Decoupling Research from Development

Reporting research and development together implies their allocation provides limited insight to investors. We construct and corroborate unique measures of development and research to evaluate this aggregation. These measures combined correlate with reported R&D expenditures at 99% (industry) and 81% (firm). In the individual measure validation tests, our research measure correlates with scientific publications (77%), while the development measure correlates with patents (71%). Our R&D measures cover 76% of NYSE firms, providing broader coverage than patents (30%), new product announcements (20%), and reported expenditures (46%). Using these measures, we investigate whether a firm’s choice to emphasize either research or development provides relevant information to investors. We find that development and research differ significantly in predicting future earnings and cash flows. The allocation between research and development also helps predict whether a firm discloses its R&D expenditures. Development intensity, but not research intensity, forecasts increased product market concentration. Investors require higher risk premia for research relative to development activity. Overall, our results are inconsistent with the hypothesis that development and research load similarly in investor-prediction models. It is puzzling that firms do not explicitly disaggregate R&D disclosures to give investors insight into their allocation decision