Technological Resources and the Logic of Corporate Diversification

Brian S Silverman

• Publisher: Taylor & Francis
• Year: 2002
• Print ISBN: 9780415271363
• eBook ISBN: 9781134487585
• Edition: 1
• Topic: Business
• Subtopic: Business General
• Index: Business

This impressive book sees the author applying and extending the resource based view of the firm to explain and predict the strategy of corporate diversification.Technological Resources and the Logic of Corporate Diversification is an original and authoritative book that will be extremely useful to academics and students in such disciplines as busin

1 Introduction

Overview

Corporate diversification¹ is a ubiquitous feature of the modern economic landscape. For most US firms, “[a] long-term continuing strategy of growth [has been based on] expansion into new geographical or product markets” (Chandler 1992: 83). The expansion of corporate boundaries to include new lines of business has occurred not only within large Fortune 500 corporations (Rumelt 1974) but also within many smaller firms (Teece et al. 1994).² Despite the prevalence of multiproduct corporations, however, until recently there has been little theory with which to explain or predict diversification.

In contrast to this theoretical dearth, there has been a great deal of empirical exploration of corporate diversifying behavior. Scholars have placed particular emphasis on (1) uncovering empirical regularities in diversification activity,³ (2) exploring the links between diversification and firm performance,⁴ and (3) constructing indices with which to measure the extent of a firm’s diversification. ⁵ However, perhaps because of the lack of theoretical foundations to guide this research, little progress has been made in the empirical literature on diversification since the landmark studies of the 1960s and 1970s (Ramanujam and Varadarajan 1989).

In the last decade, the recently developed resource-based view of the firm (Wernerfelt 1984; Barney 1986; Teece 1988) has been touted as particularly conducive to understanding corporate diversification. This conducivity notwithstanding, the operationalization of this theory has been limited to broad characterization of resources and the industries to which they might fruitfully be applied. For example, studies have found that firms with high R&D intensities tend to diversify into industries with high R&D intensities. But research into firm-specific “technical competence” (e.g., Pavitt et al. 1989; Patel and Pavitt 1994; Jaffe 1986) suggests that a particular technological ability or knowledge is useful in only a very narrow range of applications. This suggests that the operationalization of resource-based theory might be usefully informed by the narrowly defined technological capabilities proposed in the technological competence literature.

In this study I apply and extend the resource-based view of the firm to explain and predict corporate diversification. My efforts to extend the theory follow three avenues. First, I define and measure technological resources at a more microanalytic level than has been done in prior resource-based research. This enables me to integrate elements of the technological competence and resource-based literatures to shed further light on firms’ diversifying behavior. In particular, I develop a new methodology to link patents to industries in which they are likely to be applicable. Using this methodology, I am able to create a new database that links firms’ patent portfolios – which are indicators (albeit noisy) of corporate technological competence (Patel and Pavitt 1994) – to their output in product markets. This database supports the testing of more finely grained hypotheses concerning corporate diversification than previous empirical research has enjoyed.

Second, by stressing the resource-based framework’s links to transaction cost economics, I attempt to draw out a more complete set of resource-based hypotheses than has been done in the past. Most prior research in this framework has implicitly or explicitly assumed that the resources that provide value to firms are necessarily too asset-specific to allow contracting. Rather than make this assumption, I explicitly develop and test hypotheses concerning the possibility of contractual alternatives to diversification, finding evidence that firms diversify when alternatives are subject to high contracting hazards.

Third, I extend the reach of the resource-based view by deriving and testing hypotheses about aspects of diversification other than its direction and the resulting performance – in particular, about the mode by which such diversification occurs (acquisition or internal expansion).

This study should add to our knowledge about the role of resources in general, and technological resources in particular, as artifacts upon which a firm may build its downstream product portfolio. Resource-driven corporate diversification is a topic that has both scholarly and practical implications. The development of the resource-based view of the firm has led numerous scholars to prescribe that firms should focus on developing “core competences” from which business- and product-level competitive advantage will naturally flow (see, for example, Prahalad and Hamel 1990). However, the nature of these competences and their methods of exploitation remain somewhat mysterious, and rarely obvious except through ex post analysis. An enhanced understanding of how important attributes of technological resources drive a firm’s decision to diversify (or to pursue an alternative mechanism of exploitation) would provide insights into the nature of resource-based rents and their variation across firms and industries. This has both normative and positive implications for theories of competitive strategy, theories of the management of innovation, and theories of the firm.

Context of the study

The present study examines diversification behavior in a sample of US manufacturing firms during the first half of the 1980s. The sample includes 344 firms randomly selected from the 573 US firms with the largest number of patent assignments during the 1970s (see Jaffe 1986 for the entire universe of these firms), and 68 firms randomly selected from the population of all other US firms that appear in the major firm-level databases used in this study (Compustat; Who Owns Whom; and the AGSM/Trinet Large Establishment database). This study relies heavily on technology-intensive firms because the key resources under consideration are technological resources, as measured by patent portfolios.

A multi-industry sample of US firms offers several benefits for the purposes of the present study. First, numerous studies of corporate diversification have examined similar samples of firms (Montgomery and Wernerfelt 1988; Montgomery and Hariharan 1991). Since this study seeks to improve upon prior research, my conducting research on a sample of firms similar to those used before facilitates the comparison of this study’s results with those of its precursors. Second, the focus on multiple industries allows the incorporation of industry- specific variables as well as firm-specific variables to determine how the relationship between technological resources and corporate diversification patterns varies across industries as well as firms. Finally, a practical benefit of such a multi-industry study of US firms is the availability of relatively detailed information about the firms, their activities, and their technological resources. This type of detailed information is essential for a study of this nature. The sources and nature of the data used in this study are discussed below.

Sources of data

The empirical research undertaken in this study relies on several sources of data. The AGSM/Trinet Large Establishment database (Trinet) was compiled for every odd-numbered year between 1979 and 1989.⁶ Trinet includes information on corporate ownership, employment, and four-digit SIC code of operations for more than 400,000 establishments in the United States (see Table 1.1).⁷ For establishments in which multiple four-digit SIC businesses are undertaken, Trinet provides up to three separate SIC codes (with SIC-specific employee breakdowns).⁸

Information on firms’ patent portfolios came from two sources: The Micropatent PatentSearch database and the Canadian Patent Office PATDAT database. The Micropatent PatentSearch database includes every patent granted by the United States Patent and Trademark Office between 1975 and the present. For each patent, this database includes the date on which the patent application was submitted, the firm to which it was assigned (if any), and the technology class to which it was assigned, as well as other information from the first page of the issued patent (see Table 1.2). The PATDAT database includes every patent issued by the Canadian Patent Office between 1978 and 1993.⁹ In addition to providing information on the patent class to which each patent was assigned, it also provides the (Canadian) four-digit SIC industry of use and the (Canadian) four-digit industry of manufacture to which the patent was assigned. Dun and Bradstreet’s Who Owns Whom. North America (1981) was used to link patent-owning subsidiaries with their parents.

Table 1.1 Information available in the AGSM/Trinet database*

Table 1.2 Information available on the front page of US patents* (1)

Overall, these databases represent some of the most comprehensive available sources of information on corporate diversification behavior and on corporate technological resources. Nevertheless, additional sources were tapped to extend and complement this information. The “Yale study” of innovation (Levin et al.

1987) provides information on the effectiveness of various mechanisms for appropriating returns generated by technological innovation (i.e., licensing of patents, reliance on trade secrets). This information is used as a proxy for the feasibility of a firm’s using contractual alternatives to diversification to exploit its technology. The Federal Trade Commission’s Line of Business database (US Federal Trade Commission 1975, 1976), the Census of Manufactures (1982, 1987), and the Annual Survey of Manufactures (1981–1985) all provide information on industry-level variables of interest such as industry concentration, industry growth, and industry average R&D- and advertising-intensity. Finally, the Compustat database (1977–1985) provides information for firm-level variables of interest such as firm size, firm growth, and firm R&D- and advertising-intensity.

Organization of the study

This study is organized as follows. Chapter 2 examines the theoretical and empirical literature relevant to corporate diversification. The review of the various theoretical lenses suggests that there is to date only a partial understanding of the determinants of diversifying behavior. The chapter also briefly describes the technological competence literature, a subset of the resource-based view that – while not directly related to questions of diversification – suggests that existing conceptions of technological resources in the empirical literature on diversification are unsatisfactory, and offers clues as to fruitful ways to improve the operationalization of these technological resources.

Chapter 3 develops a conceptual framework around the theoretical perspectives discussed in Chapter 2: the resource-based view (with a particular emphasis on its transaction cost roots) and the technological competence literature. Several propositions are derived, primarily pertaining to the direction of diversification, the feasibility of contractual alternatives to diversification, and the mode through which diversification is effected.

Chapter 4 reviews several alternate measures of technological resources:
R&D statistics, innovation counts, and patent data. After concluding that patents are in general the best available indicators of technological prowess, and that their chief weakness is the inability to connect them to the businesses in which they offer competitive advantage, the chapter introduces a new methodology to link patents to the industries in which they are likely to be useful. The chapter also demonstrates that this newly developed patent-SIC concordance offers a more accurate linkage than the other concordances available previously.

Chapter 5 tests a set of hypotheses concerning the extent to which a firm’s existing technological resources contour its subsequent diversification activity. These hypotheses are tested with data on 2514 diversification entries and non-entries between 1981 and 1985. Chapter 6 examines the factors affecting the mode of diversification. A set of hypotheses concerning the decision to diversify via internal growth as opposed to acquisition are derived and tested with data on 1023 diversification entries between 1981 and 1985.

Chapter 7 concludes with a summary of the results of this study and a discussion of their implications for future research.

2 Review of prior theoretical and empirical research on diversification

This chapter is divided into four sections. The first provides an overview of theoretical perspectives that bear on the diversification phenomenon. It highlights the contributions that each of several efficiency-based theories – neoclassical economics (economies of scope), transaction cost economics and the resource-based view – have made to our understanding of corporate diversification. The second section reviews existing empirical studies of this phenomenon, and interprets their findings through these t...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Figures
• Tables
• Preface
• Acknowledgments
• 1. Introduction
• 2. Review of Prior Theoretical and Empirical Research On Diversification
• 3. The Effect of Resource Attributes On the Direction, Mode, and Performance of Corporate Diversification
• 4. Patent Data and Construction of a US Patent Class–SIC Concordance
• 5. An Empirical Analysis of the Effect of Technological Resources On the Direction of Corporate Diversification
• 6. An Empirical Analysis of the Effect of Technological Resources On the Mode of Corporate Diversification
• 7. Conclusion
• Appendix 1: Comparison of Scherer and Patent Concordance Assignment of Patents (Top Three Patents from Concordance)
• Appendix 2: Sample Firms
• Notes
• Bibliography