To Hunt Or to Scavenge: Optimal Investment Strategies in the Presence of Indirect Network Effects

Abstract

What determines optimal R&D investment in a market with indirect network effects? We analyze this question in a hardware-software framework, where software firms strategically invest in quality upgrades. We find that a firm's optimal investment depends predominantly on (1) its quality level relative to its competitors on the same hardware and on (2) the quality level of software firms on the same hardware relative to other hardware platforms. Using a dynamic model, we examine the effect of initial quality differences within and across platforms on firms' investment behavior. We show that small quality differences across platforms stimulate investment across firms on the same platform, regardless of their current quality level. However, large quality differences across platforms affect competition within platforms such that a firm's quality level relative to competitors on the same platform may increase or reduce its incentives to invest. In this case, responses to own and cross quality upgrades are determined by the overall market structure. The two different responses - increased or reduced investment, give rise to a taxonomy of optimal investment strategies. Since responses depend on market structure, we can then map a firm's position within the market into its optimal investment strategy.