Buyer Market Structure and R&D Effort: A Simultaneous Equations Model

Abstract

THE theoretical and empirical investigations of the relationship between market structure and the firm's incentive to invent and innovate have concentrated on the effects of the structure of the supply side of the market. This focus on the supply side originated in Schumpeter's controversial hypothesis that current market power provided the necessary conditions and future market power provided the incentive for technical change (Schumpeter, 1950). This hypothesis stimulated a vast number of theoretical refinements and empirical tests of relations between research and development activity (R&D) and supply side characteristics such as concentration, firm size, and diversification. This literature has been well summarized by Kamien and Schwartz (1975) and Scherer (1980). The analysis of the effects of the buyer market structure on the selling industry conduct and performance has been limited, both theoretically and empirically. Stigler (1964) modeled the effects of the size distribution of buyers on price competition in oligopoly. Telser (1964) devel-, oped a theory of messages in which he suggested that advertising would increase when buyers were many and small. Lustgarten (1975) tested the effects of buyer market structure on price cost margins and advertising and found the relations suggested by Stigler and Telser. Invention and innovation are forms of competition which, like price and advertising, will be affected by both buyer market structure and seller market structure. However, unlike price and advertising competition, there have not been explicit attempts to establish theoretical and empirical relations between buyer market structure and inventive and innovative activity. The effects of buyer market structure on the selling industry's incentive to invent and innovate is relevant to interpreting the effects of seller market structure on this effort. A number of previous studies of the effects of seller market structure on technical change have found that seller concentration may have a slight positive impact on inventive and innovative effort (Kamien and Schwartz, 1975, p. 20). Even this weak relation may be due to not controlling for buyer market structure. If Galbraith's (1952) countervailing power hypothesis is correct (Lustgarten, 1975, p. 128), concentration in the buyer and seller markets may be positively correlated. If buyer market concentration is positively related to inventive and innovative effort in the selling industry, the empirical estimate of the partial effect of seller concentration will be biased upward. This paper investigates the effect of buyer market structure on R&D effort. Section II draws together the literature that is relevant to the issue of the relation between buyer market structure and seller inventive and innovative activity. Section III develops a simultaneous equation model of R&D activity where R&D, advertising and concentration are endogenous variables. Section IV uses data on scientific and engineering inputs to test this model. Section V summarizes these results.