Dangerous Waffle about “the” Liquidity Trap

From May 2009 to last month, the U.S. unemployment rate has been at 9 percent or above, the longest period of such high joblessness since the 1930s. Can anything be done? In a television debate on August 14 the Nobel economics laureate Paul Krugman complained that the Federal government had not been imaginative enough. He referred to the Second World War when, according to a widely shared view, the surge in military expenditure was the key influence on the decline in unemployment. Given that argument, Krugman appealed on television to “space aliens” to threaten life on earth and create an invasion scare. He suggested that, all being well, rearmament ahead of a War of the Worlds might eliminate high unemployment in 18 months.

Krugman’s belief in the reality of space aliens and interplanetary warfare is a new strand in his intellectual development. By contrast, one of his more familiar recent themes has been standard Keynesianism for decades. This is that increases in nondefense government spending and the budget deficit are the right answer to an unemployment problem. He was one of the main advocates, along with Joseph Stiglitz and Larry Summers, of the immense so-called “fiscal boost” engineered by President Obama at the start of the present administration. That boost, which meant an addition of almost $800 billion at an annual rate to the budget deficit, was thought to be a fool-proof way of checking the Great Recession.

Keynesian emphasis on the effectiveness of fiscal activism has been accompanied by repeated claims about the ineffectiveness of monetary policy. The weakness of monetary policy is said to be particularly obvious now that Federal funds rate has dropped to zero. Krugman has said that the “zero bound” on interest rates recalls the “liquidity trap” conceptualized by Keynes in his 1936 classic The General Theory of Employment, Interest and Money. Keynes qualified his discussion in The General Theory by admitting that he had never observed a liquidity trap in the real world. Krugman today sees himself not only as reinstating Keynes’s message in theory, but as doing so in a world where two major economies—the United States and Japan—have succumbed to the liquidity trap in practice.

Evidently, the debate about the relative potency of fiscal and monetary approaches to economic stimulus is rather more than a dry, technical, and esoteric sideshow to the contemporary political scene. It will be central to the clashes between the Democrats and Republicans in the 2012 presidential election. Although neither side pretends to have a macroeconomic silver bullet, the Republicans can readily pin the blame for the United States’ ballooning public debt on Obama and his advisors. In my recently published book Money in a Free Society, I argue that the Keynesians in general, and Krugman in particular, have misunderstood both Keynes and the liquidity trap. As a result, they have grotesquely underestimated the power of monetary policy.


My critique of Krugman turns, to a large extent, on a misuse of words. The trouble goes back to Keynes himself, since he admitted at one point in The General Theory that he had “slurred over” problems in definition. Perhaps the most important of his verbal imprecisions related to the notion of “the rate of interest.”

The phrase could refer to several phenomena. People sometimes talk about “the rate of interest” on common stocks, by which they understand the dividend pay-out expressed as a yield on the market capitalization, or even “the rate of interest” on real estate investments, when they are thinking about the rental return. As we shall see, neither of these usages is silly, even if they may come across in highfalutin academic seminars as a little provincial. Usually in economics the phrase has one of two meanings. The first is the short-term money market rate, which serves as the building block of a much larger interest rate structure, including the interest rates charged on bank loans and paid on bank deposits. In the United States this money market interest rate is set by the Federal Reserve in operations with the commercial banks and is known as the “Federal funds rate.”[1] The second is the yield on bonds. This is a much trickier and more ambiguous idea than it looks. Bonds can have less than 50 weeks or over 50 years to redemption, be issued by governments or companies, and have a fixed payment coupon or a coupon payment that varies with other interest rates, with the inflation rate, or even a commodity price. (For example, Judy Shelton proposed gold-denominated bonds at the recent 29th Cato Institute annual monetary conference.)

Given the diversity of the alternatives, it might seem unwise to see “the rate of interest” as synonymous with bond yields. But without doubt this is what Keynes did in The General Theory. The point is clear and definite in the book’s chapter 15, which in one passage explained that banks could “purchase bonds for cash by bidding the price of bonds up in the market by a modest amount; and the larger the quantity of cash which they seek to create by [such] purchasing, the greater must be the fall in the rate of interest.” Further, the bond yield that concerned Keynes above all was the yield on long-dated bonds, as he thought it was particularly relevant to major investment decisions. He criticized the Federal Reserve’s activities in 1933 and 1934 when its operations had been limited to the short end of the yield curve and had “but little reaction on the much more important long-term rates of interest.”

Indeed, the logical coherence of Keynes’s liquidity trap hypothesis depended on the equivalence of the rate of interest and the bond yield in his work. As an active speculator in financial markets, Keynes was well aware that errors in judgment could lead to the loss of capital. (He almost went bust on two occasions.) He knew that, because bond prices and yields moved inversely, somebody who bought bonds at a high price (that is, a low bond yield) might suffer a hefty capital loss if the next major price move were downwards.

Fears of a capital loss could therefore deter investors from buying bonds at even higher prices, even if official operations increased the quantity of money in their hands. An increase in the quantity of money would therefore not depress bond yields or, Keynes saw it, the “rate of interest.” Once bond prices were at an unusually high level compared to long-run norms, a money injection engineered by the state could not reduce the rate of interest. In this sense the scope to conduct expansionary monetary policy had been exhausted.

In short, Keynes’s liquidity trap concerned the long-term bond yield and arose from the uncertainties facing investors as they balanced money holdings against bond investments. But here is the rub. This trap is not at all the one that bothers Krugman, despite his insistence on both the Keynesian lineage of his thinking and the “classic” character of his interpretation. In Krugman’s work monetary policy has become ineffective because the short-term money market rate cannot be lowered beneath zero by central bank operations. Frankly, Krugman says nothing in his articles and papers about the relationship between bond yields and the quantity of money. But the long-term bond yield and the short-term money market rate are not at all the same thing. It follows that Keynes’s trap and Krugman’s trap are very different ideas, and need to be clearly distinguished. (In essay 4 of Money in a Free Society I show that Krugman has devised at least three extra liquidity-trap formulae, on top of the two I have already discussed! They are all interesting and thought-provoking in their way, but not one of them is Keynes’s trap in The General Theory.)


How do these issues—apparently so theoretical, abstract, and arcane—influence today’s policy debate in the United States? Can they in all seriousness be relevant to the next presidential election?

The answer is that Krugman has leapt from his misinterpretation of Keynes to two very dangerous assertions, while President Obama has unwisely translated these assertions into policy. The first assertion is that at a zero money market rate, Federal Reserve action cannot further lower “the rate of interest.” This claim has allowed Krugman to invoke the liquidity trap and to pooh-pooh monetary policy. The second assertion is that monetary policy has been made ineffective by “the” liquidity trap, which justifies the recommendation that government stimulus efforts must take the form of a yet larger budget deficit. Implicitly, there is one, and only one, notion of the liquidity trap.

But, as I have demonstrated above (and with more rigour in Money in a Free Society), economists have a proliferation of liquidity traps and have been careless in differentiating among them. Remember that Keynes’s trap was not about the responsiveness of the money market rate to central bank operations, but rather the effect on long bond yields of changes in the quantity of money. The United States would be in his liquidity trap today only if larger increases in the quantity of money had failed to drive down bond yields. Keynes did not doubt the ability of the state to create new money without limit. In that sense monetary policy was unlimited in his era, it remains unlimited today, and it will always be so. What then is happening to the quantity of money in the United States at present?

In Money in a Free Society I argue that the correct aggregate to use in macroeconomic analysis is one that is broadly defined to include virtually all bank deposits. This is consistent with Keynes’s approach in The General Theory, since his analysis relied on the nearest alternative to “money” being a non-monetary asset (that is, a bond). It also accords with the views of Milton Friedman and Anna Schwartz, who in their path-breaking A Monetary History of the United States blessed a broadly defined measure which included time deposits. (In my view the M3 measure is the ideal. The numbers for “the quantity of money” in the next paragraph are those estimated by the UK Shadow Government Statistics advisory service for M3. The Federal Reserve ceased publishing M3 data in 2006.)

In the three years to October 2008, the quantity of money soared from $10,032 billion to $14,186 billion, with a compound annual growth rate of just over 12 per cent. The money growth rate in this period was the highest since the early 1970s. Indeed, 1972 and 1973 had many similarities to 2006 and 2007, with bubbling asset markets, buoyant consumer spending and incipient inflationary pressures. On the other hand, in the three years from October 2008 the quantity of money was virtually unchanged. (It stood at $14,340 billion in October 2011.) In other words, in the three years of the Great Recession the quantity of money did not increase at all.

What would have happened to the long bond yield if the Federal government and the Federal Reserve had, in a programme of money creation far more aggressive than anything so far attempted, decided to pump up the quantity of money by 5, 10, or 20 per cent? We do not know. Because the quantity of money has been roughly constant in the Great Recession, we can only conjecture the sequel to a big burst of money expansion. We therefore cannot say anything about whether the United States is currently in a liquidity trap or not, as Keynes and not Krugman defined that concept.

May I go further? The salient feature of the numbers just given is the contrast between the boom in money growth in the three years to autumn 2008, when Lehman Brothers failed and the Great Recession began, and the stagnation of money subsequently. (I simplify. A drop in M3 in 2010 has been offset by renewed moderate growth in the last few quarters.) If one is looking for a convincing narrative and a cause-effect interpretation of the Great Recession, it seems to me that the quantity of money is the variable that should receive most attention. In this I am self-consciously trying to revive the intellectual legacy of Milton Friedman, who in a long and often turbulent intellectual career always insisted that “money matters.” Money has mattered in the Great Recession, just as it did in the Great Depression. (By common consent, the most searching analysis of the Great Depression was given by Friedman and Schwartz in chapter 7 of A Monetary History of the United States.)


The United States is not in a liquidity trap today and never has been in one. The facts and data tell their own tale. Active fiscal policy has been useless as an antidote to the Great Recession, while money and monetary policy remain extremely powerful. Krugman and the Keynesians have misled the president. The surge in American public debt has failed to bring down unemployment, but it has led to the United States forfeiting its AAA credit rating. Astonishingly, Krugman, Stiglitz, Summers and company seem to have learned nothing from the last three years of disappointment with fiscal policy; they recommend yet more increases in government spending and another enlargement of the budget deficit. Given their indifference to evidence, one should perhaps not be surprised that the latest thinking envisages heavy expenditure on rockets and missiles to be fired at Martians and unidentified flying objects.

May I close with a few remarks on Keynes’s liquidity trap, the one that he proposed in The General Theory, but the details of which Krugman and his associates have forgotten? The validity of Keynes’s trap depended—in my view—on two assumptions that Keynes mischievously elided in his presentation of the argument. The first was the economy could be represented by a model which contained money, bonds, output, and employment—by a model, in other words, in which the only nonmonetary assets were bonds. The second was that, in books three and four of The General Theory, which developed the key ideas about money, bonds, and the rate of interest, wages per worker (and hence the price level) were held constant.

Let us think a little harder about Keynes’s model. Yes, in the real world investors are balancing money against bonds, but bonds are only part of their portfolios. Far more important in their decisions are the balancing of money against equities and real estate. In the United States today, as in most capitalist societies, bonds are only a fraction of the typical investment fund. And how plausible is the claim that never-ending expansion of the quantity of money would have no impact on the prices of equities and real estate? At highfalutin academic seminars economists might despise the possibility that unsophisticated investors compare “the rate of interest” available from bank deposits with “the rate of interest” on common stocks, equity-dominated mutual funds, real estate investment trusts, and so on. Well, the economists shouldn’t be so stuck-up. Bluntly, the viability of Keynes’s liquidity trap idea disintegrates when portfolio choice is not merely between money and bonds, but between money and a wide range of assets including equities and real estate. (This is one of the points in essay 2 of Money in a Free Society.)

And now let us also remove the assumption in books three and four of The General Theory that the price level is fixed. This assumption enabled Keynes to say that, normally, an increase in the quantity of money would lower “the rate of interest” and so stimulate investment. With the price level given, changes in the nominal and real interest rate would be identical. But nowadays alert investors form expectations about possible changes in the price level, and particularly smart investors watch trends in the quantity of money to anticipate big movements in inflation or deflation. Isn’t it possible that, if the Federal Reserve organized an enormous jump in the quantity of money (of, say, 50 per cent), investors would flee the bond market because of inflation worries? An increase in the quantity of money would then cause a rise, not a fall, in “the rate of interest.”

The troubles of the American economy today are not due to inherent weaknesses in capitalism, but to bad economic theories and poor economic advice. The Keynesians have had far too much influence since 2008. Their invoking of the liquidity trap notion in Keynes’s General Theory has been trebly unfortunate. It has been unfortunate, first, because they have misunderstood what Keynes himself said, secondly, because what Keynes said was rather foolish, and, finally, because the liquidity trap waffle has given an intellectual rationale for the disastrous budget deficits now being incurred. (And appeals to space aliens when the deficits don’t work does not help business confidence.)

American policymakers need to recall Milton Friedman’s key policy prescription over several decades, that a key condition of macroeconomic stability is stable growth of the quantity of money. The United States must never again experience the extremely volatile boom-bust in money growth of the last six years.

[1] Similar arrangements exist in other countries. In the European single currency area the bulk of the European Central Bank’s transactions with the banks are by means of “repurchase agreements” and so the main money market rate is “the repo rate.”

Also from this issue

Lead Essay

  • Tim Congdon argues that John Maynard Keynes’ latter-day followers have badly misinterpreted the theorist they profess to follow. Led by Paul Krugman, Keynesians have claimed that a near-zero Federal Funds rate is indicative of a liquidity trap. This diagnosis has several problems. First, it is not what Keynes meant by the term; second, even a rate of zero percent does not exhaust monetary policy; and third, a genuine Keynesian liquidity trap has not happened and cannot plausibly happen, in part but not solely because Keynes assumed constant prices throughout the economy, a condition that is unlikely in the face of a rising money supply. Congdon commends to readers Milton Friedman’s monetary prescription: a gradually and predictably rising supply of money, not the wild swings we have seen in recent years.

Response Essays

  • Dean Baker argues that Keynesians have not given up on monetary policy. Although the federal funds rate can’t go negative, the Federal Reserve can still set a higher inflation target, a solution both he and Paul Krugman endorse. Alongside monetary policy, Baker recommends fiscal policy: The recent economic stimulus legislation worked as intended, he argues, although the recession was more severe than the administration anticipated, and thus the stimulus proved to be too small. Policymakers have a duty to try to return the country to full employment, as the unemployed, who are suffering the most in the current crisis, are not to blame for their troubles.

  • Don Boudreaux agrees with Congdon that a monetarist policy approach would be preferable, but he draws our attention to a third relevant consideration: regime uncertainty, as described by the economist Robert Higgs. When businesses are uncertain about the major economic decisions of governments and central banks, they will defer new investments and retain cash rather than hiring new workers. Neither monetarism nor Keynesianism does anything to address the problem, which Keynes himself conceded was real.

  • Robert Hetzel reiterates that a zero lower bound for interest rates is a different phenomenon from a liquidity trap. The latter is an “irrelevant academic construct” as long as the central bank can create new money. Still, we learn little from this distinction unless we can determine the nature of the initial shock that caused pessimism among market participants; different types of shocks, monetary and real, call for different remedies. Central banks rarely use the analytical tools that would be necessary for them to evaluate their own roles in economically rigorous ways; instead, they tend to blame difficult times on the private sector, while taking credit for good ones.