the United States, Series E-89. From 1913–46, data come from the same source, Series E-84. From 1947, the data are derived from the Economic Report of the President 1990, p. 365. Again, the series are spliced together and indexed with 1982 = 100.
21. The adjusted R2, after inclusion of autoregressive terms, was .3243.
22. Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations (Oxford: Oxford University Press, 1976), book I, chap. 1, especially p. 17.
23. Joseph A. Schumpeter, The Theory of Economic Development (Cambridge, Mass.: Harvard University Press, 1949).
24. See, for example, Peter Temin, Did Monetary Forces Cause the Great Depression? (New York; W. W. Norton, 1976), chap. 3, or his Lessons from the Great Depression (Cambridge, Mass.: MIT. Press, 1989), pp. 43, 105–6. Temin seems somewhat more sympathetic to the “wages” approach used here in his 1989 book than he did in 1976, perhaps because of a spate of studies supporting that approach to explaining Depression-era conditions. For an earlier account with some Keynesian flavor, see John Kenneth Galbraith, The Great Crash, 1929 (Boston: Houghton Mifflin, 1955).
25. This subject is explored in detail in chapter 8.
26. Data on the components of GNP are consistently available only from 1929, so the estimated consumption function regressed consumption against disposable income for the period 1929 to 1989. The residuals from that regression represented deviations of the consumption-income relationship from its long-term norm. Changes in those residuals represented changes in autonomous consumption—that component of consumer spending not directly related to income. The residuals from the consumption function were then regressed against changes in labor productivity to see if a significant positive relationship existed. Similarly, functions were created where gross private domestic investment, net exports, and government purchases spending were regressed against GNP.
4
The Gilded Age
In many respects, the first three decades of the twentieth century were the golden age of the American economy. Despite the absence of national income statistics, it became very apparent during this period that America was the world’s premier economic power. The economic performance of the nation from 1900 to 1929 clearly established that this was “the American century” as far as economic growth was concerned.
Not only did real output expand at a very respectable 3.4 percent annual rate, but the growth far outdistanced that of older economies in Europe. It has been estimated that real output rose only 1.8 percent annually in the area constituting today’s European Economic Community (Common Market).1 There is some dispute as to how the fruits of this considerable economic progress were distributed.2 Yet there is no denying that the typical American family in 1929 owned a car, went to talking movies, listened to the radio, and owned a telephone—none of which was true in 1900. While his framework may be flawed, Walt Rostow makes a valid point in arguing the United States entered an “age of mass consumption” during this era.3
By the standard macroeconomic indicators of today, the first three decades of the century were highly successful. The inflation rate was 2.5 percent per annum, considered high by contemporaries but extremely low by present-day standards. Moreover, most of the inflation occurred in a few years during and immediately following World War I; during peacetime, prices were about as likely to fall in any given year as to rise.
Unemployment, as measured by the official Lebergott/BLS data, averaged only 4.67 percent for the thirty years 1900 through 1929, with the median unemployment rate being 4.30 percent. In only one year, 1921, did the unemployment rate average in double digits, and unemployment was under 3 percent more often (six versus four years) than it was above 7 percent.
Before extending our analysis of unemployment, it is very interesting to note that this era of prosperity and stability occurred during a period of relatively little government involvement in the economy. To be sure, there was an increase in some forms of regulation (passage and enforcement of antitrust laws, and railroad rate and food and drug regulation), the passage of a federal income tax amendment to the Constitution, and the beginning of institutional macroeconomic intervention with the creation of the Federal Reserve system. Collectively these factors led one distinguished economic historian to term the era 1897–1917 “the decline of laissez-faire.”4
Yet by modern-day standards, there was still a good bit of laissez-faire left in the American economy. Government purchases of goods and services absorbed only 8.2 percent of gross national product in 1929, and some 85 percent of that was carried on at the state and local level. After a high level of both spending and regulation during World War I, government involvement decreased during the 1920s, albeit to not quite as low levels as before the war.5 While it is true that total governmental expenditures rose from 7.7 percent of GNP in 1902 to 11.8 percent in 1929, the latter figure is only one-third the proportion of recent times; most of the growth in expenditures reflected high income elasticity of demand for goods and services traditionally provided by government, such as highways and education, rather than bold new forms of government involvement. The era was characterized by both relatively low levels of unemployment and relatively little state involvement in the economy.
The Unemployment Experience, 1900–1929
Moving to a less aggregated view of the period, table 4.1 shows the unemployment experience, by individual year, for the first three decades of the century. Only in 1921 did the unemployment rate average more than 8.5 percent for a single year. The notion that very high unemployment levels were commonplace in the era before major macroeconomic policy intervention by government does not seem to be supported by the empirical evidence. In 80 percent of the years, the unemployment rate averaged less than 6 percent for the year, a rate that today is considered fairly low.
When relatively high unemployment developed during recessionary periods, it did not tend to persist.6 There were three episodes (about one a decade) in which unemployment rose above 8 percent. In none of those instances did unemployment remain at an 8 percent or higher average for two consecutive years. After peaking at 8 percent in 1908, unemployment dropped sharply to slightly over 5 percent the following year. After peaking in 1915 at 8.5 percent, the unemployment rate plunged 40 percent in 1916 to a near normal 5.1 percent rate. Even in 1921, when the unemployment rate soared to 11.7 percent, recovery came quickly, with the unemployment rate falling 43 percent in the following year to under 7 percent.
TABLE 4.1 UNEMPLOYMENT IN THE UNITED STATES, 1900–1929
A similar flexibility in the unemployment rate occurred when low unemployment existed. Only once, in 1918 and 1919, was one year of very low unemployment (2.5 percent or less) followed by a second year of similar very low joblessness. Generally unemployment tended to gravitate fairly quickly back to something approximating its natural rate. Only once, from 1906 to 1908, was the unemployment rate outside the 3 to 6 percent range (annual average) for three consecutive years, and even in 1906–8, the rate went through the natural rate range, going from below 3 percent
