Introduction

In the previous chapter we considered the effect on national welfare of a minor firm's exogenous technical progress. However, technical progress usually occurs as a result of R&D investment. In this chapter we extend the analysis of chapter 1 by considering endogenous R&D investment by Cournot duopolists with initial cost differentials, and examine the structure of optimum R&D tax-cum-subsidies.

We conduct our analysis by developing a two-stage game of duopoly. In the first stage both firms decide on their cost-reducing R&D investments, and in the second they compete in a quantity-setting game. Much has been written on such two-stage models (see, for example, Bagwell and Staiger, 1994; Besley and Suzumura, 1992; Brander and Spencer, 1983; d'Aspremont and Jacquemin, 1988; Katz, 1986; Okuno-Fujiwara and Suzumura, 1993; Petit and SannaRandaccio, 2000; Rowthorn, 1992; Spence, 1984; Spencer and Brander, 1983; Suzumura, 1992; Varian, 1995). However, most of the authors work with models of symmetric oligopoly. Only Spencer and Brander (1983) consider asymmetry in marginal-cost levels. In their model rival firms which belong to two different countries compete only in a third country and the strategic use of government policies is at the heart of their analysis. They ignore the effect of R&D subsidies on consumers' surplus in the third country, and focus on only the international distribution of profits.

Although the literature on endogenous R&D is fairly large, very few authors analyse the question of R&D subsidies.