The Sensitivity of Strategic and Corrective R&D Policy in Oligopolistic Industries

We evaluate the sensitivity of the case for an R&D subsidy in an export sector when the outcome of R&D is uncertain and when the resulting product market is oligopolistic. Investments in R&D are assumed to induce either first order or mean-preserving second order shifts in the distribution of a firm's costs, with firms then competing in either prices or quantities in the product market. When R&D reduces the mean of a firm's cost distribution in the particular sense of first order stochastic dominance, we find using standard models of product market competition that a national strategic basis for R&D subsidies exists, whether firms choose prices or quantities. This national strategic incentive to subsidize R&D must be balanced against the national corrective incentive to tax R&D that arises whenever the number of domestic firms exceeds one. However, when R & D preserves the mean but alters the riskiness of a firm's cost distribution in the sense of second order stochastic dominance, we find that the national strategic basis for R&D intervention completely disappears, while the national corrective incentive is now to subsidize R&D whenever the number of domestic firms exceeds one. We conclude that the crucial determinant of appropriate R&D policy is the nature of the R&D process itself.