Question

How did US productivity compare with productivity changes in Germany and Japan between 1971 and 1992?

Answer 1

Countries fall in and out of fashion, and features identified as the secret of success in one country can be shown to be absent in other successful countries. Nevertheless, since the mid-1990s, the United States has been held up by most economic commentators as the model to emulate in policy on productivity performance, with a particular emphasis on the diffusion of ICT in the “new economy” (HM Treasury, 2000; Baily and Kirkegaard, 2004; McKinsey Global Institute, 2002; OECD, 2003). Before that, Germany and Japan were popular models with an emphasis on “flexible production” and “welfare capitalism” but these countries are now widely regarded as failing (Freeman, 1987; Dore, 2000).

International comparisons of productivity OHYHOV by industry of origin are a key measure of economic performance next to comparisons of per capita income and other aggregate measures at the economy wide level. Comparisons of manufacturing productivity have received most attention and are now frequently quoted in the media. Despite greater methodological problems, comparisons of service productivity are also of increasing interest as has appeared, amongst others, from various studies by the McKinsey Global Institute during the 1990s (McKinsey Global Institute, 1992, 1993). Recently international organizations, including the International Labor Office and the World Bank, have also begun to report industry-based productivity measures on a regular basis.1 Industry-of-origin comparisons contribute to the understanding of the determinants of differences in economic performance across countries and regions. The relative productivity standing in agriculture, industry and services is considered of fundamental importance in structural growth analysis. As the ‘structure’ of GDP on the production side involves a bigger service component than on the expenditure side, where some important services such as distribution are ‘disguised’, industry-of origin comparisons contribute to a sharper analysis of the causes of economic growth and of patterns of divergence between nations in growth accounts, catch up and convergence analysis and the exploration of lead country-follower country phenomena. It also strengthens the analysis of the locus of technical progress, in particular when supplemented by micro-oriented investigation of variance in performance between industries and between average and best practice firms. Finally these studies shed further light on the relation between productivity and competitiveness

The most solid basis for industry-of-origin comparisons is provided when all information for each country can be derived from the same primary source. For manufacturing the best source is the national census of production. It shows considerable detail on the output and input structure by industry as well as information on the sales value and quantities of specific products. For the United States we used the 1987 Census of Manufactures, which provides information on about 450 industries and 11,000 products.9 For Germany we made use of the Kostenstruktur der Unternehmen 1987, which is based on an annual survey. It includes some 175 industries, but only covers enterprises with twenty or more employees; enterprises with fewer than twenty employees are therefore excluded from our Germany-U. S. comparison. Product detail on Germany for some 6,000 items is taken from the Produktion im Produzierendes Gewerbe 1987. The basic source for Japan was the Census of Manufactures for 1987, which shows information for about 575 industries in the Report by Industries and for about 1,850 product items in the Report by Commodities. The latest year for which census or survey information is available in all three countries was 1987. For the adjustments for the hours worked per person, we made use of separate sources.

All of the G-7 countries except the United States had their largest increases in manufacturing output during the 1950–73 period. As was the case in other countries, U.S. output growth slowed after 1973, but then it grew faster after 1990, regaining and even surpassing its pre-1973 growth rate. In the other countries, the output growth rate continued to slow after 1990, or made only a partial recovery. (See chart 1, panel 2.) U.S. manufacturing employment, as well as average and total hours worked, remained relatively stable during the last half of the 20th century, compared with Japan and Europe G4. (See chart 1, panels 3 and 4.) Manufacturing employment in the United States increased before 1973, then declined slowly afterwards. This resulted in a small net gain in employment and total hours between 1950 and 2000. As in the United States, employment in Europe G-4 grew before 1973, but employment and hours fell much more steeply after 1973. A similar pattern developed in Japan, where manufacturing employment grew rapidly before 1973, but then stagnated and declined, falling rapidly after 1990. Canada was the only G-7 country to experience growth in manufacturing employment during each of the three periods. Over the entire 1950–2000 period, U.S. manufacturing unit labor costs increased less than those of most other countries, measured in national currencies. The greatest differences, however, occurred in the period after 1973 and before 1990

Currency fluctuations played an important role in determining comparative trends in unit labor costs denominated in U.S. dollars during some periods, especially after 1973, when the Bretton Woods system of controlled exchange rates was replaced by floating exchange rates. The effect on U.S. competitiveness was positive or negative, depending on the period. Looking at trends in unit labor costs denominated in U.S. dollars over the entire 50-year period, one can see that the average U.S. increases were smaller than, or the same as, unit labor cost increases in the other countries

Sectoral analysis of the US/UK case

The importance of services to the changing US/UK comparative labour productivity level in the aggregate economy over the period 1870-1990 can be seen in the sectoral breakdown of comparative productivity levels in Table 4. To get the full picture, however, requires adding to this the sectoral breakdown of employment in the two

The decline in agriculture’s share of employment had a significant impact on aggregate labour productivity because agriculture is a relatively low value-added activity. Hence countries which have remained heavily oriented towards agriculture have also remained poor. Shifting labour from agriculture into higher value added industrial and service sectors hence acted to boost aggregate labour productivity

Sectoral analysis of the Germany/UK case

The importance of services to the changing Germany/UK comparative labour productivity level in the aggregate economy over the period 1870-1990 can be seen in the sectoral breakdown of comparative productivity levels in Table 5, together with the sectoral employment data in Table 6. Again, the first point to note is that the long run trends in comparative productivity levels for the aggregate economy are less affected by trends in industry than is commonly thought. Thus, for example, between 1911 and 1990, the German labour productivity lead in industry declined while for the aggregate economy Germany went from three-quarters of the British level to a lead of more than 25 per cent. However, over shorter periods, there have been substantial

Second, although Germany’s comparative productivity position in agriculture has improved since the late nineteenth century; agricultural labour productivity remained much lower in Germany than in Britain in 1990. Again the declining importance of agriculture as a share of the labour force requires emphasis. Since the shift of labour out of low-value-added agriculture occurred much later in Germany than in Britain, and even substantially later than in the United States, this had important implications for the lateness of German catching-up at the aggregate economy level. With such a large share of the German labor force tied up in low productivity agriculture before World War II, the overall labour productivity level was bound to be much lower in Germany than in Britain. On the other hand, once Germany shifted decisively out of agriculture after World War II, overall catching-up was rapid

European countries

Comparable data are available for five non-G-7 European countries: Belgium, the Netherlands, Sweden, Denmark, and Norway. The trends in manufacturing productivity and unit labor costs in these countries over the 1950–2000 period were, on the whole, similar to the corresponding trends in the Europe G-4 countries. (See table 4.) The greatest differences were in Norway. Denmark, Norway, Sweden, and the Netherlands all achieved their highest manufacturing productivity and output growth rates during the pre-1973 years. However, employment growth was offset by declines in average hours to the point that total hours worked either remained stable or declined in all four countries. Among the Europe G-4 countries, only the United Kingdom experienced a similar decline in total hours during the 1950–73 period

. It appears that manufacturing value added was about half that of the United States in Japan, and just over a quarter of the U.S. level in Germany. In both Germany and Japan the real output in U.S. dollars in basic metals and metal products and in machinery and transport equipment is relatively large compared with the other branches.

Trends in Comparative Labor Productivity

The 1987 benchmark results for labor productivity were extrapolated on the basis of national time series for real output and labor input. Table 3 shows the annual compound growth rates of value added and labor productivity from 1950 to 1990. It appears that throughout the postwar period, Japan showed the fastest growth of output and productivity for manufacturing as a whole, although it experienced a serious setback during the second half of the 1970s, when productivity growth fell from 11.0 percent a year in the earlier period to only 5.4 percent. Japanese growth was especially rapid in the investment goods sector, and the slowdown of productivity growth during the 1970s was also less in that branch than in the others (from 12.8 percent in 1965-73 to 9.1 percent in 1973-79). In Germany growth slowed throughout the period, but the setback was particularly large during the 1980s, when the average productivity growth rate for total manufacturing was only 1.8 percent a year, and even lower in food products (0.8 percent) and chemicals (0.3 percent). U.S. productivity growth in manufacturing was slower than in the other two countries, until the 1980s, when the growth rate recovered, especially in chemicals and machinery and equipment.23 U.S. productivity growth in these latter branches was much faster than in Germany during the last decade, a fact which underlies the diverging trend in the comparative productivity performance between the two countries