A revealing window on the U.S. economy in depression and war: hours worked, 1929-1950.

• Author: Higgs, Robert
• Position: Etceteras ... - Essay

Many years after the Great Depression and World War II, controversy continues to swirl as scholars, pundits, and ordinary citizens look back at the watershed events of the 1930s and 1940s. Economists and economic historians have assessed the economy's condition during these momentous years primarily with reference to the usual macroeconomic indicators, especially the real gross domestic product (GDP) and the rate of unemployment. For these analysts, the Great Depression is almost defined as the long period when real GDP remained well below its trend high-employment capacity and the rate of unemployment stood persistently above its normal range. The war period, in contrast, stands out in the standard statistical series as a time when real GDP appeared to increase phenomenally and the rate of unemployment fell almost to zero.

Interpretation of economic events in the light of such conventional measures has been complicated, however, by institutional peculiarities unique to these extraordinary times. Both real GDP and the rate of unemployment are difficult to interpret in the usual way, the former because of the operation of a wartime command economy, complete with comprehensive price controls and a multitude of other significant departures from market pricing and resource allocation, and the latter because of large-scale, atypical forms of government employment, especially the emergency work-relief programs during the Depression and the military conscription of labor during the war.

I examine here what we can learn by focusing on a different, seldom-considered measure--namely, employment of labor as measured by hours worked. This alternative way of looking at the economy's operation helps us to avoid a number of difficulties, such as the exceptional frequency of reduced-hours employment ("work spreading") during the 1930s (Lyon et al. 1935, 830-44; Lebergott 1964, 185-86; Taylor 2008, 19-21, 29-30) and the increased prevalence of overtime work during the peak years of the war. By using hours worked as our measure of employment, we avoid the necessity of distinguishing who is employed and who is unemployed (Lebergott 1964, 184-85; Darby 1976) and of arbitrarily imposing a cut-off line for determining who has "full-time work." Not all hours worked are equal in economic significance, of course, so we do not avoid all difficulties of analysis by taking this approach, but some of these remaining difficulties can be reduced by disaggregation of the economy's total hours worked into its component sectors.

Table 1 shows the basic data to which I make principal reference here. I have drawn these data from John Kendrick's monumental study Productivity Trends in the United States (1961). Kendrick carries more significant digits than I show in the table, reporting his figures for hours in millions, but in my judgment such precision is spurious. Indeed, we are taking a substantial chance by relying even on the figures given here, in billions of hours, with one digit after the decimal point. I make this observation not to criticize Kendrick, who describes in great detail the enormous amount of careful effort that he put into making his estimates, but only to recognize that even an analyst as painstaking as Kendrick could not overcome many problems, especially those stemming from the absence of annual source data for many of his component categories. Errors of various sorts no doubt remain embedded in these figures, not all of them offsetting in the aggregates, and readers are advised to bear this unavoidable situation in mind as they consider the present discussion.

The 1930s

In most discussions of the Great Depression, the macroeconomic profile of the subject is portrayed as follows: steep continuous decline from 1929 to 1933, sharp recovery from 1933 to 1937, severe but short "depression within the depression" from 1937 to 1938, and renewed rapid recovery from 1938 onward, with the economy having fully recovered by 1940 or, at the latest, 1941. With regard to hours worked, the profile looks somewhat different, however.

Total hours worked fell substantially from 1929 to 1932. Then, unlike the standard depiction of the economy's course, they hit bottom and stayed put in a virtually flat-bottomed trough for three years, 1932, 1933, and 1934. They then rose substantially until 1937, dropped by 7 percent in 1938, then rose again thereafter. However, even as late as 1940, total hours remained below the 1929 level by 6 percent, and only in 1941, with the population vigorously engaged in mobilization for war, did total hours exceed the 1929 value, by 3 percent, Meanwhile, of course, the population and the potential labor force had grown substantially, the former by 11.6 million persons, so simply getting back to the 1929 level of hours worked represented something less than a complete triumph.

As table 1 shows, military employment remained quite low and did not vary substantially from 1929 to 1939. Similarly, farm hours worked varied little, although after remaining fairly steady from 1929 to 1933, they dropped in 1934 and never regained...