Government-sponsored Research and Technical Standards: Evidence from Standard-essential Patents

Today, we live in an increasingly interconnected world. For such a complex system to work effectively, we heavily rely on crucial technologies that not only allow individuals to exchange information with one another but also allow for communication between technolo- gies (Bekkers and Martinelli, 2012; Bekkers et al., 2020a). This type of communication is made possible by a set of common rules voluntarily adopted by producers to ensure in- teroperability that goes by the name of technical standards. Formal decisions about the design and evolution of a technical standard are often taken in the framework of Standard Setting Organizations (SSOs). SSO membership is voluntary, and SSOs’ activities could be seen as a self-governance effort made by private firms (Simcoe, 2012). In the last 15 years, economics and management scholars devoted growing attention to the role of SSOs in the standardization process and in determining the success of a technical standard (Lerner and Tirole, 2006; Chiao et al., 2007; Baron et al., 2014, among others).
In most cases, these works consider technical standards as the outcome of a purely private and market-driven process. However, several factors suggest that the contribution of gov- ernmental agencies to standards development might be underestimated. First, even though their direct involvement within SSOs is limited, governmental bodies may actively partici- pate in SSOs. Second, and more importantly, public agencies may support the development of technologies that end up in technical standards. It is well-known that the technical foun- dations of the modern Internet were laid down by the creation of the ARPANET and the adoption of the packet-switching technology by the US Advanced Research Project Agency in the 1970s (Ruttan, 2006). Mazzucato (2013) reports that cellular communication tech- nology received enormous government support in its early days. Indeed, the research grants provided by the US National Science Foundation (NSF) greatly contributed to the develop- ment of key technologies such as spectrum auctions, spectrum sharing, and massive MIMO antennas.1 Even Qualcomm, one of the most influential contributors to mobile telecom- munication standards such as CDMA, in its early days, benefited from several contracts awarded by the US Department of Defense (DoD) and National Science Foundation (NSF) in the context of the Small Business Innovation Research (SBIR) program.
Yet, the contribution of government-sp\onsored research to scientific discoveries on which technical standards build is scarcely acknowledged and studied in the economic and man- agement literature. This lack of attention is quite striking and might have far-reaching implications. A failure in tracking and quantifying the actual impact of publicly funded research may lead to underestimating the value of such investments and, thus, insufficient public support for basic and applied research. This issue is especially concerning in an era in which the private sector is reducing its investment in basic science, while the relevance of fundamental research for private innovation is not declining (Arora et al., 2018).
The main objective of this project is to fill the literature gap and shed light on the potential relevance of the link between publicly-funded research and technical standards. To address this challenge empirically, we mainly rely on the abundant information provided in patent data and, in particular, on about 19,000 patents disclosed as potentially standard- essential (declared SEP) to the European Telecommunications Standards Institute (ETSI). A growing number of studies have focused on SEPs and showed that they appear to have a particular economic and technological value as measured by conventional patent metrics, such as citation, claim counts, and renewal (Rysman and Simcoe, 2008; Bekkers et al., 2017, 2020a,b). The main question we aim to answer in this paper is whether this group of patents disproportionately relies on publicly funded science compared to a set of similar inventions never disclosed as SEPs.
Interestingly, working with patent data also allows us to follow the potential trail be- tween public funding and technology development. In recent years, several research teams systematically collected large databases of government-funded corporate patents (Rai and Sampat, 2012; Li et al., 2017; De Rassenfosse et al., 2019; Fleming et al., 2019; Argente et al., 2020; de Rassenfosse and Raiteri, 2019). However, most of these works look for the govern- ment’s direct involvement in developing an invention, and only a few devoted attention to the indirect links between patents, science, and public funding. To do so, they exploit the references to the non-patent literature (NPL) available in patent documents. Accordingly, a patent is deemed as building on publicly-funded research if it cites, as relevant prior art, a scientific article that reports the support of a government award (grant or procurement contract) in its acknowledgment section. In this work, we adopt this approach and classify a declared SEP as linked to public funding based on its references to the NPL.
Clearly, to assess whether SEPs disproportionately rely on public science, we need a suitable reference point, i.e., a group of patented inventions that are technically similar to our focal SEPs but were not disclosed as potentially essential to ETSI. To identify such inventions, we mainly rely on text similarity between patent documents, as done in de Rassenfosse et al. (2020), and select about 27,000 potential control patents.
Our main results show that, on average, disclosed SEPs that were applied during the development of the 2G and the early development of the 3G standards are 10 percent more likely to cite a scientific article as relevant prior art and about 14 percent more likely to build on publicly funded research compared to suitable controls. When we consider patents applied from 2003 onward, we find no significant difference in the average share of patents building on scientific articles supported by public agencies between treated and control patents.