Schwartz, Anna Jacobson
Milton Friedman and Anna Jacobson Schwartz, A Monetary History of the United States, 1867-1960. Princeton: Princeton University Press (for the National Bureau of Economic Research), 1963. xxiv + 860 pp.
Review Essay by Hugh Rockoff, Department of Economics, Rutgers University. firstname.lastname@example.org
On Monetarist Economics and the Economics of a Monetary History
A Monetary History of the United States, 1867-1960 by Milton Friedman and Anna J. Schwartz is surely one of the most important books in economic history, and indeed, in all of economics, written in the twentieth century. It has had a profound impact on the way economists think about monetary theory and policy. And it is still one of the most frequently cited books in economics. To some extent, it has suffered the fate of most classics: it is often cited, but seldom read. In the course of preparing this review essay, I have been repeatedly struck by the difference between what people think Friedman and Schwartz say, and what they actually say. Below I try to set out some of the reasons for the enormous impact of A Monetary History, and some of the reasons why there is such a large gap between (to subvert the title of Axel Leijonhufvud’s fine book on Keynes) “Monetarist Economics and the Economics of A Monetary History.”
The main point of A Monetary History is that “money matters: ” The quantity of money is an independent and controllable force that strongly influences the economy. This view, which is now accepted, at least in some measure, by most economists is very different from the view that prevailed when A Monetary History was published. At that time the professional consensus considered monetary policy ineffective. The job of central bankers was to keep interest rates as low as possible as long as unemployment was a problem. Following this policy would mean, however, only that investment might be bit higher than it would otherwise be and unemployment a bit lower. And although inflation might be countered with higher interest rates, the presumption was that monetary policy would have little impact. The rate of unemployment and the behavior of costs, particularly wage rates, largely determined the rate of inflation. Controlling labor unions was important for controlling inflation; monetary policy was at best a secondary consideration. The main tool for keeping the economy on an even keel was fiscal policy. It was a development in the real world, of course, the growing problem of inflation in the 1960s and 1970s, that was the main factor overturning the Keynesian orthodoxy. But A Monetary History was a powerful voice for restoring to monetary policy some of its former prestige. How did Friedman and Schwartz persuade the majority of the profession that money matters? The basic methodology of A Monetary History is to highlight “natural experiments,” occasions when the stock of money changed for reasons unrelated to the current state of the economy, so that we can then attribute the corresponding changes in the economy to changes in money.
Friedman and Schwartz offer an impressive array of case studies. To convey a sense of their approach, let me cite three of their most famous examples: (1) the contrast between 1879-1896 and 1896-1914 in terms of the behavior of the price level; (2) the contrast between World War I and World War II in terms of the behavior of the price level; and (3) the impact of restrictive actions taken by the Federal Reserve system in 1937.
(1) Prices (the NNP deflator) fell -0.93 percent per year between 1879, when the United States returned to the gold standard, and 1896, when the deflation came to an end, and then rose 2.08 percent per year between 1897 and 1914. The stock of money behaved in a similar way. Money per unit of output (money divided by real NNP) rose 2.99 percent per year from 1879 to 1896, and then rose 4.23 percent per year between 1897 and 1914. The acceleration in money growth was the result of the flow of new gold, much of it from the mines of South Africa. (High-powered money rose 3.49 percent per year between 1879 and 1896 and 4.83 percent per year between 1897 and 1914.) To be sure, the intense searches for new gold mines and new ways of refining gold ore that were rewarded when the mines of the Rand became productive and the cyanide process for refining it had been perfected, had been encouraged by rising real price of gold before 1896. But these events long preceded the post-1896 inflation. The correlation between rising money supplies and rising prices after 1896, Friedman and Schwartz argue, must be chance or must reflect a causal connection running from money to prices.
(2) Surprisingly, prices rose more in World War I than in World War II, and by about the same magnitude in World War I as in, to go outside the strict boundaries of A Monetary History, the Civil War. Yet measured in almost any conventional way (length of war, casualties, government deficits, etc.) World War I was a much smaller war for the United States than the Civil War or World War II. The monetary facts, however, are roughly in line with the inflation facts. From 1914 to 1920 money per unit of output rose 8.45 percent per year while the price level rose 10.84 percent per year. From 1939 to 1948 money per unit of output rose 7.90 percent per year while the price level rose 6.65 percent per year.
As these figures indicate, money cannot explain everything. The difference in inflation in the two wars exceeds the difference in the rate of growth of money per unit of output. Nevertheless, the striking fact is that the rate of inflation and the rate of growth of money per unit of output were broadly similar in the two wars. One would have expected, based on the degree of mobilization, far more money growth and inflation in World War II.
Part of the reason that the United States could “get away with” slower monetary growth in World War II was that the deposit-reserve ratio of the banking system was lower during World War II. The government, therefore, received a larger share of the revenues produced by increases in the stock of money. High-powered money, the main channel through which the government acquires seigniorage, rose 10.78 percent per year in World War II compared with 12.25 percent per year in World War I. Friedman and Schwartz conclude that the correlation between prices and money per unit of output suggests causation running from money to prices, rather than the common effect of some third factor, such as the intensity of the mobilization.
(3) One of the most famous and most hotly debated examples offered by Friedman and Schwartz is the 1937-1938 recession. In early 1937 the Federal Reserve doubled the required reserve ratios of the banking system with the purpose of immobilizing reserves and preventing future inflation. After some months, this action was followed by declines in the stock of money and real output. Money fell -0.37 percent between 1937 and 1938 while prices fell -0.50 percent, and real output fell -8.23 percent. High-powered money, responding to other forces, rose by 7.95 percent during the same year. Friedman and Schwartz conclude that the correlation between the decline in the stock of money and the decline in economic activity must have resulted from chance or from causation running from money to economic activity.
These case studies, I should note, arose in three different institutional regimes. In case (1) the United States was on the gold standard, and there was no central bank. In case (2) the Federal Reserve was constrained by the need to finance large wartime government deficits, and had to follow the Treasury’s lead. In case (3) the Federal Reserve was relatively independent, and could follow its own judgments about appropriate monetary policy. Drawing examples from different institutional environments strengthens the argument. In each case there is a rough correlation between monetary changes and changes in the economy, yet the factors determining the supply of money are very different. This suggests that the proposition “money matters,” represents a fundamental economic relationship, and is not the adventitious result of some particular set of institutional arrangements.
None of these “natural experiments” or the many others cited in A Monetary History, was conducted in a laboratory. Many variables were changing, and it is always possible, although not always easy, to construct an alternative explanation based on some other key factor. An extensive literature, for example, has grown up elaborating and contesting the Friedman-Schwartz interpretation of case (3), and attributing the 1937 downturn to other factors, such as fiscal policy. But for someone seeking to overturn A Monetary History, contesting one of these explanations is only the beginning. What gives weight to Friedman and Schwartz’s argument is the multiplicity of examples. So far, I would argue, none of Friedman and Schwartz’s critics has been able to forge an alternative explanation – whether based on fiscal policy, or labor union militancy, or technological change, or whatever – that fits all of the examples explored in A Monetary History. Indeed, to my way of thinking, the major advances since A Monetary History, have been the attempts by Brunner and Meltzer, Bernanke, and others to enrich the picture of how disturbances in the financial sector, and in particular the banking sector, affect the rest of the economy, rather than attempts to explain macroeconomic events from totally different perspectives.
Perhaps the greatest mystery is not that the Friedman-Schwartz methodology was persuasive, but rather that despite the enormous impact of A Monetary History, few economists use its methodology. Typically, when an economist attempts to persuade other economists, the first step is to feed the numbers through the computer and in the process strip away the historical circumstances that adhere to them.
Friedman and Schwartz’s interpretation of the Great Depression is both figuratively and literally at the heart of their book. The detailed discussion occupies about 30 percent of the total, and the episode is referred to by way of contrast in discussions of other episodes. Princeton University Press later issued this section as a separate volume, The Great Contraction.
Their point, as most college students of economics now know (or should know), is that the Great Depression could have been greatly ameliorated by better monetary policy. Today, only a few dyed-in-the-wool Keynesians reject any causal role for monetary policy, although many economic historians would place the major blame for the Depression on other factors, and relegate bad monetary policy to a secondary role. The Friedman-Schwartz interpretation of the Depression was crucial, moreover, to the revival of confidence in market-based economics. The Great Depression, and the way it was interpreted by Keynesian economists, convinced a generation of American intellectuals that only socialism (or near-socialism) could save the American economy from periodic economic meltdowns. If Great Depressions could be prevented through timely actions by the monetary authority (or by a monetary rule), as Friedman and Schwartz contended, then the case for market economies was measurably stronger.
It has been objected that Friedman and Schwartz don’t prove that monetary forces caused the Great Depression. They merely describe the Great Depression in great detail as if monetary forces were the causal factor. This objection is true, but not as decisive as it might seem at first glance. From the point of view of proving the importance of money, the Depression is merely another period, although a particularly revealing one, in which to search for natural experiments. It provides additional evidence, such as the case of the doubling of required reserve ratios in 1937 discussed above, and other episodes, but this one short period by itself cannot prove anything.
Friedman and Schwartz are doctors writing up the results of a detailed clinical examination of a patient who entered the hospital on the verge of death. Their observation that the patient was suffering from a bacterial infection is not by itself proof that the infection caused the patient’s illness. The fact that other patients with the same symptoms and the same infection have been seen at other hospitals in other places and at other times is what makes their argument persuasive. And it is the evidence taken as a whole that makes the prescription offered by Drs. Friedman and Schwartz, that the patient should have been given a strong dose of antibiotics (high-powered money), appear so sensible.
Perhaps the most misunderstood aspect of A Monetary History is the way that Friedman and Schwartz treat Nonmonetary factors. Their approach is to assume a “real” business cycle, which is then pushed a pulled by monetary factors. I use the term “real” with some trepidation. What Friedman and Schwartz have in mind is the sort of cycle described by Wesley C. Mitchell, Arthur Burns, and other scholars at the National Bureau of Economic Research in work that preceded A Monetary History, rather than what now goes by the name “real business cycle.” Yet there is a family resemblance worth stressing. Friedman and Schwartz, unfortunately for us, say little about the sources of this cycle, although at times they make some interesting observations about the tendency of good harvests in the United States to occur at the same time as bad harvests in Europe, and a few other factors. Nevertheless, it is clear that various supply-side shocks including technological shocks that now appear important to macroeconomists would fit easily into the Nonmonetary cycle that forms the backdrop for Friedman and Schwartz’s analysis.
The real cycles in which Friedman and Schwartz impound other factors are often forgotten when economic historians recount “monetarist” interpretations of historical episodes. I have heard economic historians claim that Friedman and Schwartz “say” that the recession of 1937 was caused by the doubling of reserve requirements in 1937. In fact, they write the following.
“Consideration of the effects of monetary policy [the increase in required reserve ratios] on the stock of money certainly strengthens the case for attributing an important role to monetary changes as a factor that significantly intensified the severity of the decline and also probably caused it to occur earlier than otherwise” (p. 544).
Similarly, I have heard economic historians claim that Friedman and Schwartz say that money caused the Great Depression, or that the stock market crash did not cause the Great Depression. In fact their statements on both points are more circumspect, and assume a Nonmonetary contraction of some magnitude. Of the stock-market crash Friedman and Schwartz write that “… its [the stock market crash’s] occurrence must have helped to deepen the contraction in economic activity. It changed the atmosphere within which businessmen and others were making their plans, and spread uncertainty where dazzling hopes of a new era had prevailed” (p. 306).
The crucial turning point in the Depression, according to Friedman and Schwartz, was late 1930 or early 1931, when they thought the contraction might have come to an end in the absence of the banking crises. But they acknowledge that even so, the contraction of the early 1930s “would have ranked as one of the more severe contractions on record” (p. 306).
In their counterfactual discussion of the effects of an open market purchase of $1 billion, they conclude that if undertaken between January 1930 and October 1930 the open market purchase would have “reduced the magnitude of any crisis that did occur and hence the magnitude of its aftereffects” (p.393). If undertaken between September 1931 and January 1932, the open market purchase would have produced a change in the monetary tide and as a result “the economic situation could hardly have deteriorated so rapidly and sharply as it did” (p. 399).
In discussing the banking panic of 1907, to give an earlier example, Friedman and Schwartz conclude that “There can be little doubt that the banking panic served to intensify and deepen the contraction: its occurrence coincides with a notable change in both the statistical indicators and the qualitative comment. If it had been completely avoided, the contraction would almost surely have been milder” (p. 163).
In short, Friedman and Schwartz tried to show that good monetary policy – best of all, as Friedman argued elsewhere, a monetary rule – would make the world a better place; they never promised a rose garden.
Although the central thesis is “money matters,” Friedman and Schwartz follow a large number of closely related threads. These range from the determinants of the greenback price of gold after the Civil War, to the relative effects of mild inflation and mild deflation on long-term economic growth, to the effects of deposit insurance on the stability of the banking system, and so on. Their discussions of these episodes are invariably intelligent, and often at variance with what was the conventional wisdom at the time they wrote. Not only do these discussions help us to understand these particular episodes; they also increase our confidence in their central thesis. They convince us that we are reading economic historians of outstanding ability who have explored every nook and cranny of American monetary history.
As most readers of A Monetary History recognize the book also succeeds in part because of how well it is written. Friedman and Schwartz employ a style that might be called high-NBER. It is written for the intelligent lay person. No special knowledge of statistics is required to read it, and no equations appear in the text, although there is an appendix on the determinants of the stock of money that uses equations. The quantity theory of money never appears in algebraic form. The sentences flow in magisterial fashion, and yet one is aware that the authors have thought about what they are discussing and are eager to make sure that the reader understands. In many ways their book, with its myriad of examples and its telling analogies, is the most similar, among all the classics of economics, to The Wealth of Nations. One can’t help but feel that the former lecturer on rhetoric would have approved of Friedman and Schwartz’s polished yet straightforward style.
For all these reasons, my choice for the most significant book in the field of economic history in the twentieth century is A Monetary History of the United States, 1867-1960 by Milton Friedman and Anna J. Schwartz.
There is a large and growing literature on A Monetary History. Here I will mention just a few sources that I have found particularly useful.
Bernanke, Ben S., Nonmonetary Effects of the Financial Crisis in Propagation of the Great Depression. American Economic Review. Vol. 73 (3): 257-76, June 1983.
Bordo, Michael D., editor, Money, History, and International Finance: Essays in Honor of Anna J. Schwartz. National Bureau of Economic Research Conference Report series. Chicago: University of Chicago Press, 1989. (This volume, a Festschrift for Anna J. Schwartz, contains a number of relevant essays, including one by Bordo that focuses explicitly on the contributions of A Monetary History.)
Brunner, Karl and Allan H. Meltzer, “Money and Credit in the Monetary Transmission Process.” American Economic Review. Vol. 78 (2): 446-51, May 1988.
Hammond, J. Daniel, Theory and Measurement: Causality Issues in Milton Friedman’s Monetary Economics. Cambridge: Cambridge University Press. 1996. (Hammond discusses all of the Friedman-Schwartz work on money focussing on methodological issues and the large volume of criticism their work generated).
Leijonhufvud, Axel, On Keynesian Economics and the Economics of Keynes: A Study in Monetary Theory. New York: Oxford University Press, 1968.
Lucas, Robert E, Jr., “Review of Milton Friedman and Anna J. Schwartz’s A Monetary History of the United States, 1867-1960.” Journal of Monetary Economics. Vol. 34 (1): 5-16, August 1994. (Lucas lays out what he considers the most important contributions of A Monetary History.)
Miron, Jeffrey A., “Empirical Methodology in Macroeconomics: Explaining the Success of Friedman and Schwartz’s A Monetary History of the United States, 1867-1960. Journal of Monetary Economics. Vol. 34 (1): 17-25, August 1994. (Miron explains why members of the younger generation of macroeconomists, even those not trained at Chicago, found A Monetary History so persuasive.)
Steindl, Frank G., Monetary Interpretations of the Great Depression. Ann Arbor: University of Michigan Press, 1995. (Steindl provides a useful overview, which compares and contrasts the Friedman-Schwartz interpretation of the Great Depression with the interpretations offered by other monetary historians.)
Temin, Peter, Did Monetary Forces Cause the Great Depression? New York: Norton, 1976 and Temin, Peter, Lessons from the Great Depression. Cambridge, MA: MIT Press, 1989. (Temin presents a detailed and extremely skeptical reading of the Friedman-Schwartz interpretation of the Great Depression.)
|Subject(s):||Macroeconomics and Fluctuations|
|Geographic Area(s):||North America|
|Time Period(s):||20th Century: Pre WWII|