and traffic safety has been represented in in- 

 creased R&D funding. 



Character of Work 



While development activities accounted for a 

 considerable larger proportion of the Federal 

 intramural total than either basic or applied re- 

 search throughout the 1961-72 period, the frac- 

 tion devested to development declined from 55 

 percent in 19t)l to an estimated 52 percent in 

 1972. Part of this decline was due to the Apollo 

 program phaseout. Meanwhile, Federal intra- 

 mural basic research increased from 11 percent 

 to 13 percent of Federal intramural R&D funds, 

 and applied research rose from 34 percent to 35 

 percent (figure 64). 



Figure 64 



Federal Obligations for Intramural R&D 

 Performance, by Character of Work, 

 FY 1961-72 



{Billions of DollarsI 

 5 



■66 '67 



Fiscal year 

 la) GNP price deflatof was used to convert current to constant dollars, 

 SOURCE National Science Foundation. 



72 



(est.1 



INDUSTRIAL RESEARCH AND 

 DEVELOPMENT 



Company Size 



and Research and Development 



Research and development is an activity con- 

 fined largely to large companies. The distribu- 

 tion of industrial R&D among different-size 

 firms — as measured by the full-time-equivalent 

 number of scientists and engineers engaged in 

 R&D — is shown in figure 65. In 1971, 85 per- 

 cent of all R&D scientists and engineers were 

 employed by firms with more than 5,000 em- 

 ployees, as compared with 70 percent in 1958. 

 Companies with fewer than 1,000 employees 

 accounted for a declining proportion of indus- 

 trial R&D scientists and engineers, from 20 per- 

 cent in 1958 down to an estimated 6 percent in 

 1971. There was little change for companies of 

 intermediate size. 



The above statistics must be treated with cau- 

 tion. The structure of industry has changed over 

 this period. "Research and development," as an 

 industrial activity, was in its early stages of 

 growth during the 1950's. While the number of 

 companies with 5,000 or more employees grew 

 by about 50 percent between 1958 and 1967, the 

 number of firms with less than 1,000 employees 

 actually declined slightly. Some small com- 

 panies naturally grew in size; others were ab- 

 sorbed by larger corporations. A further reason 

 for caution in interpreting these data is the 

 quality of respondent reporting, especially for 

 small companies, during the early years of the 

 survey. 



Despite these cautions the rapid apparent de- 

 cline of R&D in small firms may be a danger sig- 

 nal which calls for further investigation to 

 determine its true significance. There is ample 

 historical evidence to suggest that small firms 

 have produced more than a proportionate share 

 of major innovations, particularly those innova- 

 tions which have been one of a family of succes- 

 sive innovations whose cumulative effect has 

 been to create an entirely new industry. It is 

 possible that the statistics quoted here do not 

 accurately represent the true situation, but it is 

 also possible that they signal a declining rate of 

 technological innovation. 



R&D Intensiveness 



An indication of the degree of "R&D 

 intensiveness" of an industry is the proportion 



77 



