Diversification Strategy

  • Neil M. Kay


This chapter will base its examination of aspects of the diversification problem on existing analyses. We shall start by using data from a number of existing studies in looking at patterns of diversification in relation to technological change. Then we shall look briefly at the extent of R & D diversification in different US industries. The next section is concerned with relations between industrial inventive activity and diversification, and the third section is concerned with examination of the portfolio risk theory of diversification; in both cases puzzles thrown up by empirical studies are discussed in the light of the approach developed earlier.


Technical Skill Inventive Activity Industry Level Median Strategy Portfolio Theory 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Notes and References

  1. 1.
    The ratios for each company were calculated from quantitative and qualitative information compiled from a variety of sources; registers, annual reports, prospectuses, books and articles.Google Scholar
  2. 2.
    Rumelt’s assignment of links between activities was based on subjective evaluation of whether or not significant relationships existed between activites.Google Scholar
  3. 3.
    This is more reasonable in the unrelated case than might appear at first sight, since 2-digit industry classification is generally broad enough to cover a variety of businesses having sufficiently weak links to merit the description ‘unrelated’.Google Scholar
  4. 4.
    ‘Dominant-vertical firms are vertically integrated (vertical ratios of .7 or more) that produce and sell a variety of end products so that the specialisation ratio is less than .95’ (Rumelt, 1974, pp.31–2). Of the 27 dominant business firms, only one paper and one furniture firm were not classified as dominant-vertical. (Rumelt, 1974, p.98).Google Scholar
  5. 5.
    The single mining-metals firm is excluded from this analysis since R&D figures were not obtained from the available data.Google Scholar
  6. 6.
    Indivisibilities occurring at different levels of output in the respective production stages may require an extremely large final output if potential economies of scale are to be fully exploited.Google Scholar
  7. 7.
    Also, the dominant-vertical firms had the lowest average return on capital employed of any strategic category (Rumelt, 1974, p.92) Thus, profitability is low and the firms generally do not have a high level of internally generated funds for diversification purposes.Google Scholar
  8. 8.
    Petroleum is an obvious exception.Google Scholar
  9. 9.
    For details of Xerox’s phenomenal pattern of growth, see Rumelt, 1974, pp.97, 223–4. Xerox operates in a highly technological environment, but was a dominant product firm in 1969.Google Scholar
  10. 10.
    In distinguishing between science-based and non-science-based industries, the following industries were classed as science-based in each case; aerospace, chemicals, pharmaceuticals, engineering (light, heavy and electrical). The following industries were treated as non-science-based; food, drink, tobacco, petroleum, metals, materials and minerals, vehicles (excluding aircraft), textiles and clothing, glass, shipbuilding, paper and packaging printing and publishing.Google Scholar
  11. 11.
    Gort comments that the result means that ‘diversification … depends upon … technical skills’ (1962, p. 135). Amey bases his use of the T-ratio on the fact that one reason commonly cited for the apparent growth of diversification has been the tremendous increase in the amount of organised research carried on within firms (1964, p.267). Wolf argues that the T-ratio is ‘assumed to be a reasonable proxy for the firms accumulated useful stock of technical knowledge’ (1977, p. 181). Hassid states that the ‘existence of highly developed skills … makes easier the adjustment of already familiar production processes to the requirements of other industries’ (1975, p.389). Gorecki argues that R&D activities are liable to throw up opportunities, many of which ‘have characteristics which weigh the scales in favour of their exploitation by diversification rather than sales’ (1975, p. 138).Google Scholar
  12. 12.
    Sutton (1973) also argues from a behavioural standpoint that R&D staff may dominate decision-making in high technology industries. This might be expected to promote diversification as a result of R & D staff empire building and successfully pursuing their own objectives. Thus, internal R & D stimulates diversification in this view also. See Sutton (1974) and Grant (1974) for further discussion.Google Scholar
  13. 13.
    The other variables were firm sales, after tax profits and patents per scientist and engineer.Google Scholar
  14. 14.
    That firms do ‘fine-tune’ their strategies within industries but that industry effects can dominate and swamp these firm level effects, was argued for the firms R&D decision in Kay (1979, Chapter 8). Grabowski’s analysis discussed here provided useful evidence in this analysis of the determinants of inventive activity in the firm. It also provides an alternative explanation of why the petroleum industry in Grabowki’s analysis did not record a significant correlation between diversification and inventive activity (Kay, 1979, pp. 197–203).Google Scholar

Copyright information

© Neil M. Kay 1982

Authors and Affiliations

  • Neil M. Kay

There are no affiliations available

Personalised recommendations