Research and Productivity

We model research as a signal on an unknown parameter of a technology. We distinguish applied from basic research and show that firms in the same industry can optimally choose different research portfolios, and that basic research can seem to have a higher rate of return than applied research, even though it really doesn't -- essentially, firms on a 'fast track' upgrading policy opt for basic research but fast and slow-track upgrading policies can coexist in a long-run equilibrium. We also derive the lag structure for how R&D affects the firm's stock of knowledge. To a first approximation, the lags decay geometrically (as is typically assumed in practice) but the rate of decay is endogenous, and depends on how fast the firm is upgrading its technology.