Quality Improvement, Cost Reduction, and New Goods and Services

Industries that replace existing products with unique products rely heavily on scientific discovery. Siegel (1994) provides the number of new products introduced for broad categories of U.S. industry. Having defined new products as those appearing in different time intervals (1972-77 and 1977-82), Siegel calculates the percentage of shipments in 1982 (in dollar values) devoted to these new products for 20 manufacturing categories (SIC codes 20-39). The two industries with the lowest proportion of sales in new products from 1977 to 1982 were primary metals and electric machinery, with 6.6 and 9.0 percent, respectively, while the two with the highest proportion were nonelectric machinery and instruments, with 26.3 percent and 32.0 percent, respectively. For the 1972-77 period, the percent sales for the instruments category was again highest, this time at 47.2 percent. The second highest category in 1977, however, was petroleum, at 38.2 percent, and the lowest two categories were furniture at 7.4 percent and apparel at 9.4 percent. In 1977, primary metals registered 22.8 percent and electrical machinery 25.1 percent, which differ dramatically from their percentages in 1982.
5 The inconsistency across periods suggests that overall trends in industrial use of R&D could be difficult to observe in terms of new products.

If a particular good or service is defined in terms of scientifically measurable quantities, then measurement difficulties involving quality changes could sometimes be avoided. For example, Nordhaus (1994) examined "lighting," in which he defined quantities of light in lumens, the same measurement employed by physicists studying light flow. He found that, because of scientific and engineering advances in lighting technology and in energy generation and transmission, the amount of human labor needed to produce light throughout history has gone down dramatically. In particular, the amount of light that, in 1800, required 5.4 hours of total labor for all aspects of production, required only 0.22 hours in 1900, 0.00060 hours in 1990, and 0.00012 hours in 1992. Nordhaus does not associate these achievements with the R&D expenditures that made them possible. Nonetheless, he demonstrates that, in terms of basic, physical definitions of progress (such as lumens produced per hour of human labor), the benefits of scientific advancement can be quite large and tend to be much larger than what is often revealed by traditional, economic measures of change.




5 Because his study focused on price-indexing, Siegel does not offer an explanation for why these percentages across time were different. A simple explanation does not exist, although the circumstantial, individual successes of a small number of new products might provide a partial explanation.


Page
Top
Previous
Section
Next
Section
Chap
TOC
Refs
Seind96
Home
Abbrev
Help
SRS
Home