Thursday, March 8, 2012

Epic Fail from William L. Anderson

Consider this statement made by William L. Anderson:
“The only think [sic] in the end that Keynesians have is inflation, and they believe that if the government inflates enough, somehow this will ‘rescue’ the economy.”
William L. Anderson, “Heading for double-digit inflation?,” Krugman-in-Wonderland, March 1, 2012.
I assume Anderson is using the word “inflation” in the sense of monetary inflation (though no doubt price inflation is envisaged as a consequence of the monetary inflation).

There are four obvious points to be made:
(1) Whatever the merits of partial or even total monetisation of a budget deficit, in relation to the type of Keynesianism practised by governments around the world today, if a government engages in a stimulus by way of a deficit, virtually all governments cover the deficit with dollar-for-dollar bond issues (some governments like Australia once had a tap system of direct purchases of Treasury bonds by the national central bank when the private sector did not wish to buy all bonds at some issue of government debt, but that system is long gone).

In the context of the US (though this also applies to other nations), when the government covers the deficit with dollar-for-dollar bond issues this process does not add to money supply as measured by M0, M1, M2 and M3 (now discontinued).

Some nations (such as Japan and Hungary) include government bonds in their highest monetary aggregate (M4 or L), but this is understood to be a measure of money plus liquid or reasonably liquid assets. The advocates of MMT say that a government deficit adds net financial assets to the private sector, and that is true. It is also obvious that, even if one were to include government bonds as money, the actual purchases made by people holding bonds must be so small as to have a marginal or insignificant contribution to price inflation. Who, for example, pays for ordinary goods and services in government bonds? Certainly the common items on the CPI are not bought with bonds.

Expansionary fiscal policy via deficits is the main policy proposal of Keynesian economics to stimulate aggregate demand. But, when government bonds are issued to cover budget deficits, the process does not expand the money supply, except by the highest monetary aggregate. And even then the newly created government bonds can only have marginal effects on price inflation, if they have any effect at all. The assertion “the only thing in the end that Keynesians have is [monetary] inflation” is rubbish.

Furthermore, the Austrians have their own idiosyncratic measures of the money supply, as follows:
(1) the True Money Supply (TMS) or TMS 1 (also called the Austrian Money Supply, or the Rothbard Money Supply, which can be found on

(2) Shostak’s Austrian Money Supply (AMS or TMS 2), and

(3) Mike Shedlock’s M Prime (M'), which is an alternative measure of (2).
True Money Supply (TMS 1) is made up of the following:
the currency component of M1 + total checkable deposits + savings deposits + US government demand deposits and note balances + demand deposits due to foreign commercial banks + demand deposits due to foreign official institutions.
Therefore the True Money Supply does not include government bonds. Nor do Shostak’s Austrian Money Supply (AMS) or Mike Shedlock’s M Prime. None of these Austrian measures of the money supply includes government bonds. On what basis do the Austrians argue that a bond-financed government deficit increases the money stock?

(2) Central banks expand reserves by open market operations or the more radical version of this called quantitative easing. That process is distinct from expansionary fiscal policy done by budget deficits. Yet Keynesian economists have often emphasised the feeble and comparatively futile effect of monetary base expansion in situations of depression, debt deflationary environments, or poor business expectations (“pushing on a string” was the old neoclassical synthesis Keynesian phrase). This monetary “stimulus” is unlikely to have a significant effect on aggregate demand in such circumstances.

Witness the actual facts about the base money created in QE1: it mostly stayed at the Fed. Aggregate base money went from $800 billion in 2008 to about 2 trillion in early 2010. The $1.2 trillion created in QE1 did not enter the economy, but has been held by the banks as excess reserves. QE2 did much the same thing, but added to the reserves of foreign banks and Eurodollar markets, adding to the US dollar carry trade. That US dollar carry trade no doubt contributes to asset price inflation and commodity speculation, but the inflationary effects of this on national CPIs are indirect, mostly via cost push inflation from factor input price increases.

(3) If a budget deficit were (1) covered by bonds or (2) partially or totally monetised (and in the latter case there would be monetary inflation), in a situation of depression or significant idle resources and unused capacity in an economy open to international trade, the primary effect of deficit spending is to increase output and employment – employing people, raising their income and increasing their standard of living. Price inflation is no doubt a secondary effect, just as it always would be even if the spending causing it under the same circumstances were private.

Keynesian stimulus policies confer benefits that outweigh the costs of secondary price inflation. The creation of greater employment and output drives real GNP to hit its potential. This makes the community richer than it would otherwise have been, if real GNP (that is, real output) had fallen below its potential, and provides a greater level of income owing to higher levels of employment.

But here the Austrians must switch to the meaning “price inflation” if they were to complain that “the only thing in the end that Keynesians have is inflation.” And even then they would be wrong.

(4) With regard to price inflation and movements in the price level, there are Austrians who argue that these have complex real and monetary causes:
“the essence of inflation is not a general rise in prices but an increase in the supply of money, which in turns sets in motion a general increase in the prices of goods and services .... While increases in money supply (i.e., inflation) are likely to be revealed in general price increases, this need not always be the case. Prices are determined by real and monetary factors. Consequently, it can occur that if the real factors are pulling things in an opposite direction to monetary factors, no visible change in prices might take place. In other words, while money growth is buoyant – i.e., inflation is high – prices might display low increases.”

Frank Shostak, “Defining Inflation,” Mises Daily, March 6, 2002.
That is essentially correct, and in environments of severe unemployment, idle resources and idle capital goods monetary inflation does not necessarily lead to sigificant price inflation.

Anderson, William L. 2012. “Heading for double-digit inflation?,” Krugman-in-Wonderland, March 1.

Shostak, F., 2002, “Defining Inflation,” Mises Daily, March 6

1 comment:

  1. You're shooting fish in a barrel again, I see! Not very sportsmanlike.