Money in a Free Society: Keynes, Friedman, and the New Crisis in Capitalism, Tim Congdon, Encounter, 486 pages
Tim Congdon’s new set of essays could not arrive at a better time. The global economic crisis has exposed serious flaws in the standard model used by macroeconomists. Often dubbed “New Keynesianism,” this model suggests that macroeconomic policy is all about manipulating the short-term interest rate, especially the overnight rate on interbank loans. As Tim Congdon puts it, “In the New Keynesian schema, it became in effect the only policy instrument, the factotum of macroeconomics.”
Yet by 2009 the most famous New Keynesian policy, the Taylor Rule, suggested that nominal interest rates should be negative. Since cash has a zero nominal interest rate, negative rates are virtually impossible. Thus the New Keynesian policy regime froze up at a most inopportune time. It was as if the steering mechanism in a car had locked just as it approached the guardrail along a cliff.
Tim Congdon has had a long and distinguished career within the British monetarist tradition, and he has now published a set of 18 essays that depict how that tradition has evolved over time. The early chapters feature Congdon battling against a British establishment that was even more hopelessly stuck in “old Keynesianism” than the elite in America. There’s a lot of interesting history here, such as the infamous letter signed by 364 British economists in 1981 complaining that the Thatcher budget cuts would push Britain deeper into recession—just as British GDP was about to turn higher. Many of the signers are now in positions of authority and would just as soon forget that episode.
The old Keynesian tradition that was so popular in Britain might be called “extreme Keynesianism,” as it takes some of Keynes’s most provocative assertions, particularly the ineffectiveness of monetary policy, and places them right in the core of the model. Congdon points out that Keynes’s actual writings were much more nuanced than is often appreciated—for instance Keynes clearly thought that monetary policy was highly important. However, one has to be careful when discussing “what Keynes really thought,” as he adopted all sorts of seemingly contradictory views at various points in his career. It does no good to cite quasi-monetarist statements made in the Tract on Monetary Reform, written right after the famous post-World War I hyperinflations, when Keynes’s fame rests on his work from the early 1930s, which is much more skeptical of the efficacy of monetary policy.
At one point Congdon compares Paul Krugman unfavorably to Keynes, noting that Krugman seemed to give up on monetary policy during the recent recession, when he called for massive fiscal stimulus. I see these two figures as being much more similar than Congdon does. Indeed, one could argue that it is Krugman who has the more nuanced view of monetary policy. Keynes did argue at various times during the 1930s that monetary stimulus would be ineffective at near-zero rates and that fiscal stimulus was needed. In contrast, Krugman’s famous liquidity-trap study shows that monetary injections expected to be permanent are always effective in boosting demand, as they raise the future expected price level and hence lower real interest rates—even for an economy stuck at zero nominal rates. He reformulated the liquidity trap as an “expectations trap,” and argued that the real problem in Japan was that the public correctly understood that monetary injections were not permanent and would be withdrawn at the first sign of inflation.
Krugman has a far more sophisticated understanding of expectations than Keynes. Not because he is smarter, but because he lived through different trend rates of inflation in the era after the end of the gold standard, something Keynes never experienced. But there are also lots of similarities. Both are very skilled polemicists. Both can be frustratingly hard to pin down on monetary policy. Most of Krugman’s readers assumed that he was arguing that monetary policy is ineffective at zero interest rates. But a close reading of his New York Times blog shows he was arguing that central banks were not likely to engage in the type of unconventional policies that might be effective. One could argue something similar about Keynes.
The biggest weakness in Congdon’s work is an insufficient focus on expectations. For instance, Congdon has a long discussion of the Japanese economy, which has experienced deflation since the early 1990s. Like other monetarists, Congdon points to the dramatic slowdown in the growth rate of the broad money supply during the 1990s and 2000s. Krugman would reply that the Japanese monetary base (cash and bank reserves) rose sharply during the past 20 years, and the broader aggregates (which include bank deposits) failed to grow because Japan was stuck in a liquidity trap and banks weren’t making loans. I happen to agree with Congdon that Japan was not “trapped,” but I don’t think he has sufficiently addressed the Keynesian counterargument. After all, the Fed recently injected huge amounts of base money into the U.S. economy, yet most of that infusion is hoarded by banks in the form of “excess reserves,” whereas the broader aggregates have shown little increase.
The real problem is expectations. The Bank of Japan pulled much of the new base money out of circulation in 2006 and raised interest rates out of misplaced fear of inflation, which never occurred. Monetary stimulus didn’t fail in Japan because of a liquidity trap, it failed because each time the rate of deflation approached zero, in 2000 and 2006, the Bank of Japan tightened monetary policy. The Bank of Japan acted like it was targeting inflation at no higher than zero percent, and by that criterion the policy was a huge success. They succeeded in preventing inflation for a period of nearly two decades. If they had set a 2 percent inflation target, and acted in a fashion consistent with that target, there’s no reason to think they would not have achieved it.
I don’t wish to leave the impression that Congdon has completely overlooked the problems raised by liquidity traps. He discusses ideas such as buying longer-term securities, whose yields have not fallen to zero. In my view this sort of “mechanical” monetarist approach might be sufficient, if pursued aggressively enough. And Congdon also has a sophisticated understanding of the political environment that central banks operate in, especially their reluctance to take extreme measures. This is why central banks need to focus on shaping expectations through explicit price level or nominal GDP targets. If the central bank doesn’t try to manage expectations, the level of bond purchases required to achieve their (implicit) objectives may be politically intolerable.
Congdon also has a lot to say about the relationship between monetary and fiscal policy and broader questions about the role of government: “As its critics understood, monetarism was not—and is not—politically neutral. It was and is an ally of a certain disposition towards political problems. This disposition was basically liberal, in the best nineteenth-century sense of valuing the freedom of the individual.”
I think that’s probably right, but let me offer two qualifiers. First, the Keynesian model is not necessarily left-wing. Government spending can be used to stimulate an economy in recession, but so can tax cuts. In addition, the Irving Fisher/Milton Friedman view of the Great Depression—that monetary stimulus was needed as a remedy—was actually quite progressive at a time when the gold standard reigned supreme.
But in a deeper sense I think Congdon is right. The most famous monetarist narrative is certainly Milton Friedman and Anna Schwartz’s Monetary History of the United States. Before their book was published, the standard view among intellectuals was that the Great Depression had exposed the folly of laissez-faire capitalism. By showing that the Depression was actually due to a failure of monetary policy, which allowed the price level to fall by 25 percent, Friedman and Schwartz made the world safe for free-market reforms and New Keynesian economics. By the 1990s, the standard model emphasized the importance of deregulation and privatization and assumed that the central bank, not the fiscal authority, would stabilize the business cycle. Friedman and Schwartz were not the only reason for the neoliberal revolution—the failure of communism played a role—but they certainly advanced free markets by radically changing the economics profession’s view of the 1930s, which had been seen as the greatest failure of capitalism.
The eighth essay in Money in a Free Society contains a fascinating look at how Milton Friedman’s views evolved over time. In 1948 he still accepted the standard Keynesian dogma that the budget deficit played a major role in determining aggregate demand and output. By 1996 he was arguing that fiscal stimulus was almost completely ineffectual, as monetary policy drives growth in nominal expenditures. Friedman suggested that this change reflected “empirical evidence” over the intervening half century.
Congdon discusses five reasons why fiscal stimulus might be ineffective, but I’d like to add a sixth. If central banks are targeting inflation, then any fiscal stimulus that would otherwise raise demand (and hence inflation) will be offset by monetary tightening. One might argue that this doesn’t actually mean fiscal stimulus is ineffective in a technical sense; rather its effects are offset by monetary policy. But that’s to lose sight of the fact that from the perspective of fiscal policymakers, the only question that matters is “how much more rapidly will GDP grow if we spend a lot of money on fiscal stimulus, compared to the alternative scenario where we refrain from stimulus?”
If the answer is zero, then in any meaningful political sense the fiscal multiplier is zero. It’s probably no coincidence that the estimates for the fiscal multiplier have been sharply scaled back in the era of inflation targeting.
Congdon is at his best in the final two essays, where he discusses the role of monetary policy in the current recession. He shows that when the crisis hit in 2008, central banks abandoned the standard monetary policy playbook and embarked on experimental efforts to bail out banks, reflecting the “credit view” of people like Ben Bernanke. This was supplemented by fiscal stimulus. But it rapidly became clear that these policies were not likely to work, and as a result asset prices (especially equities) plunged in anticipation of a deep slump. Only after the Fed and the Bank of England adopted monetary stimulus—“quantitative easing”—in March 2009 did the asset markets begin to recover. Many readers will wish to focus on these sections of the book and skim over some of the earlier material.
I think Congdon is exactly right about the role of monetary policy in the Great Recession, which puts us in a very small minority. Most people on the left saw monetary policy as an irrelevance and focused on fiscal stimulus. On the right many worried that the large monetary injections would lead to high inflation. Neither view is supported by the evidence. Fiscal stimulus has not proved effective. In contrast, monetary policy never runs out of ammunition, at least as long as the central bank has paper and ink. Many conservatives underestimated the need for central banks to meet the public’s increased demand for liquidity in an environment of falling asset prices and near-zero interest rates.
Perhaps the biggest flaw in Congdon’s approach, and in traditional monetarism in general, is the excessive focus on the monetary aggregates. In fairness, Congdon doesn’t see any simple mathematical relationship between money and inflation. And these aggregates do provide a rough idea of the stance of monetary policy. But they can be misleading indicators when conditions are very unstable—which is when we most need a reliable indicator. I’ve recently participated in the development of a new strand of monetarism, dubbed “market monetarism.” As the name suggests, we believe that policymakers need to adjust the money supply to the point where markets expect on-target growth in prices (or nominal GDP). This approach is less sensitive to sudden changes in the demand for money, when the monetary aggregates provide misleading signals.
Despite these reservations I recommend this book to anyone searching for alternatives to the standard model, which both Congdon and I believe let us down in the recent crisis.
Scott Sumner is professor of economics at Bentley University and blogs at TheMoneyIllusion.com.