The relationship of size to research in American industry
O'Brien, Bruce L.
MetadataShow full item record
Spending for research and development accounts for a significant and integral part of the nation's gross national product. In 1953, private industry conducted or financed more than five billion dollars worth. The major purpose of this thesis is to determine if a relationship between size and research in American industry can be found through the use of a statistical model. Within the various size groups in private industry, the small firms encounter situations which help explain variations in research spending among the companies. When a small company attempts to embark upon a research program of its own, the financial barrier is often too great to overcome. To operate their own research facilities requires a large initial outlay of money, and there is a certain minimum annual expenditure that must be make to keep the program running. While outside facilities such as the university type laboratories, the trade association laboratories, and the consulting laboratories offer inexpensive research programs, the small firms have, for the most part, preferred to do RD work in their own laboratories. Although certain unique advantages are thereby secured, it is apparent that more research spending would be incurred if small firms directed more research dollars into the excellent outside facilities. At the other end of the size scale, the large firms do not encounter these financial difficulties and have established a more imposing record. While only 10% of the small firms conduct or finance research and development programs, 90% of the large firms have programs. Moreover, the large firms account for 70% of all the industrial RD done, although they employ only 40% of those people who work for firms doing research. Despite the fact that there is a lack of agreenent among theorists as to what kind of market conditions does the most to foster industrial research, large firms, just by their size alone, possess a superior position in the abiltty to conduct sound RD programs. More specifically, large firms are a necessary factor in insuring high levels of private spending for research and development in American industry. Government research contracts accounted for one third of the total research and development spending done by American industry in 1953. Because of a system of sub-contracts which allows a company without expensive apparatus to bid on, or otherwise obtain, government RD contracts, small and medium sized firms have been given an opportunity to add to their research division's programs. Although it has been found that small firms do not supplement government RD work with their own, to the extent that larger firms do, it is suggested that the government should offer smaller firms a greater share of the contracts in order to stimulate more competition in industry. In attempting to determine the relationship of size to research, government RD contracts, market conditions, and growth were considered in order to determine that effect they might have on RD spending. Through the use of multiple correlation, it was found that the four independent variables explained only 40% of the variation in the dependent variable, RD spending. The size factor, by itself, exnlained 34% of the variation, the other three independent variables explaining only negligible amounts. While the results of the statisticel study aid not significantly explain the levels of RD spending and a lack of suitable data made a proper explanation of the remaining variation impossible certain sound conclusions were drawn from the correlations. First, market conditions, growth and government contracts (subject to imperfections in their use in the mathematical operations) are not a criterion in explaining what determines variations in RD spending among industrial firms. Also, size as a measure, while not explaining a large enough percentage of the variation to be considered significant, i.e. the major criteria, it does go a long way towards filling the gap. In conclusion, it must be pointed out that the question of what causes variation in RD spending among industrial firms was examined from only one direction and even within this narrow area, up-to-date and the most suitable data was not available.
Thesis (M.A.)--Boston University