4
Need for a Conceptual Framework

David Mowery of the University of California at Berkeley's Haas School of Business pointed out that the policy agenda and specific policy concerns are continually changing and lack the continuity needed to guide a national statistical program incorporating time series information. Furthermore, policy and program data requirements tend to be narrowly focused and are unlikely to accord high priority to collecting the core data needed to establish a foundation of information on industrial innovation and its impact on the economy as a whole. Accordingly, many participants in the workshop called for more work on developing a conceptual framework of the innovation process to guide data collection activities and to provide a base for analysis of the data. The specific policy-making needs for information on industrial innovation summarized in the previous examples provide partial but not sufficient guidance for the National Science Foundation and other programs responsible for national statistics on innovation.

The most common, intuitive conception of innovation as a linear progression from research effort to invention to commercial application, development, and marketing misrepresents the way innovation occurs in a complex industrial economy. Anecdotal evidence reveals multiple factors interact to affect the development and commercialization of new technology. Workshop participants from industry, academia, and government offered descriptions of parts of the process but conceded that no theory adequately describes and explains private-sector innovation processes, how and why they are changing, and the implications of the changes for the performance of industry and the national economy. One function of an analytical framework is to help guide data collection and analysis of the fundamental determinants of innovation and performance. Participants noted that there has been speculation about what these determinants are,



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 13
--> 4 Need for a Conceptual Framework David Mowery of the University of California at Berkeley's Haas School of Business pointed out that the policy agenda and specific policy concerns are continually changing and lack the continuity needed to guide a national statistical program incorporating time series information. Furthermore, policy and program data requirements tend to be narrowly focused and are unlikely to accord high priority to collecting the core data needed to establish a foundation of information on industrial innovation and its impact on the economy as a whole. Accordingly, many participants in the workshop called for more work on developing a conceptual framework of the innovation process to guide data collection activities and to provide a base for analysis of the data. The specific policy-making needs for information on industrial innovation summarized in the previous examples provide partial but not sufficient guidance for the National Science Foundation and other programs responsible for national statistics on innovation. The most common, intuitive conception of innovation as a linear progression from research effort to invention to commercial application, development, and marketing misrepresents the way innovation occurs in a complex industrial economy. Anecdotal evidence reveals multiple factors interact to affect the development and commercialization of new technology. Workshop participants from industry, academia, and government offered descriptions of parts of the process but conceded that no theory adequately describes and explains private-sector innovation processes, how and why they are changing, and the implications of the changes for the performance of industry and the national economy. One function of an analytical framework is to help guide data collection and analysis of the fundamental determinants of innovation and performance. Participants noted that there has been speculation about what these determinants are,

OCR for page 13
--> at both the firm and industry levels, due to the lack of adequate empirical data. The approach taken by some researchers has been to look at the relative importance of a host of firm characteristics and practices, industry demand conditions, the nature of technological opportunity as afforded by knowledge flows and the vitality of the underlying science, and the means of capturing profits from innovation as determinants of the level of effort and effectiveness of innovation. Although there has been considerable analysis of some manufacturing sectors, there was general agreement that it is not clear how to assess innovation determinants in the emerging industries and the services sector generally. A coherent analytical framework is also needed to help identify and assess observed changes in the innovative activities of firms. Because they appear to represent structural changes, several widely perceived trends in industrial research and innovation should be captured in national data series or should at least be considered candidates for systematic tracking. These trends include Changes in the composition and orientation of research and innovation, including an apparent shortening of the time horizon of corporate investments, a focus on incremental improvements in technology, and shifts in R&D performance among sectors of the economy; Changes in the organizational structure of R&D and innovation, including greater corporate reliance on external sources of technology and collaborative arrangements with public and private institutions domestic and foreign; Changes in the location of R&D and innovation, including domestic regional clustering of technology-intensive enterprises and increased foreign direct investment in and out of the United States; and Innovation across manufacturing and service industries in information technology that is developed largely but not exclusively outside the user industries and is often supplied by intermediaries such as systems integration and consulting firms. As a starting point in the discussion of a conceptual framework, innovation has been defined as the practical use of an invention to produce new products or services, to improve existing ones, or to improve the way in which they are produced or distributed. Innovations include technologically improved products or processes, where processes may involve changes in equipment, human resources, or working methods (Statistics Canada, 1996). Fred Gault, Director of the Science and Technology Redesign Project at Statistics Canada, suggested a narrower and simpler definition of innovation as ''the commercial use of invention," as distinct from pure knowledge-creation activities commonly referred to as basic or fundamental research. This definition leaves the status of industrial reorganizations and new management approaches that can greatly affect performance ambiguous.

OCR for page 13
--> Workshop participants were in general agreement that industrial innovation is a nonlinear process with continuous interaction of elements including Research and development, Adoption and adaptation of new technology, Design, Training linked to innovation, Tooling up and start-up activities, Marketing, Investment in machinery and equipment, and Production engineering. When analyzing innovation using the nomenclature of "inputs" and "outputs," it is important to keep in mind the interactive and simultaneous nature of these many components. All of the above elements may be considered, in a sense, inputs to the innovation process. The term "inputs to innovation" generally refers to the resources committed to innovation, including the investment in R&D and intellectual capital. "Innovation output" refers to the new products and processes produced. Finally, there is the "outcome" or impact of innovation on the firm, economy, and society. The relationship between the inputs and outputs or outcomes provides information on the productivity or effectiveness of the R&D and other investments. Of interest here are the factors and conditions that determine the incentives for, and success of, innovation investments. Adam Jaffe, Brandeis University professor of economics and coordinator of the Project on Industrial Technology and Productivity of the National Bureau of Economic Research, noted that when discussing data on the inputs, outputs, outcomes, and productivity of innovative activity, it is important to distinguish not only between outputs and outcomes or impacts but also between the underlying concept of concern and the proxy or measure used to indicate it. Some concepts and their proxies for innovation inputs, outputs, and outcomes are given in Table 4-1. It is not exhaustive. Another categorization of the components of innovation was offered by Bill Long, president of Business Performance Research Associates, Inc., who identified technology production inputs as including not only tangible capital equipment and scientists and engineers but also factors influencing the research methods employed, such as the impact of computer technology on DNA sequencing and Internet communications. Although much analysis of the inputs to innovation focuses on R&D expenditures, it was frequently remarked at the workshop that this is only one of the

OCR for page 13
--> TABLE 4-1 Innovation: Concepts and Proxies Category Concepts Proxies Inputs (Investments) Person-years, equipment-years Expenditures Outputs (Products) Ideas, discoveries Papers, prizes   Inventions Patents, invention disclosures   Human capital Degrees awarded Outcomes (Impacts) Broad advance of human knowledge Papers, citations, expert evaluations   New products Patents, citations, innovation counts, new product sales   Productivity improvements Measured productivity growth   Income growth Benefit/cost ratio or rate-of-return estimates   Source: Presentation by Adam Jaffe, Brandeis University factors contributing to the innovative effort. Analyses of the total costs of innovation necessarily incorporate non-R&D inputs to innovation such as prototype design, testing, production engineering, start-up investment, and marketing. A key feature of innovation outputs and outcomes is that they often occur as spillover benefits to other firms or parts of the economy. Spillovers occur when beneficial effects of a firm's research and innovation activity accrue to other firms or industries without compensation to the investing firm. In this sense they refer to the failure of markets to fully capture the economic value of an innovation. Spillover effects can occur as productivity improvements or as new products and processes. Although studies have attempted to estimate interindustry spillovers, the constraint has been the lack of empirical data on the flows of knowledge and technology, an uncertainty of what to look for, and inadequate measures of output and outcomes that may indicate the effects of the spillover. Many workshop participants remarked on the importance of this empirical problem, noting that for many industries or technologies the spillover effects are at least as significant as the appropriated returns. Some analysts seek to track and model the generation, transfer, absorption, and application of new knowledge through the economy. Adam Jaffe noted, however, that describing or measuring knowledge-flow transactions improves understanding of the economics of R&D, but it conveys little about the spillover effects. To measure spillover effects on productivity, for example, one needs to look at different kinds of outputs and outcomes at different levels of aggregation. This underscores the importance of improving output and outcome measures. Fred Gault of Statistics Canada noted that some students of the process have

OCR for page 13
--> used the concept "national innovation system" to capture the array of a country's social institutions and practices that influence innovation, while others believe innovation systems are local or regional phenomena, albeit dependent upon global markets. Other participants in the meeting commented that better links between science and technology data and regional economic data are needed to increase understanding of the inputs to innovation and its economic development results. Wesley Cohen, professor of economics and social sciences at Carnegie Mellon University, remarked that because our understanding of innovation is incomplete and is itself changing, data collection efforts should try to incorporate new information that could contribute to the development of a conceptual framework of innovation. For example, statistical agencies should systematically identify and probe through pilot surveys new phenomena that anecdotal evidence suggests are important. An example mentioned by several participants is that alliances between and among otherwise competing firms are becoming more common.