Are There High-Tech Industries or Only High-Tech Firms? Evidence from New Technology-Based Firms

Abstract

Considerable attention has been directed at understanding the structural changes that are generating an increased need for skilled workers. These changes are perceived to be the result of developments associated with the emergence of the new knowledge economy, whose potential is often linked to the growth of new technology-based firms (NTBFs). Where are these firms to be found? Related work on changes in technology and innovativeness has been accompanied by the creation of taxonomies that classify industries as high-tech or high-knowledge, based primarily on the characteristics of large firms. There is a temptation to use these taxonomies to identify new technology-based firms only within certain sectors. This paper uses a special survey that collected data on new firms to argue that this would be unwise. The paper investigates the limitations of existing classification schemes that might be used to classify industries as high- or low-tech, as advanced or otherwise. Characteristically unidimensional in scope, many of these taxonomies employ conceptual and operational measures that are narrow and incomplete. Consequently, previous rankings that identify sectors as high- or low-tech using these measures obscure the degree of innovativeness and human capital formation exhibited by certain industries. In a policy environment wherein emotive scoreboard classifications have direct effects on resource allocation, the social costs of misclassification are potentially significant. Using a comparative methodology, this study investigates the role that conceptualization plays in devising taxonomies of high- and low-tech industries. Far from producing definitive classifications, existing measures of technological advancement are found to be wanting when their underpinnings are examined closely. Our objective in the current analysis is to examine the limitations of standard classification schemes, particularly when applied to new small firms, and to suggest an alternative framework based on a competency-model of the firm. This framework differs from previous attempts in several important respects. First, it constitutes a multidimensional approach to industry classification. As different concepts--such as innovation, technology use, and worker skills--can be used to define high- and low-tech industries, we integrate each of these measures into a unified framework that captures the different dimensions of technological prowess. This, in turn, lessens the degree of bias that may arise due to narrow or incomplete conceptualization. Second, our competency-based approach focuses directly on the population of interest-new small firms. Often at the forefront of product development and advanced technology use, it is these firms that are seen as critical in the transition to knowledge-based production. Basing industry classification on new small firms thus alleviates the bias in favor of large firm characteristics that arises with the use of industry-level aggregates. The competency-based approach developed herein reveals several important findings. First, industry rankings are highly sensitive to the conceptual measures that form the basis for classification. Accordingly, classification schemes that rely on a single-measure of technological prowess, as many do, may incorrectly rank industries and/or classify sectors. Second, firms that possess the advanced competencies that contribute to technological prowess are found in many industries, and are not as sector-specific as previous attempts at classification suggest. Simply stated, low-tech industries are not devoid of high-tech firms, nor, for that matter, are high-tech industries comprised exclusively of high-tech firms. Consequently, broad generalizations at the industry-level may prove dubious. An additional issue addressed herein focuses on the appropriateness of using a single competency-based index as a basis for classifying industries. In our view, the competency-based approach represents a considerable advance over previous efforts: it formally recognizes the multidimensional nature of technological prowess, and it focuses on the correct firm population. That said, can we then infer that a single-ranking derived from such a framework provides sufficient grounds for classification? Or alternatively, should a series of rankings be used when attempting such a task? We find that there are numerous dimensions to technological prowess and that the use of any single index may, in fact, obscure some of these. Consequently, different industries will emerge as high-tech leaders when different aspects of technological prowess are emphasized.