and thus do not exactly reflect price 

 structures specific to the costs of R&D 

 in a country. This is somewhat of a 

 limitation in terms of making precise 

 international comparisons with re- 

 gard to R&D; nevertheless, the use 

 of PPPs to convert R&D expenditure 

 data and national income data still 

 yields a more accurate picture of rel- 

 ative spending levels than the use 

 of market exchange rates. For a de- 

 tailed discussion of the methodol- 

 ogy for computing PPPs, see OECD, 

 Purchasing Power Parities and Real 

 Expenditures in the OECD (Paris: 1985). 



basic research 



Although basic research is de- 

 fined in virtually identical terms in 

 both the United States and Japan (i.e., 

 research that is conducted for the 

 purpose of advancing scientific 

 knowledge and without specific ap- 

 plications being directly sought), the 

 interpretation of this definihon seems 

 to differ in Japan and the United 

 States. Understandably, the alloca- 

 tion of research as basic or applied 

 is a subjective process, and even in 

 the United States there is disagree- 

 ment over the distinction between 

 these two types of research.^ 



The conceptual differences be- 

 tween Japanese and American per- 

 ceptions arise in allocating research 



that is not "truly" basic research (e.g., 

 particle physics). In the United States, 

 the relationship between basic and 

 applied research is typically seen as 

 both mutually exclusive and linear: 

 applied research cannot be basic, and 

 applied research occurs after basic 

 research in a unidirectional flow of 

 research activity. Once a goal or ob- 

 jective other than the pure pursuit 

 of scientific knowledge is assigned 

 to a research program, then it is tra- 

 ditionally seen as applied — not 

 basic — research. 



In Japan, however, basic and ap- 

 plied research are perceived in a more 

 multidimensional and interactive 

 way than in the United States. One 

 of the key differences between the 

 two countries seems to be that in 

 considering whether research is ap- 

 plied or basic, the type of science 

 (basic or applied) being used is also 

 considered. Thus, a long-term ap- 

 plied research program (one with a 

 problem or application in mind) that 

 extensively uses basic science ex- 

 perimentation for its problem reso- 

 lution might be classified as basic 

 research in Japan.* Consequently, 

 Japanese data on basic research ex- 

 penditures probably include some 

 types of research that would be clas- 

 sified as applied research in the 

 United States. Note that the Japa- 

 nese exclude R&D in the social sci- 

 ences and humanities in the 

 allocation of research by character of 

 work. 



^See Arthur Gerstenfeld, Scienct' PoliOf Perspec- 

 tives: USA'Japnn, (New York, New York: Academic 

 Press, Inc., 1982), especially "Part I: Perceptions of 

 the Nature of Basic and Applied Science." 



^See Gerstenfeld, op. cit., and Institute for Future 

 Technology, op. cit. 



manufacturing 

 industries 



At the aggregate level, the Stand- 

 ard Industrial Classifications for 

 manufacturing industries in Japan 

 and the United States are very sim- 

 ilar, with a key exception being the 

 placement of computers in the clas- 

 sification scheme. In the United 

 States, computers are classifled as 

 "office, computing, and accounting 

 machines" under general machin- 

 ery; in Japan, they are classified un- 

 der electrical machinery as 

 "communications and electronic 

 equipment." To make the R&D data 

 in this report as comparable as pos- 

 sible, the U.S. general machinery 

 subgroup "office, computing, and 

 accounting machines" data has been 

 moved to "communications and 

 electronic components" under elec- 

 trical machinery. 



This adjustment was made be- 

 cause (1) the U.S. data were readily 

 available, and (2) only one straight- 

 forward statistical adjustment was 

 required to make the two categories 

 (general and electrical machinery and 

 their subgroups) completely com- 

 parable. To adjust the Japanese data 

 would have required several statis- 

 tical changes for which there were 

 insufficient data. 



As a last note, the Japanese Stand- 

 ard Industrial Classification does not 

 contain a separate "3-digit level" 

 category for aircraft and missiles. 

 Aircraft are included under "other 

 transportation equipment"; missiles 

 are not identified in the classifica- 

 tion schedule. 



47 



