Sources of Technological Progress: An Empirical Investigation

The aim of this study was to empirically investigate the sources of technological opportunitiesas one of the major determinants of technical progress at the industry level - using data from Switzerland. This question was looked at from the perspective of Swiss industry as a whole, as well as from the perspective of interindustrial differences. The analysis was based on a survey conducted among Swiss experts (mostly R&D executives of selected firms) in 1988. Of the 940 experts questioned, 358, or approximately 40%, responded. They represented 127 different lines of business. The most important results can be summarized as follows: 1. Market (profit-oriented) organizations make the most important contributions (of all kinds: financial, individual, informational, etc.) to technical progress. The most important source is firms within the same industry; second is product users; and third, suppliers of materials and equipment used in manufacturing. 2. The contributionof non-market organizations seems relatively unimportant. University research, other government research institutions, state companies and agencies, professional and technical associations and individual inventors make small contributions. 3. The contributions of market and non-market organizations vary from one industry to the other. 16 4. Science also contributes to technical progress, even if only selectively. Education and training in physics, computer science, materials science, electrical engineering, mechanical engineering and applied chemistry are all considered relevant to technical progress in Switzerland. 5. Generally, university research (domestic and foreign) is not considered as relevant to technical progress in the industries surveyed. In certain fields, such as computer science, materials science and electrical engineering, university research does, however, seem relevant to technical progress. These empirical results are important for a science and technology policy of both the state and individual firms. They show what aspects of the innovation process the responding experts find particularly relevant. The results concerning the sources of technological opportunities and those concerning the relevance of education and research in basic sciences, applied sciences and in engineering are particularly important for technical progress. For they exemplify in what areas policies can be pursued. One possible economic policy implication of this study is the necessity of strengthening the institutional infrastructure of technical progress. This would include. (1) encouragement of cooperation in R&D-projects between firms within the same line of business, between producers and users and between the former and the suppliers of inputs and equipmeñt (2) the fostering of cooperation between the institutions of basic science and applied science and between these and private research laboratories, especially in the fields of science and technology that are most relevant to, technical progress. All these measures ought to take into consideration that there are important differences between industries with respect to the nature, mechanisms and institutions of technical progress - as this and other empirical studies on technical innovations have suggested (cf. Dosi, 1988; Cohen/Levin, 1989; Nelson 1988). The empirical results further emphasize the economic policy implications of new developments in the theory of technical progress. Rosenberg summarizes them as follows: "In addition to nourishing the supply side in a broader range of areas, intelligent policies must be directed; at institutional aspects of the innovation process, working to encourage the interaction of users and producers, as well as the iterative interactions between more basic and applied research 17 enterprises... Useful policies would be those directed at the provision of information, from basic research institutions in the non-market sector to private firms and laboratories, as well as from users to producers concerning desired products and characteristics." (Rosenberg 1982: 237f