I am extremely happy to announce that my blog this week will feature four guest posts by Clark Johnson. I have for some time tried to convince Clark to write something for my blog so I was very happy when Clark’s manuscript for his paper “Keynes: Evidence for Monetary Policy Ineffectiveness?” arrived in my in-box recently. Clark and I have decided to split up the paper in four parts which will be published in the coming four days.
Clark Johnson is not only a brilliant economic historian and author of the great book “Gold, France, and the Great Depression, 1919-1932”, but he is also a clever observer of the current monetary policy debate. Clark last year authored an insightful paper on the causes of the Great Recession, which in my view already is a Market Monetarist classic.
Keynes: Evidence for Monetary Policy Ineffectiveness? (Part 1)
by Clark Johnson
Lawrence Summers, who was President Obama’s chief economist during 2009-2010, and who by accounts continues to be an important advisor, recently called on the US and other government to increase borrowing at current very low interest rates. He observes that, based on inflation-protected bond rates, current Treasury borrowing costs for securities of five and ten year maturities are negative. He adds that interest rates elsewhere in the world – Germany, Japan, and Britain – are also extremely low. He then argues that governments should look on such rates as an opportunity to borrow cheaply and thereby improve their long-term fiscal positions.
Summers is presumably correct from a fiscal management perspective about benefits of borrowing when interest rates are low. But as a macroeconomic strategy for recovery, that is only the beginning. Whether we turn to John Taylor on the right or Paul Krugman on the left, the essential element for fiscal stimulus to succeed is to stabilize expectations: will the stimulus continue for long enough to drive expectations, so that market participants know the boost will ongoing and not soon withdrawn? Summers’ then shifts to monetary policy, where his case is weaker. He says there is no point in “quantitative easing”, the open-market mechanism the Federal Reserve uses to inject reserves, as interest rates are already rock-bottom – and monetary easing works, he explains, through the mechanism of lowering interest rates. Presumably, this is what he has been telling Obama since 2009.
Summers’ reasoning draws at least in part, on John Maynard Keynes’ discussion about “absolute liquidity preference” that occur when interest rates are very low, and demonstrates a key argument used in policy circles against more aggressive use of monetary policy. I believe Keynes was, and Summers is, mistaken. Literature on Keynes is abundant. To gain a different perspective, I want to look at various evidence Keynes adduced against monetary remedies. I will then return to arguments he used in the General Theory (1936) and elsewhere, hopefully with fresh perspectives.
KEYNES’ ILLUSTRATIONS FOR MONETARY POLICY INEFFECTIVENESS
A portion of Keynes’ reputation as an economist, and of his place in history, rest on his diagnoses of crisis situations and his proposed remedies. Well-known examples include his tract on the post-World War One Versailles Conference, The Economic Consequences of the Peace (1921), and subsequent writings on hyperinflations and then on British deflation during the 1920s. Another, less well-known, was his discussion of French monetary and political crises during 1925 and 1926, which I credited in my own work on the period. His two-volume Treatise on Money (1930) provided detailed and often shrewd observations on a wide range of economic policy matters.
In contrast, the General Theory, the heart of Keynes’ contribution to economic ideas, is light on historical or even contemporary illustration. So the reader seeks to fill in the gaps by turning to other writings. Consider four prominent cases as they reflect on Keynes’ view of roles of monetary and fiscal policy.
a) British Deflation in the 1890s
An unexpected embrace of fiscal activism comes in the Treatise discussion of the deflation of the early 1890s, where Keynes argued that the Bank of England’s gold reserves were abundant and credit was easy. But prices in Britain and the world nevertheless went into decline, which undermined profit and investment and reduced employment. He wrote:
I consider, therefore, that the history of this period [1890-1896] is a perfect example of a prolonged Commodity Deflation – developing and persisting in spite of a great increase in the total volume of Bank-Money. There has been no other case where one can trace so clearly the effects of a prolonged withdrawal of entrepreneurs from undertaking the production of new fixed capital on a scale commensurate with current savings.
Keynes then concluded (anticipating his arguments a few years later, including in the General Theory,) that monetary expansion does not always work, and that there might therefore be a role for public investment projects to boost demand.
Keynes’ discussion of the 1890s misses the point. Britain in the late nineteenth century was part of an open world economy, with easy movement of goods, people, and especially capital. Keynes neglected to mention that system-wide demand for gold rose much more than the supply from the 1870s through the 1890s as nearly two dozen countries adopted or re-adopted the gold standard, and hence needed to accumulate reserves. Indeed, demand drove the commodity-exchange value of gold to the highest level it was to reach in four centuries of record-keeping — the flip-side of commodity price deflation. The commodity price decline reduced profits and chilled investment demand; but commodity prices were determined in international markets, not in Britain.
While demand for gold was surging, the world’s monetary gold supply in the mid-1890s was at its lowest point it was ever to reach relative to its 1800-1920 trend line. As the mines in the South African Rand cranked up production in the 1890s, relative gold supply and commodity prices increased, nearly in tandem after 1896 – thus ending the Commodity Deflation, and initiating a gentle inflation. A growing money stock affected not just the supply of credit (as reflected in a declining interest rate), but also the demand for it. A result was nearly two decades of economic growth in all of the industrial powers, which was sadly interrupted by the First World War.
Monetary events were at the heart of both the origins of and recovery from the depression of the early 1890s. Keynes himself gave this backhand acknowledgement with his comment a few paragraphs later that, “the fall of prices [in the early 1890s] could only have been avoided by a much greater expansion of the volume of bank-money.” It is revealing that Keynes could discuss price trends during that period without mentioning the geographic expansion of the gold standard – easily the most important monetary development of the era.
b) The onset of the Great Depression
Moving to then contemporary events, Keynes’ discussion of the “slump of 1930,” also in the Treatise, builds on similar themes. Gustav Cassel and Ralph Hawtrey had argued a few years earlier that the undervaluation of gold following restoration of gold standards at prewar gold prices would force world-wide monetary contraction, especially as former belligerents Britain, France, Germany, and Italy restored their gold standards. Keynes, in contrast, told the Royal Commission on Indian Currency in 1926 that central banks would adjust their currency reserve cover ratios if their gold stocks became inadequate – which allowed him to dismiss the danger. Keynes underestimated what we might call the mystique of gold money.
Keynes listed factors driving interest rates higher during the 1920s: corporate borrowing for new industries; governments borrowing to pay reparations and war debts; central banks borrowing to add reserves; and speculators borrowing to buy shares of stock. He identified but was less able to explain the collapse internationally in anticipated returns in investment – what he would later call the marginal efficiency of capital — that occurred in the mid-1920s. As in considering the early 1890s, he did not connect the fall-off in real yields on new investment with systemic monetary constraint. Parallel to what happened in the 1890s, the middle and late 1920s saw a commodity deflation as key countries adopted or returned to gold standards. He thought monetary expansion worked through lowering interest rates, without directly affecting demand for goods and services. He wrote that the only ways to boost demand were by lowering interest rates, especially long rates, further – or by government fiscal activism. He did not understand that the world required a higher gold price to restore gold-to-currency reserve ratios, or perhaps needed a departure from gold money altogether.
c) The Roosevelt Recovery in 1933
Keynes’ comments in January 1934 on the monetary-fiscal mix in the US were baffling. In one of his initial acts after Roosevelt’s accession to power in March 1933, the dollar was allowed to depreciate against gold. This was a momentous event in monetary history – the underlying cause of the interwar deflation had been removed, and the gold standard was never restored with the same conviction. Keynes nevertheless wrote:
One half of [Roosevelt’s] programme has consisted in abandoning the gold standard, which was probably wise, and in taking various measures … to depreciate the gold value of the dollar… [But i]t is not easy to bring about business expansion merely by monetary manipulation. The other half of his programme, however, is infinitely more important and offers in my opinion much greater hopes. I mean the effort to cure unemployment by large-scale expenditure on public works and similar purposes.
This summary scarcely acknowledges the results of the real-time experiment in expansionary monetary policy undertaken in the US within the previous year. Depreciation succeeded at least to the extent any advocate could have hoped. Industrial production soared by 57 percent during the first four months of the Roosevelt Administration beginning in March 1933 – this was the actual increase, not an annualized rate — making up half of what had been lost since 1929. It was the fastest four-month rate of expansion in industrial production in the history of the US. Yet Keynes apparently considered this event to be “infinitely” less important than the boost to come from fiscal borrowing for public works programs.
Had the experiment continued for a few months more, pre-crash production levels might have been recovered. Unfortunately, the NIRA (National Industrial Recovery Act, announced in July 1933, brought micro-policy changes that had the effect of stopping the recovery in its tracks. The NRA (National Recovery Administration), set up under NIRA, then negotiated specific sets of codes with leaders of the nation’s major industries; the most important provisions were anti-deflationary floors below which no company would lower prices or wages, and agreements on maintaining employment and production. Within a short time, the NRA reached agreements with most major industries. In a phrase, the NIRA wanted to increase prices by restricting output rather than by increasing demand. Scott Sumner provides several rounds of evidence for the contractionary impact of NIRA policy in his soon-to-arrive book, The Midas Curse: Gold, Wages, and the Great Depression.
Lest this appear suspect as a predictable right-wing narrative of the New Deal, consider that Keynes himself pointed to the “fallacy” of the NRA approach: He noted that “rising prices caused by deliberately increasing prime costs or by restricting output have a vastly inferior value to rising prices which are the natural result of an increase in the nation’s purchasing power.” He added that it was “hard to detect any material aid to recovery in the National Industrial Recovery Act.” Within six months after the NRA went into effect, industrial production had dropped twenty-five percent, erasing nearly half of the gains recorded during Roosevelt’s more successful initial months in office.
So here we are. We saw an historically sharp recovery for four months during 1933, driven almost entirely by a decision to break the straightjacket imposed on monetary policy by the international gold standard. Keynes had previously been an able critic of the gold standard, for example in the Tract on Monetary Reform (1923) and then in several chapters in the Treatise. The 1933 recovery was then stalled by micro-policies of which he was explicitly critical. Yet Keynes seemed to dismiss this entire episode in his call a few months later for fiscal stimulus!
 Lawrence Summers, “Look beyond interest rates to get out of the gloom,” Financial Times, 3 June 2012
 H. Clark Johnson, Gold, France, and the Great Depression, 1919-1932 (Yale, 1997)
 John Maynard Keynes, Treatise on Money (1930), Ch. 30 (ii).
 Roy Jastram, The Golden Constant (1977)
 League of Nations chart, reproduced in Johnson, p. 52.
 Treatise, Ch. 37 (iv) “The Slump of 1930.”
 Keynes, “Roosevelt’s Economic Experiments,” The Listener, 17 January, 1934.
 Keynes, “Mr. Roosevelt’s Experiments,” London Times, 02 Jan 1934.