Economists take a keen interest in patent rights and their effect on innovation. The primary argument for the existence of patents is, after all, that they incentivise entrepreneurs to seek profit through innovating. This column looks at how patent rights affect innovation by small and large firms, finding that the results vary greatly depending on size.
Modern macroeconomic models emphasise that innovation is central to economic growth (Aghion and Howitt 1992). The consensus among economists is that there is underinvestment in research and development, with the social returns of innovation estimated as more than twice as large as the private returns (Bloom et al. 2013). This is the primary justification for government support of research and development investments, and the patent system is one of the key policy instruments for this purpose.
The overall impact of patent rights on research and development investments depends on two questions: how they affect the innovation incentives for the patent holder, and how they affect follow-on innovation by other firms. On one hand, patents are expected to promote research and development investment by enhancing the ability of the patent holder to capture rents from their innovations, either by directly commercialising the innovation or by licensing it. On the other hand, the cumulative nature of the innovation process implies that patent rights on upstream technologies may affect the incentives to invest in downstream (follow-on) innovations, and potentially could block such innovation (Green and Scotchmer 1995). These dangers have been prominently voiced in public debates on patent policy in the US (Federal Trade Commission 2011) and recent decisions by the Supreme Court (e.g. Association for Molecular Pathology v. Myriad Genetics, 133 S. Ct. 2107, 2013). Similar concerns have also been raised in European policy discussions (European Commission 2011).
A growing empirical literature has focused on the impact of patent rights on follow-on innovation (e.g. Murray and Stern 2007, Williams 2013, Sampat and Williams 2015, Galasso and Schankerman 2015a). Overall, these studies show that patents have a negative effect on subsequent innovation by other firms, but only in very particular contexts where bargaining frictions in technology licensing transactions appear to be more severe. Perhaps surprisingly, the academic literature has largely overlooked the impact of patents on the subsequent innovation by patent owners.
Removal of patent rights through judicial decisions
In recent research (Galasso and Schankerman 2015b), we investigate how patent rights affect the rate, and direction, of innovation activity by different types of patent owners across a range of technology fields. The challenge in studying this relationship empirically is to identify comparable innovations with and without patent protection. We exploit patent invalidation decisions by the US Court of Appeal for the Federal Circuit, a specialised court with exclusive jurisdiction in patent appellate cases. We use comprehensive data on 1,469 Federal Circuit decisions from 1983 to 2010, and record whether each patent was invalidated. About 40% of the decisions in our sample lead to a loss of patent rights over the innovation.
Our empirical work is guided by a simple theoretical framework that shows how the loss of patent rights can affect the incentives to innovate. The basic mechanism is as follows. A firm is assumed to build on its patents in subsequent rounds of innovation. When a patent is invalidated, the firm still retains the knowledge embodied in that patent for future use, of course. However, the loss of patent protection opens up the innovation competition to other firms now able to exploit this knowledge without a license. The resulting competition for the second generation patent reduces the incentives and thus the research and development level of the original owner.
To estimate the effect of patent rights on innovation, we compare the subsequent number of patent applications by firms (or individuals) owning patents that are invalidated to those by firms whose litigated patents are upheld by the Federal Circuit Court, in a five-year window following the decision. A fundamental challenge with this approach is that invalidated patents may differ from those that are upheld in a variety of dimensions which may also affect innovation investments by the patentee. It is crucial to address this ‘endogeneity’ issue in order to estimate the true causal impact of patent rights on innovation incentives.
Our empirical strategy exploits the fact that judges are assigned to patent cases through a computer programme that randomly generates three-judge panels, with decisions governed by majority rule. The variation in the propensity to invalidate patents among randomly allocated judges allows us to identify the causal impact of patent rights. Essentially, we compare patenting behaviour by firms whose patents are invalidated because they were randomly assigned to judge panels with high propensity to invalidate, with patenting by similar firms whose patents were not invalidated because they were randomly assigned to judge panels with low propensity to invalidate. In conducting this analysis, we control for a number of confounding factors related to the patent, the technology field and the patentee.
The patents litigated in the Federal Circuit are not representative of the overall population of patents. They are typically higher value patents, as they have gone through the costly litigation process up to the appellate level. However, for the purposes of studying how patents affect innovation incentives, it is reasonable to start by analysing Federal Circuit patents because the distribution of patent values is highly skewed (Schankerman and Pakes 1986) and the incentives generated by these patents are likely to be more important for welfare.
Large and small firms
We present three main, robust empirical findings:
- First, Federal Circuit invalidation causes, on average, a 50% decrease in future patenting (in a five-year window) by the focal patentee.
- Second, the effect of patent protection depends critically on the size of the firm, the centrality of the technology for the firm’s research strategy and the competitive landscape.
The analysis shows that the effect is entirely driven by firms with small patent portfolios losing patents on technologies that are core to their research focus. There is no significant response by small firms that lose a non-core patent (one in a technology field where the firm does not concentrate its activity). Even more strikingly, the evidence shows that large firms (with more than 75 patents in their portfolio) do not reduce their level of innovation when they lose either a core or a non-core patent.
- Finally, we show that patent rights affect the direction of innovation by large firms, even though there is no significant impact on their level of innovation. The invalidation of a non-core patent induces large firms to increase their patenting in different (though related) technology fields.
Losing patent rights over core technologies does not induce any refocusing of innovation for large firms.
These findings can be potentially explained by three main channels through which patent rights affect innovation incentives for small firms:
- By strengthening their market position in competing with large incumbent firms in the product market and by strengthening their ability to license their innovations to large firms for commercialisation;
- By facilitating access to financing and venture capital; and
- By enhancing their bargaining power in negotiations for patented inputs needed for their research and development activity.
The evidence supports only the first hypothesis – we find that patent invalidation has a large impact on small firm innovation only in technology fields where they face many large firms. In contrast, our analysis does not support the capital markets or licensing channels. We show that the effect of losing a patent is no stronger for patents that have been pledged as loan collateral, which contradicts the capital market hypothesis, and the impact of invalidation is no larger in technology fields where patent ownership is highly fragmented, as would be predicted by the licensing negotiation hypothesis.
‘One size does not fit all’ – toward a more nuanced patent policy
Our findings complement those in our earlier study of how patent rights affect follow-on innovation by firms other than the patentee (Galasso and Schankerman 2015a). The earlier study shows that Federal Circuit invalidation causes an increase in follow-on innovation (as measured by patent citations to the focal patent) by external innovators, on average, but the impact is highly heterogeneous with the strongest effect is in technology fields with high likelihood of bargaining failure in licensing. Moreover, the effect is entirely driven by cases in which invalidation of patents owned by large firms triggers more follow-on innovation by small firms.
Taken together, these two studies show that patent rights affect innovation by small and large firms very differently. If we are willing to interpret judicial invalidation as equivalent to a marginal reduction in the strength of patent rights for firms, these findings would imply that reducing the strength/scope of large firms’ patents is likely to encourage follow-on research by small firms and unlikely to reduce innovation incentives for the large firms. In contrast, weakening patent rights held by small firms diminishes their innovation incentives without spurring additional patenting by large firms. This interpretation is consistent with the recent macroeconomic research by Acemoglu et al. (2013), who show that fiscal stimulus policies for research and development by large incumbents are less effective than support targeted at small innovative firms. However, we emphasise that our findings do not imply that complete removal of patent rights of large firms would improve innovation incentives or welfare. In the presence of patent rights, research is conducted under the expectation of rents from the product market and licensing to follow-on innovators. These rents would be expected to (largely) disappear in a regime without patents and this would reduce, perhaps sharply, incentives to conduct such research and development.
The law and economics literature has discussed ways to design the patent system to make it more effective in recognising that the heterogeneous role of patent rights for different types of innovators and technology fields. Possible policy instruments include the use of patent application fees, patent renewal fees, and the application of legal rules such as injunctive relief and presumption of validity by courts. However, there are serious practical challenges in implementing such policies effectively. Our results suggest that more research on these approaches is warranted.
Acemoglu, D, U Akcigit, N Bloom and W R Kerr (2013), “Innovation, Reallocation and Growth,” NBER Working Paper 18993
Aghion, P and P Howitt (1992), “A Model of Growth through Creative Destruction,” Econometrica, 60: 323-351.
Bloom, N, M Schankerman, and J Van Reenen (2013), “Identifying Technology Spillovers and Product Market Rivalry,” Econometrica 81: 1347—1393.
European Commission (2011), “Implementing enhanced cooperation in the area of the creation of unitary patent protection”, Commission Staff Working Paper.
Federal Trade Commission (2011), The Evolving IP Marketplace: Aligning Patent Notice and Remedies with Competition, Washington D.C.: Government Printing Office.
Galasso, A and M Schankerman (2015a), “Patents and Cumulative Innovation: Causal Evidence from the Courts”, Quarterly Journal of Economics 130: 317-369.
Galasso, A and M Schankerman (2015b), “Patent Rights and Innovation by Small and Large Firms,” NBER working paper 21769.
Green, J and S Scotchmer (1995), “On the Division of Profit in Sequential Innovation,” RAND Journal of Economics 26: 20-33.
Murray, F and S Stern (2007), “Do formal intellectual property rights hinder the free flow of scientific knowledge? An empirical test of the anti-commons hypothesis”, Journal of Economic Behavior and Organization 63: 648-687.
Sampat, B, and H Williams (2015), “How do patents affect follow-on innovation? Evidence from the human genome”, NBER Working Paper 21666.
Schankerman, M and A Pakes (1986), “Estimates of the Value of Patent Rights in European Countries during the Post-1950 Period,” Economic Journal 96: 1052-1076.
Williams, H (2013), “Intellectual Property Rights and Innovation: Evidence from the Human Genome,” Journal of Political Economy 121: 1-27.Tags: IP innovation economic model, IP rights, Patent rights