub 1973). The latter does provide an intermediate step whereby a larger money supply allows further price increases and sustained price inflation without causing macroeconomic problems induced by shortage of money and credit.

A rising broad money stock means that banks require more reserves for their clearing of debts and transactions conducted in bank credit money. Central banks provide those reserves.

Therefore the process runs:
credit money demand → broad money supply increase → base money increase. (Moore 2003: 118).
This should be quite clear because the money supply is endogenous: most of the money stock is “broad money” or bank money, and the major driver of the expansion of this type of money is credit expansion in the form of bank loans (the creation of ordinary demand deposits and saving accounts is also an important factor).

But what causes the demand for and changes in the level of bank loans?

For businesses, it is investment and often changes in factor input bills. Wage rises can be induced by wage bargaining and other institutional factors, and factor input prices generally rise because of administered pricing decisions or demand/supply dynamics in flexprice markets.

Conversely, when a severe price deflation occurs accompanied by a contraction in the money supply, the direction of causation between the two is highly complex. Falls in prices in both flexprice markets and even fixprice markets can be induced by severe demand collapses, administered price decisions, or falls in factor input prices. To the extent that credit growth is reduced by these factors, money supply growth from credit will also fall.

And an actual monetary contraction can also be the consequence of other factors such as a collapsing financial system, the contraction in broad money as credit money (or bank money) is destroyed as people scramble for the higher form of money (such as cash), and repayment of bank loans and debt further contracts broad money.

The issue is complicated by debates between Post Keynesian “horizontalists” and “structuralists” on the role of the interest rate and how credit supply is determined, but that need not concern me for the purposes of this post.

The idea that inflation is “always and everywhere a monetary phenomenon” is unacceptable because it assumes an exogenous money world and the wrong direction of causality.

This is why the solution to accelerating inflationary outbreaks in capitalist economies is:
(1) incomes policy, especially policies to stop excessive wage rises and wage–price spirals;
(2) price stabilisation of fundamental factor input prices through buffer stocks, and
(3) demand management.
Trying to control money supply growth rates, as in Friedmanite monetarism, is pointless, because (1) the direction of causation is backwards and (2) central banks do not have direct control over the rates of money supply growth anyway.

Davidson, Paul and Sidney Weintraub. 1973. “Money as Cause and Effect,” The Economic Journal 83.332: 1117–1132.

De Grauwe, P. and M. Polan. 2005. “Is Inflation Always and Everywhere a Monetary Phenomenon?,” Scandinavian Journal of Economics 107: 239–259.

Flynn, D. O. 1984. “Use and Misuse of the Quantity Theory of Money in Early Modern Historiography”, in E. van Cauwenberghe and F. Irsigler (eds.), Münzprägung, Geldumlauf und Wechselkurse: Akten des 8th International Economic History Congress, Section C7, Budapest 1982. Verlag Trierer Historische Forschungen, Trier. 383–417.

Ingham, Geoffrey K. 2004. The Nature of Money. Polity, Cambridge, UK and Malden, MA.

King, J. E. 2002. A History of Post Keynesian Economics since 1936. Edward Elgar Publishing, Cheltenham, UK and Northampton, MA.

Moore, B. 2003. “Endogenous Money,” in J. E. King, The Elgar Companion to Post Keynesian Economics. Edward Elgar, Cheltenham. 117–121.

Robinson, Joan. 1970. “Quantity Theories Old and New: Comment,” Journal of Money, Credit and Banking 2.4: 504–512.

Rogers, C. 1989. Money, Interest and Capital: A Study in the Foundations of Monetary Theory. Cambridge University Press, Cambridge.

Thirlwall, A. P. 1999. “Monetarism,” in P. A. O’Hara (ed.), Encyclopedia of Political Economy: L–Z (vol. 2). Routledge, London. 750–753

Smithin, J. 2012. “Inflation”, in J. E. King (ed.), The Elgar Companion to Post Keynesian Economics (2nd edn.). Edward Elgar, Cheltenham. 288–294.


  1. Nothing in your article persuades me that inflation is not "always and everywhere a monetary phenomenon”.

    You correctly identify that there is not a direct correlation between base money and the price level. Banks operating on fractional reserve principals will cause broad money to be many times greater than base money. If banks vary the proportion of reserves they hold this will cause broad money itself to be elastic. In addition the demand to hold money (liquidity preference) will mean that a given level of broad money may drive varying levels of spending and this will have a direct effect on the price level.

    So one can envisage a scenario where:
    - Workers negotiate higher wages
    - banks extend more credit to firms to meet these higher wages
    - inflation expectations pick up and cause people's demand to hold money to decreases
    - This cause them to spend more and this adds to the inflationary cycle.

    So it is possible to get inflation without an increase in the monetary base but soon (without an increase in the base) money demand will bottom out and banks will not be able to extend any more credit and inflation will end. Its only if the base is expanded (which may well happen for political reasons) that the cycle will continue.

    Further: A central bank that wished to control inflation could intervene at the start of the above cycle and REDUCE the monetary base sufficiently to prevent the credit expansion from occurring and adjust to the change in the demand for money.

    As long as a CB control the monetary base it has control over the price level albeit indirectly.

    BTW: Genuine question, can you give any examples of "incomes policy, especially policies to stop excessive wage rises and wage–price spirals" that actually worked ?

    1. BTW: Genuine question, can you give any examples of "incomes policy, especially policies to stop excessive wage rises and wage–price spirals" that actually worked ?

      (1) WWI and WWII.

      (2) Australia's "Prices and Incomes Accord" in the 1980s.

      (3) the so-called "polder" model as used in the Netherlands from the 1980s

      (4) even Nixon's wage and price controls worked when some serious attempt was made to enforce them.

      If you bothered to look hard enough you could find other examples from, e.g., Finland and other Continental European countries.

    2. There is discussion of incomes policy, with examples, in King, J. E. "Wages Policy," in G. C. Harcourt and Peter Kriesler (eds.), The Oxford Handbook of Post-Keynesian Economics, Volume 1: Theory and Origins, 2013.

  2. A post that fails on every way imaginable.

    First, no monetarist nowadays believes that there is a "direct and proportionate increase between increases in the money supply and the price level." Things like expectations, matter. An increase in the money stock perceived to be temporary, a doubling of the base in year 1, with the expectation that it will be withdrawn in year two, does little or nothing to increase the price level. On the other hand, a sustained, year on year increase in "vertical money" or the MB will easily increase velocity and pickup inflation.

    Number 2, velocity isn't stable, I know you want to flog a red herring and dead horse by attacking the "old" monetarism, but this tired attack doesn't get you anywhere. No market Monetarist believes that V is stable. thats why we advocate NGDPLT.

    Three, "horizontal" money can't increase without "vertical" money. (Even MMT'ers believe in a modified version of the QT with deficits instead of the monetary base) The idea that the central bank cannot control the money stock is laughable. OMP's are done every day LK. The central banks can use it like a sledghammer to stop inflation, buy selling their bonds and or gold reserves at ten cents on the dollar, or to increase it. by committing to buying a certain sum of bonds "at market"

    Four, INCOMES POLICIES? Are you serious??????

    Five, the seventies were a decade that central banks around the world gave excuses as to why inflation wasn't under their control. They all turned out to be false. P.V. stopped inflation in America at great cost. If you want to argue that it was unnecessary, fine. If you want to say that its costs were too severe, fine. But don't say that CB"S don't have the POWER. of course they do.

    1. "The idea that the central bank cannot control the money stock is laughable. OMP's are done every day LK"

      OMPs? Presumably you mean open market operations?

      And the fact that central banks conduct OMOs does not mean that they directly control the quantity or growth rate of the broad money stock. While the Fed can increase the base money when it wants to, that does not translate into control over broad money growth.

    2. "While the Fed can increase the base money when it wants to, that does not translate into control over broad money growth"

      Of course it does. It just makes adjustments to the base until it has the desired effect on broad money. It may be indirect but it can still achieve any outcome it chooses.


    4. "Clearly the reason pushing on string doesn't work is that we're not pushing on string hard enough. Why, no, I will never consider that I might just be misidentifying the causal relationship." - A Monetarist, without a trace of irony

    5. Reserves have no channel into the real economy. Increasing the monetary base cannot have any significant effect on what you call the money supply because bank lending is not constrained by reserves. Any loan officer can tell you this.

  3. The Fed should keep making adjustments until Novembers are sunny in Scotland.

  4. If the base causes nothing directly how can the mythical "expectations" change things? I fully expect the monetary base to keep increasing and I don't care in the slightest. 1000x zero is still zero. I expect some shamans will pray for rain. Does it change my expectations of rain if I expect them to pray more this month? How is this a serious argument?

  5. You should use some granger causality tests in order to move to a more serious level this debate. At this point what I see is that both explanations remain as hypothesis to be empirically tested.

    1. This has already been done by Moore (1988), Thomas Palley (1994), Howells and Hussein (1998) and Coporale and Howells (2001).

      The fact that you find some feedback from deposits to loans is still consistent with endogenous money theory: see also Philip Arestis, Malcolm C. Sawyer (eds.), A Handbook of Alternative Monetary Economics pp. 60-62.

  6. PeterP,

    If what you were saying is true, it would still be worthwhile for the central bank could buy up all the debt in the country several times over with no effect on prices, simply to eliminate the debt load The monetary base could be increased to infinity and the gain on those reserves and currency would be infinite.

    Obviously this scenario is ridiculous, and so is the argument that the base causes nothing

    1. It's interesting you write that, given Australia and Canada have no reserve requirements which (according to the money multiplier) means those banking systems do in fact have infinite quantities of reserves. And there's zero effect on price nor are their infinite "gains" as you call it.

    2. Um, you can only buy the debt once, not "many times over", because there is nothing to buy after you bought all of it. You monetize it all and nothing would change except the state wouldn't pay us interest on our holdings of its IOUs.

  7. Out of curiosity, Lord Keynes, have you read through Book V of The General Theory?


    Lord keyens i hope you will able to read how the government of israel stabilized the inflation from 450 percent annual to 6 percent annual by post keynsian moves its amazing how its true