Industrial Research 



121 



company surveyed the agricultural scene and found tliat 

 mechanization had not been extended to the small farm 

 of less than 100 acres. It brought out a 1-plow tractor 

 and a 5-foot combine, which the trade ijrcdicfod would 

 not sell. They not only sold but substantially increased 

 this company's share of the market and competitors 

 were soon in the field with new models of their own 

 design. 



The influence of research on the growth of industries 

 has been clearly shown in the development of the 

 Diesel engine. For j^ears the apphcations of the 

 Diesel engine had been limited to stationary power 

 and marine uses. Gradually the light-weight, high- 

 speed engine was developed. In recent years, the 

 Diesel was applied in trucks, tractors, and locomotivc- 

 imits and the companies which developed thes(^ new 

 uses have notably bettered their comi)etitive positions 

 in their various fields. 



The accumulation of many such cases almost forces 

 the conclusion that there is a positive correlation be- 

 tween research expenditures and growth in earning 

 power. This, however, is rather difficult to measure 

 because of the lack of a generally accepted definition 

 of research, the secrecy on the part of many companies 

 regarding expenditures and the time lag between such 

 expenditures and an earnings return. Research ex- 

 penditures represent a sacrifice of immediate earnings 

 in anticipation of a gi-eater return later on, the time 

 lag bemg greater in the case of pure research than in 

 the case of product development. 



In view of the importance of research as a factor in 

 the growth of companies and industries, it was (h^'ided 

 to tabulate the size of research personnel ami expendi- 

 tures for a number of leading conipanies representing 

 a broad cross section of American industry. Accord- 

 ingly the following letter was directed to a selected list 

 of companies: 



As an investment counsel organization, we are making a 

 study of research expenditures in various industries. Will 

 you be good enough to give us the following information with 

 reference to your company: 



"The approximate amount spent for scientific research and 

 development of products in each of the past few years and the 

 number of men engaged in such work." 



While it is our intention to make the conclusions drawn from 

 our study available to industry generally, we shall treat any 

 information pertaining to your particular company confidential, 

 if so requested. 



Since the primary purpose of the inquiry was to 

 obtain whatever data on research might be available, 

 the request was worded in very general terms, leaving 

 it to the individual companies to determine such ex- 

 pendittires in accordance with their accounting policies. 

 It soon became obvious that there was wide variation 

 in the definition of research and that such data had 

 little value for comparing the activities of individual 



companies. Furthermore, many companies which had 

 reported their research personnel to the National 

 Researcli Council, refused to give out any information 

 on expenditures. Obivously, a statistical tool was 

 needed for estimating expenditures on a comparable 

 basis. 



The first figure sought was the average research and 

 development expenditure per worker, including both 

 salaries and the pro rata cost of supplies, equipment, 

 and overhead. Once this was obtained, it would be 

 possible to estimate the expenditures for each company 

 and also for industry as a whole, based on the number 

 of workers reported to the National Research Council, 

 which data arc ])robal)ly more nearly comparable than 

 any other. 



From all the I'ejjlies received, those were selected 

 which stated both the number of personnel and ex- 

 penditures for research and development. While 

 there is un(iouI)te(lly consiileraide difl'erence of computa- 

 tion among companies, the ratio between the number 

 of workers and expenditures for any one company is 

 highly significant, since they both come under that 

 particular company's definition of research, whatever 

 it may be. 



Although replies were received from a great many 

 additional companies, which gave incomplete data, 31 

 companies reported both the personnel and expenditures 

 for research and development in 1937. This is sum- 

 marized in the following tabulation, without revealing 

 the names of the individual companies, which furnished 

 this information in confidence. 



Reported lexearrh ixpenditnies and personnel for representative 

 CO nipa nies — 1937 



