funding in Japan and the United 

 States, both countries exhibit similar 

 patterns of R&D by performer. In 

 1985, industry expended the major- 

 ity of R&D funds (67 percent for Jap- 

 anese industry and 73 percent for 

 U.S.), higher education accounted 

 for the second largest share of R&D 

 expenditures, and government ac- 

 counted for nearly all of the remain- 

 der (chart 6). In terms of the flow of 

 R&D funds from source to per- 

 former, there is a higher level of funds 

 transfer among sectors in the United 

 States than in Japan, especially from 

 government to industry. In Japan, 

 the flow from government to higher 

 education is greater.'' 



'Japan and most European countries include in 

 R&D a category' of expenditures called "general uni- 

 versity funds" (GUF), which are nationally allocated 

 funds for the general operaring expenses of colleges 

 and universities. The United States has no equiv- 

 alent to GUF, since nearly all nationally allocated 

 R&D monies are "separately budgeted R&D " Thus, 

 the larger Japanese flow from government to higher 

 education is mainly due to the component of gen- 

 eral university funds in government expenditure 

 statistics. U.S. State Governments support more than 

 one-half of all American universities, but virtually 

 none of the general operahng expenses of these 

 State universities are included in US, R&D expend- 

 itures. (For a more detailed discussion of these dif- 

 ferences, see appendix A.) 



r&d by character 

 of work 



While conceptual distinctions can 

 be established between basic and 

 applied research and between R&D, 

 there are obvious difficulties in mea- 

 suring and assigning specific R&D 

 activities to the respective catego- 

 ries. These difficulties are increased 

 when making international compar- 

 isons of absolute levels of R&D by 

 character of work. Therefore, greater 

 stress should be placed on evaluat- 

 ing trends within each country over 

 time.* 



By 1985, the overall distribution of 

 total R&D funds within the two 

 countries by character of work was 

 very similar, although it appears that 

 the share of Japanese R&D devoted 

 to basic research has declined some- 

 what over the past 10 years. In 1975 

 and 1980 basic research accounted 

 for 15 percent of R&D expenditures; 



'For a discussion of differences in U.S. and Jap- 

 anese definitions and treatments of basic research, 

 see appendix A. 



by 1985, however, this share had de- 

 clined slightly to 13 percent (table B- 

 6).^ In the United States, basic re- 

 search as a share of total R&D re- 

 mained constant during 1975-85 at 

 12-13 percent. In 1985, Japan de- 

 voted a marginally higher propor- 

 tion of R&D to applied research than 

 did the United States (25 percent and 

 22 percent, respectively); in con- 

 trast, the United States devoted rel- 

 atively more expenditures to 

 development — 66 percent and 62 

 percent, respectively (charts 7 and 

 8). This difference largely reflects the 

 greater U.S. defense R&D effort, 

 which is primarily development. 



The distribution of R&D by char- 

 acter of work among performing 

 sectors is also very similar for the 

 two countries, with higher educa- 

 tion placing nearly equal emphasis 

 on basic research (56-58 percent) and 

 industry placing similar emphases 

 on basic research (4-6 percent), ap- 

 plied research (20-22 percent), and 

 development (72-76 percent). 



For the past several years, Japa- 

 nese research leaders — including the 

 Prime Minister's Council for Science 

 and Technology — have urged an in- 

 creased emphasis on "fundamental 

 research." (The combination of basic 



Chart 6. National R&D expenditures by performer: 1985 



Percent 

 100 



80 



60 



40 



20 



^^ Japan 

 ^^H United States 



rk 



100 



80 



60 



40 



20 



Government 



Industry 



Higher education 



Other 



SOURCE National Science Feaindalion, SRS. table B-5 



'Because of changes in the methodology of the 

 key Japanese national R&D survey, it is not possible 

 to calculate basic research from these data prior to 

 Japan fiscal year 1974. However, in the Science and 

 Technology Agency's White Paper on Science and 

 Technology, 19S3 (Tokyo, Japan, 1983), it was esH- 

 mated that basic research was 28 percent of R&D 

 expenditures in 1967. 



This relative decline in basic research expendi- 

 tures in Japan has occurred exclusively in the higher 

 education sector Here, basic research declined from 

 72 percent in 1975 to 56 percent in 1985; in turn, the 

 share of applied research increased by the same mag- 

 nitude, rising from 20 percent to 36 percent. How- 

 ever, real higher education expenditures on basic 

 research doubled during this period, indicating that 

 the relative dccbne is principally the result of a higher 

 rate of growth in applied research expenditures. It 

 is also possible that, over time, Japanese researchers 

 view the distinction between basic and applied re- 

 search in a manner more similar to US. percep- 

 tions. (See appendix A.) This would imply that the 

 relative decline of basic research (and the concom- 

 mitant increase in applied research) is also a func- 

 hon of a reclassiHcation of research activity, rather 

 than simply a reflection of a shift in research focus. 



