The paper, by Sasan Bakhtiari and Robert Breunig, uses administrative data from firms in Australia that conduct research and development (R and D) in an examination of how R and D activity of other firms and public institutions affect a firm’s own R and D expenditure.
The authors state
We distinguish between the impact of peers, suppliers and clients. We examine whether geographical proximity and industrial clustering affect R and D spillovers. Overall, we detect positive effects on R and D expenditure from spillovers from peers and clients to firms that are nearby; within 25 or 50 km. R and D expenditure by academia, unlike by government bodies, has a positive influence on a firm’s own R and D expenditure within state boundaries. We fail to find any significant role for industrial clusters in augmenting spillover effects.They comment
Research and development play a central role in long-run productivity and economic growth. Theory suggests that R and D spillovers (where the R and D activity of a firm affects the well-being of consumers or the profitability of other firms) also play an important role in economic growth and that the benefits of R and D extend well beyond the firm that makes the R and D investment. Thus, the social returns to R and D as a whole may be greater than the sum of the private returns to firms who make R and D investment decisions. The existence of R and D spillovers may also increase the incentive for firms to invest in R and D if other firms’ R and D is complementary — that is if it makes a firm’s own R and D more productive. This could happen if spillovers from other firms make a firm’s R and D more likely to succeed or if knowledge from other firms’ R and D combines with an individual firm’s R and D to increase the returns to a firm’s own R and D expenditure.
However, the existence of R and D spillovers also has the potential to disincentivise firms from investing in R and D. The partially public nature of knowledge and competition in markets reduces the firm’s ability to appropriate rents from their innovative activities. This may lead firms to reduce their R and D expenditure. A priori, it is not known whether the positive or negative effects of spillovers on firm-level R and D expenditure will dominate.
Recognising the latter possibility, governments around the world offer a range of incentives such as patents and licenses (that grant a temporary monopoly to the inventor) or R and D grants and subsidies. These measures compensate for the lack of incentive for firms to invest in R and D.
In this paper we focus on the R and D expenditure decisions of individual firms and how they are affected by the R and D activity of other firms. We find that overall, the negative disincentives dominate. The presence of spillovers results in firms making less R and D investment than they otherwise would.
However, both distance and relationship matter. For peers and clients, we find a positive role for proximity. After trying a few discrete radii, we find that spillovers from peers within 25 km, and for clients within 50 km, result in higher R and D expenditure. For suppliers, we find that spillover effects are always negative but less strongly so at greater distances. This gives us insight into which types of spillovers might be most important for which types of relationships, as we discuss below.
We also test for the role of industrial clustering in spillovers. We find that R and D activity is on average higher in industrial clusters but that clustering does not amplify the effect of spillovers on R and D expenditure. For public sources of R and D expenditure, we find that higher education expenditure has a positive influence on firm-level R and D expenditure. Direct government spending on research seems to crowd out private R and D expenditure.
In the next section we elaborate on the conceptual background of our approach. We provide a short literature review in Section 3 and a discussion of our data in Section 4. Our model and methodology are described in Sections 5 and 6. Our results are presented in Section 7 with the geographical refinements presented in 7.2 and the results on industrial clustering discussed in 7.3. We conclude in Section 8.'Connect the Dots: Patents and Interdisciplinarity' by Michal Shur-Ofry in (2017) University of Michigan Journal of Law Reform comments
This article unravels a troubling paradox in the ecosystem of innovation. Interdisciplinarity is widely recognized as a source of valuable innovation and a trigger for technological breakthroughs. Yet, patent law, a principal legal tool for promoting innovation, fails to acknowledge it in an explicit, consistent manner. Moreover, while the scientific understanding of the significance of interdisciplinarity for innovation increasingly relies on big data analyses of patent databases, patent law practically ignores patent data as a source of information about interdisciplinary innovation. This article argues that patent law should connect the dots: explicitly recognize interdisciplinarity as a positive indication in the decision whether an invention deserves patent protection, and use information derived from patent databases to evaluate the interdisciplinarity of inventions. Relying on cutting edge research in economics and network-science, the article explores nuanced manners for implementing these proposals, calling, ultimately, for the development of an algorithmic “recombination metrics” that would allow courts and patent offices to identify interdisciplinary inventions in an accessible, standardized, manner. The adoption of this article’s proposals would align patent doctrine with its ultimate goal of promoting high-risk, socially valuable, innovation; would inject an objective and measurable criterion into various patent doctrines famously criticized for their ambiguity and unpredictability; and would also allow patent law to realize some of the enormous potential of patent data — a treasure that current patent doctrine leaves untapped.
