Guest post: Part 2 – Keynes: Evidence for Monetary Policy Ineffectiveness (continued)
By Clark Johnson
(See the previous post in this series here)
a) The 1937-38 Contraction in the US
A few years later, Keynes disregarded evidence of the role of monetary policy in triggering a sharp relapse into near-depression conditions in the US during 1937-1938. The dollar depreciation of 1933 and the formal increase of the gold price to $35/ ounce in 1934 meant automatic revaluation of central bank gold stocks and gave impetus to increased gold exploration and production – concentrated, as it happened, in the Soviet Union. (Keynes noted the irony that increased Soviet efficiency in mining of gold was bailing out world capitalism!) He also noted that new gold reserves were bringing increased effective demand to the world economy that might result in “abnormal profits.”  Keynes understood (at least some of the time) the role of growing liquidity in the economic recovery of the mid-1930s.
In what now appears as one of the worst mis-steps in its history, the Federal Reserve, responded to rising wholesale prices in 1936 by deliberately sterilizing new gold inflows. A money supply measure (M2) that increased by 12 percent annually during 1934 -1936, suddenly turned flat and even slightly negative from about January 1937 to July 1938. Real GDP fell by 11 percent during this period, and industrial production fell by 30 percent. Rather than sterilize gold, had the Fed intervened in financial markets to target a modest rate of increase in any of a number of variables – a price index, industrial production, either real or nominal GDP growth, even a money supply indicator – most of the 1937-1938 contraction could have been avoided. By August 1938, the sterilization policy was jettisoned, and economic recovery resumed.
In February 1938, Keynes offered advice in a private letter to President Roosevelt that mentioned little of this. He did acknowledge that addressing “credit and insolvency problems” was an essential step toward recovery, as this created a necessary “supply of credit” – while, one infers, demand for that credit would have to come from elsewhere. This comment reflected Keynes’ ongoing view that expected returns on investment – the schedule of marginal efficiencies of capital — was independent of monetary policy. He went on the recommend that the US could “maintain prosperity at a reasonable level” only through “large-scale recourse to … public works and other Investments aided by Government funds or guarantees.” 
Despite Keynes’ recommendations, the lesson of all four of the illustrations here is that increasing money balances – through open market purchases, or through new gold or foreign exchange reserves – does affect expected returns on investment in plant and equipment, in equities, and in real estate.
ARGUMENTS FOR FISCAL ACTIVISM
We could stop here, having assembled evidence of Keynes’ dubious conclusions about relative un-importance of monetary factors in specific pivotal events. Indeed, evidence from these cases points strongly in the opposite direction, toward the crucial role of such factors. But the prominence of Keynes’ fiscalist legacy requires that we go further. Evidence aside, what was Keynes’ argument? In fact, he had a sequence of arguments.
In 1929, Keynes offered a comparative argument in favor of fiscal stimulus, and against monetary stimulus, specific to economic circumstances in Britain at the time. Keynes anticipated some portion of an argument Robert Mundell was to make decades later regarding the “policy mix,” that is, the appropriate mix of monetary and fiscal policy to meet both domestic output and external exchange rate targets. Britain in 1929 was on the international gold standard, hence was constrained externally by the need to maintain gold reserves. The Bank of England could not simply create credit, because, Keynes reasoned, “such credit might find its way to foreign borrowers, with the result of a drain of gold out of the Bank.” Hence Keynes proposed fiscal stimulus to increase domestic demand and employment, alongside monetary constraint to maintain Britain’s reserve and exchange rate targets.
This well-grounded argument also offers possible insight into the 1890s, where demand for gold reserves among central banks generated monetary contraction. Keynes, as we saw, did not make that argument – but we can construct it post facto. While the best solution might have been some international agreement to increase demand by modifying the international gold standard, a purely national approach could have looked to a fiscalist demand boost. But Keynes soon abandoned this policy-mix argument.
a) Removing external constraint on Monetary Policy
The US had freedom of action in monetary policy in 1933 and 1934. By March 1933, the dollar had been floated against gold, hence removing the external policy constraint – and, in any event, the US had by then accumulated vast gold reserves. In Keynes’ comments in January 1934, he had moved beyond his 1929 analysis. His newer interest was to argue that fiscal activism was preferable to monetary expansion even if the latter was not constrained.
Keynes in the General Theory (Ch. 15, “Incentives to Liquidity,”) offered the argument that monetary policy was specifically unsuited to boost economic demand when interest rates approached zero percent. In conditions where interest rates could not be lowered further, he reasoned, a condition of “absolute liquidity preference” held, later dubbed a “liquidity trap.” He observed, “In this event, the monetary authority would have lost effective control over the rate of interest.” This argument is cited endlessly by later-day Keynesians in support of a fiscalist agenda. (For example, see the reference to Summers mentioned at the outset.)
But the argument establishes much less than Keynes needed for his fiscalist agenda. Near-zero interest rates did not prevail in any of the four situations discussed earlier – yet Keynes wanted fiscal activism in all of them. So his case against monetary activism went beyond situations of absolute liquidity preference.
As noted earlier, Keynes pointed to a collapse in the marginal efficiency of capital as the trigger for the “slump of 1930.” The General Theory does much more to advance the concept that investment volume is unstable. Much of Keynes’ vision for government intervention, including fiscal activism, follows from his discussion of the fickleness of financial markets (Ch. 12, “Long Term Expectations.”) Noting the instability of private sector investment volume, he advocated a larger role by the government in stabilizing investment demand, often through direct outlays.
Keynes’ argument often shifted from the instability of the investment function to concern that investment was and would remain chronically weak – hence the conclusion that high unemployment was not self-correcting, but could persist for years. In Ch. 17 on the “Essential Properties of Interest and Money,” Keynes noted situations where the:
…rate of interest declines more slowly, as output increases, than the marginal efficiencies of capital-assets measured in terms [of the same asset].
As formulated in one of several instances in Ch. 22 (“Notes on the Trade Cycle”):
A more typical, and often the predominant, explanation of the crisis is, not primarily a rise in the rate of interest, but a sudden collapse in the marginal efficiency of capital.
This pattern of falling marginal efficiencies of capital made Keynes increasingly skeptical of monetary remedies.
A counter-argument is that adding liquidity – through open market purchases, gold inflows, or variations on these – might directly boost demand, and hence boost the marginal efficiency of capital, by increasing cash balances. But Keynes usually argued, to the contrary, that monetary policy worked mainly through raising or lowering interest rates –this was certainly a premise of the “liquidity trap” argument in Ch. 15. Further on, he wrote that “the primary effect of a change in the quantity of money on the quantity of effective demand is through its effect on the rate of interest.” In the Treatise chapter on “Control of Investment,” where he calls for open market operations a outrance, the goal is to bring “the market rate of interest … down to the limiting point.” In 1937 articles on “finance,” where Keynes stressed the crucial role of monetary policy, he again emphasized the channel of lowering interest rates.
 Keynes, “The Supply of Gold,” Economic Journal, Sept 1936.
 That is, coupling purchases of gold with offsetting sales of other central bank assets to drain liquidity
 Doug Irwin, Gold Sterilization and the Recession of 1937-1938, NBER Working Paper No. 17595, Nov 2011.
 In Collected Works of JM Keynes, Vol 21, pp. 434-39.
 Keynes, “A Program of Expansion (General Election, May 1929),” in Essays on Persuasion (1931), p. 124f.
 For ex., Robert Mundell, The Dollar and the Policy Mix (1973)
Keynes, General Theory of Employment, Interest, and Money (1936), p. 236.
 General Theory, p. 315.
 Axel Leijonhufvud offers a variation on this argument with the comment that in in Ch. 37 of the Treatise “the assumption that entrepreneurs are right was dispensed with” – that is, entrepreneurs became, in Keynes’ judgment, excessively bearish. In “Keynes and the Effectiveness of Monetary Policy,” Information and Coordination (1981). Leijonhufvud argues that Keynes’ subsequent arguments therefore relied more on fiscal intervention.
 General Theory, p. 298.
 For ex.,Keynes, “The ‘Ex Ante’ Theory of the Rate of Interest,” Economic Journal 46 (1937)