The experience curve and limit pricing as means of integrating portfolio matrices into capital budgeting.

  • Paul S. Marshall

Student thesis: Doctoral ThesisPhD


This thesis develops a model for the net present value to the firm of a strategy of increasing an SBU's market share as a function of four quantitative variables: the firm's initial market share, the current position in the product life cycle, as viewed by the firm, the competitor's relative experience curve slope and the competitor's view of the shape and size of the product life cycle. The net present value is calculated using traditional definitions, but includes a limit pricing strategy by the firm and assumes the existence of the experience curve. The prime reason for the development of this model is to test quantitatively the capital budgeting implications of the portfolio matrices proposed by the Boston Consulting Group and McKinsey and Company. And, to suggest a format that better integrates corporate strategy and finance. The most important finding of this research is that the inclusion of experience curve effects causes relative market share to assume dominance over market growth rate in the BCG matrix and causes business strength to assume dominance over industry attractiveness in the McKinsey matrix, at least within the limitations and assumptions of the model. Put into laymen's terms, that means that corporate planners should abandon attempts to convert low share but high growth businesses (what BCG calls "Problem Children") into "Stars", if such conversion can only be accomplished .through price, or price equivalent competition, as long as both participants are equally competent. A second important finding is that other variables, beyond growth and share, can be successfully and quantitatively incorporated into the model. This means that the two-dimensional approach of the Boston Consulting Group can be improved upon by adding additional variables or that businessmen can make better use of the McKinsey approach to strategy and its investment implications, by logically quantifying their variables.
Date of Award1985
Original languageEnglish
Awarding Institution
  • University of Bath

Cite this