of the total human and financial resources which 

 it invests in R&D. A high level of R&D, how- 

 ever, does not necessarily insure that an indus- 

 try is highly innovative. For example, an indus- 

 try may direct its R&D activities toward prod- 

 uct differentiation, rather than to more basic 

 technological innovation. 



Two frequently used indicators of R&D 

 intensiveness are (a) the number of R&D scien- 

 tists and engineers as a percentage of all em- 

 ployees, and (b) the percent of net sales devoted 

 to R&D. These indicators are used in the follow- 

 ing to estimate the level of R&D intensiveness of 

 U.S. industry as a whole, as well as of manu- 

 facturing and nonmanufacturing industries, and 

 to determine its relationship to company size. 



U.S. Industry. The first of these indicators 

 shows that for U.S. industry as a whole, the 

 number of R&D scientists and engineers in- 

 creased from 25 per 1,000 employees in 1960 to 

 30 in 1064-b5, but then declined to 24 in l^yo. 

 Similarly, the percent of net sales devoted to 

 R&D rose from 4.2 in 1960 to 4.6 in 1964, and 

 then declined progressively to 3.8 in 1970 (figure 

 66), 



The changes between 1960-70 in both indi- 

 cators show that the R&D intensiveness of U.S. 

 industry, in terms of scientific and engineering 

 manpower, increased by 20 percent between 

 1960-64 but then declined to a level in 1970 

 which was lower than 1960. The percent of net 

 sales shows the same trend; the ratio rose by al- 

 most 10 percent between 1960-o4 and then fell 

 continuously to a level in 1970 well below that of 

 I960. 



Manulacturiiig buiuslriei. Individual manu- 

 facturing industries differ greatly in the extent 

 of their R&D intensiveness. Based on the two 

 indicators, each of the 15 major industries in the 

 manufacturing sector was placed into one of 

 three groups according to the level of its R&D 

 intensiveness in 1970. The resulting groups, 

 each consisting of five industries, are presented 

 in the table on page 79; Group I industries are the 

 most R&D intensive and Group III the least. 



The level of R&D intensiveness remained 

 essentially constant for industries in Groups II 

 and III between 1960-70, but declined for Group 

 I industries after 1964 (figure 67). The extent of 

 the latter reduction was approximately 25 per- 

 cent by 1970, as measured by each of the indi- 

 cators. The declines in the most R&D-intensive 

 group appear to be due almost entirely to reduc- 

 tions in Federal funding of defense- and space- 

 related R&D in these industries. 



Nonmanufacturing huiuitries." Only a small 

 number of companies (estimated at 1,100-1,200) 

 in this sector perform R&D. The R&D inten- 

 siveness of those which perform any R&D in- 

 creased by 50 percent between 1961-66, and re- 

 mained at essentially that level in later years.'" 



" These include (but are not limited to) agriculture, 

 mining, transportation, public utilities and sanitary serv- 

 ices, wholesale and retail trade, finance, insurance, business 

 services, medical and dental laboratories, and engineering 

 and architectural services 



'"National Science Foundation, Reiearch and Development in 

 huiuitrn. 1970 (NSF 72-309). 



78 



