Nonlinear capital market payoffs to science-led innovation

This study explores the extent to which cross-sectional differences in firms' propensity to absorb and utilize scientific knowledge matter for their stock market and operating performance. Unlike previous empirical studies and given the coexistence of both benefits and potential costs to science-guided search, I conceptualize innovation as a recombinant search to hypothesize a non-monotonic relationship. Using patent data from the European Patent Office (EPO) relative to 3281 international firms over the 1999-2015 period, empirical tests yield evidence in support of the predictions. However, the effect varies with firm-level absorptive capacity, across industries and countries. As suggested by a limited attention model, I find a stronger payoff predictive power of science for firms with lower investor attention and higher valuation uncertainty. Besides, the association between science and future operating performance increases with time, indicating that firms' scientific capabilities are a potent driver of long-term profitability. The conclusions are robust to controlling for the endogenous nature of firms' use of scientific knowledge. By providing a useful basis upon which to judge the economic merit of firms' innovation endeavors, this study contributes to reducing capital market imperfections that curtail access to external finance and ultimately spurring private-sector investment in R&D. It also provides evidence of diminishing marginal payoffs to science-led search. Finally, the evidence is likely to be relevant to policymakers, who might wish to influence the innovation ecosystem in a way to sustain long-term prosperity.