Choice of Strategy

  • Neil M. Kay


We have arrived at the stage where we can begin to tie threads together to build models of corporate behaviour based on the structuralist theory developed in the earlier chapters. Here we shall construct a simple model of firm decision-making which we shall apply in the chapters following in an examination of some problems in industrial organisation. It should be borne in mind throughout that we are looking from the point of view of decision-makers (managerial) rather than owners (shareholders).


Indifference Curve Corporate Strategy Decline Phase Portfolio Theory Survival Constraint 
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.
    See Markowitz (1952), (1959).Google Scholar
  2. 2.
    This is a standard formulation in the literature.Google Scholar
  3. 3.
    This example is adopted from Scherer (1970), pp.101.Google Scholar
  4. 4.
    Dyckman and Stekler show that if r = 0 and expected profit levels are identical for each plant, the coefficient of variation declines by 1/√N with increases in N, where N is number of plants in the combination (1965, p.214–18 and cited in Scherer, 1970, p.101)Google Scholar
  5. 5.
    An example of an exception to this would be the aluminium tennis racquet and ski example discussed earlier; the complementary nature of the respective seasonal cycles reduces profit variability by a greater amount than if the activities were unrelated, while synergy benefits are also obtainable.Google Scholar
  6. 6.
    Effectively this is shifting the risk onto a second party, and depends on the other party’s ability to bear risk, and attitude towards risk.Google Scholar
  7. 7.
    There will be associated costs of stockholding.Google Scholar
  8. 8.
    Strictly speaking, vertical integration could also be considered in single period analysis; it is not considered in portfolio theory because exploitation of its advantages requires managerial control and intervention, and does not simply result from financial combination.Google Scholar
  9. 9.
    See Oi and Hurter (1965) for detailed discussion of some circumstances where these may be appropriate.Google Scholar
  10. 10.
    This, of course, assumes the parallel condition that disclosure of information will not be used opportunistically by the potential funder.Google Scholar
  11. 11.
    Plus any profit the capital market may be able to extract.Google Scholar
  12. 12.
    A further point worth making here is that a standard argument in portfolio theory is that optimal diversification may be achieved if returns are negatively correlated due to the associated effects on risk of the total portfolio. While this may be the case for holding of financial assets, it is less liable to be the case for corporate diversification, since there are two major cases in which negative correlations obtain. Firstly, counter-cyclical activities in the economy. If they are desirable to one firm for risk reduction purposes they will be attractive to many, with drastic effect on profitability of these activities. The process of competition will trade off risk reduction against profitability. Secondly, substitute goods. This is where identification of shared constituents is useful. Negative correlation in the fortunes of two activities may quickly be converted into a positive correlation if a third substitute is developed and creates catastrophe linking. Thus, short run negative correlations may hide serious costs to weigh against such risk reduction.Google Scholar
  13. 13.
    It should be borne in mind that we are dealing with constant risk here (by which we mean a constant coefficient of variation). An additional problem that is generally serious in the early stage of the life cycle is true or unmeasurable uncertainty. For the moment we leave this problem aside but return to it below.Google Scholar
  14. 14.
    We assume that if the constituent innovation (the catastrophe) is adopted by the firm, this effectively creates a new product market.Google Scholar
  15. 15.
    Gort (1966) has also questioned the worth of diversification to solve the inconvenience of uneven earnings and suggests instead that the ‘risks’ really worth diversifying are long-term ‘risks’, one of which is decline in earnings. In this context, he mentions switch to shorter-lived product lines as being important.Google Scholar
  16. 16.
    That is, in principle. In practice, the examples used by Schumpeter are frequently interpretable as constituents (see p.83–4)Google Scholar
  17. 17.
    A more sophisticated analysis would attempt to reconcile the problem of comparability by establishing criteria for ordering of strategies involving different mixes of shared constituents.Google Scholar
  18. 18.
    This is reasonable if the possibility of take-over raiders being attracted by a high level of liquid reserves is recognised. Take-over threat is dealt with by keeping liquidity at a level which may create vulnerability to a second type of threat due to technological change. Hedging is an alternative way of anticipating technological threat.Google Scholar
  19. We are not concerned here with how the firm actually dispenses its excess liquid funds; profits may be a source of founds for sub-objectives such as growth, dividends, etc. The use of the funds may differ according to circumstance.Google Scholar
  20. 19.
    This version of the utility function is the simplest possible given our analysis of potential catastrophe and synergy effects. A more complex version of the utility function was included in Kay (1978) but this encountered difficulties in obtaining determinate solutions.Google Scholar
  21. 20.
    The assumption of constant level of P may be seen to be more reasonable to the extent that firms tend to operate within 2-digit industries, and to the extent that these industries involve a family resemblance between technologies; for example, a food firm would tend to have low catastrophe potential associated with its range of activities, while an electronics firm would have consistently high catastrophe potential for its range of activities. However, later analysis might explore the implications of recognising the possibility of mixing levels of P associated with the bundle of activities operated by the firm.Google Scholar
  22. 21.
    There is a difficulty in that the process of competition may lead to a violation of (δπ*/δr)>0 and (δπ*/δp)>0 due to the changing level of profits along ZZ’. However it may be reasonable to maintain the assumption (δπ*/δr)>0: holding P constant, the distinguishing features of each Srp are not individual activities but rather the combinations of activities. Therefore, for constant P modification of π* for Srp on ZZ’ through the process of competition should have an accompanying effect on other strategies within the relevant industry through the modification of profitability on individual activities that make up individual strategies.Google Scholar
  23. Maintaining (δπ*/δr) > 0 means that utility may always be increased for firms operating to the left of ZZ’ if they increase r (as indicated by the horizontal arrow in Fig. 5.5). This ensures that firms will be distributed along the locus of ZZ’.Google Scholar
  24. 22.
    We would expect Beta-Two to be more profitable if its greater size permits exploitation of economies within divisions.Google Scholar

Copyright information

© Neil M. Kay 1982

Authors and Affiliations

  • Neil M. Kay

There are no affiliations available

Personalised recommendations