Great Wave

by Fischer, David Hackett;

  Kuznets cycles or “long swings” of approximately twenty years have been much discussed by American economists, but this pattern has not been so interesting to European scholars or so visible in the history of their nations. See Simon Kuznets, Secular Movements in Production and Prices (Boston, 1930); “Long Swings in the Growth of Population and Related Economic Variables,” Proceedings of the American Philosophical Society 102 (1958) 25–52; Arthur F. Burns, Production Trends in the United States since 1970 (New York, 1934); Moses Abramowitz, “Resource and Output Trends in the United States since 1870,” American Economic Review 46 (1956) 5–23; Brinley Thomas, Migration and Economic Growth (Cambridge, 1954); John C. Soper, “Myth and Reality in Economic Time Series: The Long Swing Revisited,” Southern Economic Journal 41 (1975) 570–79. This rhythm is sometimes thought to be demographic in its origin, but Friedman and Schwartz argue in Monetary Trends in the United States and United Kingdom, 599–621, that long swings are episodic in their origin and monetary in their expression. Many economists agree with them.

  The Labrousse cycle (or intercycle) of roughly 10 or 12 years is much favored by European historians but rarely appears in American scholarship. It has been used in studies of French history.

  Juglar cycles or trade cycles (7 or 8 years) have been found by many scholars—by Goubert in Beauvais, Parenti in Tuscany, Spooner in Udine, Hauser in Paris. The classic work is Clément Juglar, Des crises commerciales et leur retour périodiques en France, en Angleterre, et aux Etats-Unis (1889) rpt. New York, 1967).

  Kitchin cycles or business cycles (3.5 years, or forty months) were first observed in the American economy during the nineteenth and twentieth centuries, and also in Europe during our own time. The classical text is Joseph Kitchin, “Cycles and Trends in Economic Factors,” Review of Economics and Statistics 5 (1923) 10–16. They are sometimes called “inventory cycles” and are thought to rise from the structure of modern business enterprise. But several historians have also reported them in price data as early as the fifteenth century, and Pierre Chaunu has discovered them in the rhythm of Séville’s transatlantic trade.

  For general discussions of business cycles, see Wesley C. Mitchell, Business Cycles (New York, 1927); Arthur F. Burns and Wesley C. Mitchell, Measuring Business Cycles (New York, 1946); Joseph A. Schumpeter, Business Cycles: A Theoretical, Historical and Statistical Analysis of the Capitalist Process (New York, 1939); Geoffrey H. Moore, The Cyclical Behavior of Prices (Washington, 1971). Historians will find a rapport with E. R. Dewey and E. F. Dakin, Cycles: The Science of Prediction (New York, 1950), which argues that these rhythms are themselves variable through time and space—a conclusion that is certainly correct.

  Cyclical patterns are often extracted from the data by “detrending” a time series—that is, by removing the secular trend so as to expose fluctuations more clearly. The great waves in this work are not extracted by filtering or detrending the data. They are the secular trends, and appear on the surface of the evidence. For problems of method, leading works are James D. Hamilton, Time Series Analysis (Princeton, 1994), and T. W. Anderson, The Statistical Analysis of Time Series (New York, 1971). Also helpful is Nathaniel J. Mass, Economic Cycles: An Analysis of Underlying Causes (Cambridge, Mass., 1975).

  APPENDIX F

  Toward a Discrimination of Inflations

  The many uses of the word “inflation” make an interesting study in scholarly semantics. The term has been defined in different ways. Some of the most common meanings incorporate a particular theory of inflation in such a way as to exclude all other theories. The result is a family of mutually contradictory theory-driven definitions. Each of them claims a universal validity. All are more unitary than the phenomenon that they purport to describe.

  An amusing example appears in Webster’s New World Dictionary. The second college edition of this work offers two contradictory theory-centered definitions on the same page. The term “inflation” itself is defined as “an increase in the amount of money in circulation, resulting in a relatively sharp and sudden fall in its value and rise in prices.” Just below it is “inflationary spiral,” which is defined as a “continuous and accelerating rise in the prices of goods and services, primarily due to the interaction of increases in wages and costs.”

  One of these definitions insists that inflation is exclusively a monetary phenomenon, caused by an expansion in the money supply. The other requires us to subscribe to a “cost-push” model. These theoretical definitions are narrow and specific. They are also mutually exclusive. If the “cost-push” model is correct, then inflation is not always caused by an increase in the amount of money in circulation.

  Further, both definitions also include specific historical descriptions of inflation. One of them demands that we think of inflation as “sharp and sudden.” Another insists that inflationary spirals are “continuous and accelerating.” These historical models of inflation are not only at odds with one another. They are also mistaken, both in general historical terms and in their specific theoretical linkages. Monetary inflations are not necessarily “short and sharp.” Wage-price inflations are not always “continuous and accelerating.”

  These usages often recur in learned discourse. It is very common for American economists to define the term “inflation” in exclusively monetary terms, and then to use it to describe an historical process which is not exclusively monetary in its cause.

  An example is an assertion by American economist Milton Friedman that inflation is “always and everywhere primarily a monetary phenomenon” (New York Times, February 19, 1984). Many of his colleagues agree with this statement. There is no necessary error in it. As long as it is confined within the constraining context of monetarist theory, Friedman’s statement is not merely true but tautological. Given certain theoretical assumptions, a rise in prices can always be translated into monetary terms. If the discussion were exclusively theoretical, there is no error here. The trouble comes when the term is defined in this way, and then used to describe the operative cause of an actual rise of prices in the real world—where price-increases sometimes have a monetary cause, but often rise from other roots.

  Outside of the learned professions, the word inflation is understood in other ways. In ordinary speech, it tends to be an omnibus term for any sort of increase in prices generally (which is not the same as an economist’s idea of the “general price level”).

  Professional usage in the learned disciplines seems to be shifting in this direction. Increasingly, historians and economists are growing more eclectic in their ideas of inflation. Two economists, Paul Samuelson and William Nordhaus, write, “Like illnesses, inflations occur for many reasons.” They divide inflations into three types, mainly by speed of advance: “moderate inflation” as in the industrial nations during the late twentieth century (1–10 percent), “galloping inflation” as in Latin America or Israel during the same period (10–1000 percent), and “hyperinflation” as in post-Wilhelmine Germany (1000 percent or more). This taxonomy brings to mind a mortality bill by an eighteenth-century New England physician who believed that all forms of disease shared a single etiology, and who classified deaths as “sudden” or “slow.” This is a primitive idea in medicine and history, but sometimes it has its uses.

  Another and better approach is to make a discrimination of price-inflations not by velocity but by cause. Historians tend to think of inflations in pluralistic terms, as rising from a broad variety of causal conditions. At least seven types of inflation might be distinguished by cause.

  One common variety of price inflation is caused by an expansion of the money supply. This is sometimes a slow creeping movement. It can also become a sudden surge of hyperinflation, of which the classic example is the German inflation of 1922–23. When the infant Weimar Republic was unable to meet its obligations by taxes or loans, it deliberately resorted to the printing press. The number of German marks in circulation rose from 5,807 trillion in January 1922 to 202 trillion-trillion in December 1923
, a number so large that it requires 30 digits: 202,232,341,000,000,000,000,000,000,000 marks. As a consequence, the wholesale price index in Germany rose from 100 in 1913 to 142 trillion in 1923. German burghers who suffered through this event told the story of a man who went to a grocery store with a wheelbarrow full of money to pay for his family’s food. A thief stopped him, threw away the money, and stole the wheelbarrow. What was still more dramatic about the German inflation was its sudden end. Monetary stability was restored in 1924 by the issue of a new currency that was very stable. The German hyperinflation of 1922–23 had many social and political consequences, but it did not become embedded in the structure of the economy, and disappeared when the inflated marks were withdrawn from circulation. There have been many monetary inflations of this sort, and other monetary inflations of a more gradual variety.

  A second type of inflation rises from increases in aggregate demand. One common example is war-inflation. Government spending for military purposes has often stimulated demand throughout an economy, at the same time that a shift of workers from productive labor into the armed forces causes a decline in aggregate supply. Other demand-inflations have risen from population growth, particularly when the general population increases more rapidly than the work force. In the twentieth century, demand-inflations have also been caused by rising expectations, and by higher standards of living.

  A third form of inflation is caused by contractions in supply—for example, by runs of bad weather which drive up agricultural prices. This happening was very common in medieval and early modern Europe, when a large proportion of family income was spent on grain and other farm products. The supply shock of reduced harvests reverberated through the entire economy.

  A fourth variety is cost-push inflation. It occurs when wages and prices begin to spiral upward, each driving the other in its turn. This mechanism was clearly operating in the middle stages of the price-revolution of the twentieth century.

  A fifth variety might be called the inflation of administered prices. It has happened in the United States as a result of collusive price-fixing in oligopolistic industries. Recent examples include the manipulation of oil prices by OPEC nations during the 1970s. Oil shocks had an impact on general price levels through the world economy.

  A sixth variety might be called bubble-inflation, caused by a surge of speculative activity, which when it rises rapidly and reaches broadly through an economy, distorts price levels in a general way. Examples might include the Dutch tulip mania in 1634 and the French Mississippi Bubble in 1717.

  A seventh variety might be called the “inflationary-expectations” model. It occurs when people begin to raise prices not because of actual changes in supply or demand or costs or the size of the money supply, but out of fear that some such change might happen.

  These different types of inflation often coexist. In actual practice, price-revolutions are complex phenomena that characteristically include many different types of inflation. Most have begun as demand-inflations, to which the effects of monetary-inflation, supply-inflation, and administered-inflation later added, and had the effect of reinforcing the momentum of the price revolution.

  It is interesting to observe that the effect of short-term inflations varies according to their timing within price-revolutions and price equilibria. For example, the inflation associated with the Civil War, the Crimean War and the Franco-Prussian Wars in the nineteenth century did not cause a permanent elevation of price levels. Prices surged during the wars, then rapidly declined in the peace. In the United States by the 1880s, prices had returned to levels of the late 1850s. During the major wars of the twentieth century things were a little different. Inflation surged after America joined the World War I in 1917, then declined after 1919, but not to prewar levels. After World War II, Korea, and Vietnam, war-inflations were not followed by a decline at all. Prices continued to climb. What was different here was the underlying dynamic of the price system.

  All of this suggests a need for price theory that incorporates a component of historical thinking, and also for historical models that include a generous measure of economic theory. Historical trends and contexts make a major difference. So also do the dynamic relationships that are modeled in economic theory.

  Economics is what Windelband called a nomothetic discipline. It seeks knowledge through generalization. History is an idiographic discipline. It studies things in their particulars. The two approaches are different, but also complementary. Together they can help us understand the many varieties of price-inflation, and also their common characteristics.

  APPENDIX G

  Money of Exchange and Money of Account

  A student of price history must confront a vast diversity of monetary units in the world—not merely in the variety of coins and paper currency, but also in the structure of monetary systems themselves. In the early modern era, these systems were in some ways more complex than those of our own time.

  One dimension of that complexity appeared in the difference between two types of monies: money of exchange and money of account. Alexander Justice wrote in 1707, “Money in general is divided into two sorts, imaginary and real.” (A General Treatise of Monies and Exchanges [London, 1707], 1; quoted in John J. McCusker, Money and Exchange in Europe and America, 1600–1775: A Handbook [Chapel Hill, 1978], 3).

  Justice’s “real money” is money of exchange. It is issued by virtually all sovereign states and consists of coins and paper that pass physically from hand to hand. Justice’s “imaginary money” is called money of account. It exists only as an idea, and is used in bookkeeping and credit transactions.

  The distinction between real and imaginary money seems unnatural and absurd to Americans today, who use the dollar as both money of exchange and as money of account. But practices were different in earlier periods. Real money and imaginary money existed side by side.

  A case in point was eighteenth-century England, and English-speaking North America. Money of exchange consisted primarily of two coins: the silver shilling and the golden guinea, which was worth twenty-one shillings. There was also a silver crown (worth five shillings), and various other coins of smaller denominations.

  At the same time, the most important money of account was a different unit: the pound sterling, worth twenty shillings or 240 pence. This was “imaginary money.” Pounds did not actually exist as coins or paper currency until the nineteenth century, but they were the standard money of account throughout the English-speaking world for many years. In the United States, elderly people continued to keep their books in pounds sterling as late as the 1830s, half a century after independence.

  By the mid-nineteenth century, Americans abandoned this dual system. But even today the people of Britain still use different money of exchange and money of account. By a curious irony of monetary history the major units reversed their roles in Britain. The pound sterling became the leading money of exchange—first in the form of elegant banknotes, then small and clumsy coins of base metal that make a dreary clunking sound when dropped on a modern plastic counter. Guineas, on the other hand, have become a money of account. They rarely circulate but are used to reckon prices of luxury items. As recently as 1990, the author was billed in guineas by a private physician in Harley Street, but the bill was settled by the passing of pounds. Rolls Royce automobiles and tickets to opulent Commemoration Balls in Oxford Colleges are priced in guineas but paid in pounds.

  Dual systems of this sort were widespread in the early modern era. Their complexity was compounded by multiple moneys of account. The great merchant banks of medieval Italy kept their books in imaginary money of account, which had a value that was unique to each house. This “banco money” rose and fell with the reputation of each banking house, even where monetary units were nominally the same.

  The difference in value between one money of account and another was called by Italian bankers the aggio, or premium. That word entered common usage throughout Western world. It was often written agio in Frenc
h, German, Spanish and English, and is still used in Europe.

  Money of exchange also had many complexities. It consisted mostly of silver coin in medieval Europe. During the late medieval and early modern era a bimetallic standard was widely adopted. Gold and silver coins were minted in great variety, but a few units of value became common through many monetary systems. In the seventeenth and eighteenth centuries, roughly the same value attached to the French écu, the Spanish peso, the Dutch rijksdaalder, the German Reichsthaler, and, a little later, the Yankee dollar. All were worth about five English shillings, or one-quarter of an English pound.

  England’s golden guinea (after 1726) was approximately equivalent to the French louis d’or, which was also called the “French guinea.” Before 1726, the louis d’or and Spanish pistole were about the same. Dutch and German ducats were roughly equal to Portuguese escudos, at a little less than half an English pound.

  All of these coins passed current in every nation. When a British general fell overboard near Boston, his baggage was found to contain 694 5/8 joannes, 37 moidores, 300 English guineas, 8 1/2 pistoles, 1 French guinea, 1 dollar, 1 copper halfpenny, 26 “small heart” bits of silver, 6 pieces of gold, and 7 small pieces of silver. It was common for raw unminted lumps of gold and silver to be used as money. Value was determined by weight of gold or silver, measured in grains and later grams of precious metal. See W. T. Baxter, The House of Hancock; Business in Boston, 1724–1775 [Cambridge, 1945], 15, 17–21.

  In our contemporary world, money of exchange has become predominantly paper currency. This trend began as early as the seventeenth and eighteenth centuries in countries where gold and silver coins were very rare: New England, New France, Scandinavia and parts of eastern Europe.