Relative Labor Productivity: Definitional and Operational Considerations
Relative labor productivity can be used as a summary measure for many aspects of a country’s economy, but it is used here primarily as an indicator of internal and external returns to scale. Returns to scale, in turn, are defined as increasing if equiproportionate increases in all factors give rise to more than proportionate increases in output, constant if increases in output are proportionate, and decreasing if increases in output are less than proportionate.
Increasing returns tend to be associated with industries that intensively use physical and human capital. As a result, labor productivity in these industries will be relatively high, as it will be for a country specializing in such industries, and increasing rapidly.
Other possible measures of increasing returns to scale would be plant size, length of production runs, and the like. Unfortunately, consistent cross-national data for the late nineteenth and early twentieth centuries do not exist for these indicators. More important, these measures reflect only internal economies of scale. Only relative labor productivity captures both internal and the more diffuse external economies.
The limited sectoral data that do exist tend to confirm the validity of relative labor productivity as an indicator of national specialization in increasing returns industries and production techniques. In 1924–25, 1969, and 1976, the United States was found to be consistently more productive than the United Kingdom across all industries studied. In only 1947–48 and in only two sectors (beet sugar and manufactured ice) was Great Britain found to be more productive.1 Similarly, in 1968 and 1976, Germany was found to be more productive than the United Kingdom in all but one sector.2 Other studies have found similar patterns.3
With the rapid spread of technology and the practice of industrial targeting, however, this clustering of sectoral productivity levels may be breaking down. In a recent study of sixty Japanese and American industries, six Japanese industries were found to be more productive than their American counterparts despite an aggregate level of Japanese labor productivity less than half that found in the United States.4 Thus, relative labor productivity may now be a less valid indicator of national specialization in constant or increasing returns industries than it was in the past. If present trends continue, it may be even less valid in the future, especially as differentials in labor productivity between countries decline, although large differences—as, say, between the United States and Argentina or Brazil—will still be telling.
As described in Chapter 1, relative labor productivity is defined as national output per worker-hour relative to the average national output per worker-hour in the other middle- and large-sized countries. In restricting the comparison group to the middle- and large-sized countries, it is implicitly assumed that countries compare their gains from trade only against the largest and most prominent members of the international economy. There is considerable indirect evidence to validate this assumption. Before World War I and during the interwar period, for instance, the United States, United Kingdom, France, and Germany appear to have perceived each other as their principal trade rivals and not the often more productive but smaller European nationstates.5 Nonetheless, Table A.1 presents relative labor productivity calculations on a base of sixteen countries (column 1) and a base of the four middle- and large-sized nations (column 2) for 1913. From these comparisons, it is evident that France and Germany would be classified as spoilers and the United Kingdom and United States as opportunists at this time regardless of which method was adopted.
In addition to the problem of external validity, there are two problems of internal validity in the measurement of relative labor productivity. First, there is large debate on the utility of output measured by monetary value of production as compared to output measured by physical volume of production.6 The difficulty of matching actual goods produced and their quality over time and across countries clearly precludes the use of physical output for this study, whatever the merits of the measure.
Second, even productivity comparisons across countries based on monetary values of output are fraught with statistical and methodological problems. These have been discussed elsewhere in the literature and need not be reviewed here.7 I recognize that a considerable margin of error may exist in the data; they should be taken to indicate only general magnitudes and trends.
Most studies of labor productivity are concerned with intranational comparisons over time. Only two sources present their results in the form required here. The first, Colin Clark’s The Conditions of Economic Progress, last rewritten in 1957, is largely out of date.8 The second, Angus Maddison’s “Long Run Dynamics of Productivity Growth,”9 presents national productivity estimates in 1970 U.S. dollars and is used here. For comparison, Clark’s “real output per man-hour” was recalculated as relative real output per worker-hour and compared to the recalculations of Maddison’s estimates presented in the text. The results for 1913 are presented in Table A.2. The two sources are relatively consistent for the United States and the United Kingdom. Clark’s estimates are considerably lower for France and higher for Germany. Nonetheless, the categorizations of these four nation-states as opportunists and spoilers would remain the same regardless of which source was used.
1A. W. Flux, “Industrial Productivity in Great Britain and the United States,” Quarterly Journal of Economics 48 (November 1933): 1–38; Marvin Frankel, “Anglo-American Productivity Differences: Their Magnitude and Some Causes,” American Economic Review 45 (May 1955): 94–112; and A. D. Smith et al., “International Industrial Productivity: A Comparison of Britain, America, and Germany,” National Institute Review, no. 101 (August 1982), pp. 13–25.
2Smith et al., “International Industrial Productivity.”
3These studies are reviewed in Angus Maddison, Phases of Capitalist Development (New York: Cambridge University Press, 1982), p. 103.
4Cited in ibid., p. 103.
5See Ross J. S. Hoffman, Great Britain and the German Trade Rivalry, 1875–1914 (Philadelphia: University of Pennsylvania Press, 1933); Matthew Simon and David E. Novack, “Some Dimensions of the American Commercial Invasion of Europe, 1871–1914; An Introductory Essay,” Journal of Economic History 24 (December 1964): 591–605; and Frank A. Vanderlip, “The American ‘Commercial Invasion’ of Europe,” Scribner’s Magazine 31 (January–March 1902): 3–22, 194–213, and 287–306.
6See Jean Fourastie, Productivity Prices and Wages (Paris: Organization of European Economic Cooperation, 1957).
7See International Labour Office, Measuring Labour Productivity (Geneva: International Labour Organization, 1969); and Franz-Lothat Altmann et al., eds., On the Measurement of Factor Productivities: Theoretical Problems and Empirical Results (Gottingen: Vandenhoeck and Ruprecht, 1976).
83d ed. (New York: St. Martin’s, 1957).
9Banca Nazionale de Lavoro Quarterly Review 128 (March 1979): 3–43.