Biotech is in a structural decline---only the plan can reverse that.

Studies deploying different methods, measuring innovation through different means and over different time periods, collectively provide consistent and persuasive evidence that the current innovation system is becoming increasingly inefficient and unproductive. While intellectual property and other exclusive rights are critical to continued firm investment in research and development, too much intellectual property actually decreases efficiency through siloing and high transaction costs. Firms retrench to lower-risk projects. The consequence is a declining level of growth due to innovation, rending the argument that governments ought to invest more in innovation precarious. As a result, governments curtail funding, leading university researchers to become increasingly risk-adverse, particularly when facing a peer review system that rewards short term citations over long-term contribution. At some point, as West (2017) speculates, inefficiency leads to more loss of productivity that, in turn, leads to greater declines in efficiency.

Whether the innovation system reaches such a decline in efficiency depends, in part, on whether the actors in the innovation system can curtail cost escalation and increase productivity. Following David (2003), we can fix some of these difficulties by re-establishing an equilibrium between proprietary and open science models of innovation. The last several decades witnessed a slide away from open science toward proprietary models, upsetting a balance that has sustained science for three centuries. The belief that science was only done in the academy and commercialization in industry (Chesbrough, 2019) is not only factually incorrect – the vaccine to fight Ebola was created, developed and trialed solely by the public sector (Herder et al., 2020) and the largest pre-clinical drug deal in Canadian history arose out of an OSP (“The Ontario Institute for Cancer Research and the Structural Genomics Consortium Develop and Give Away New Drug-like Molecule to Help Crowd-Source Cancer Research” n.d.) – but blinds us to advantages of open science partnerships in accelerating commercial development.

The Structural Genomics Consortium transitioned from science to commercial development through novel approaches to intellectual property (Ali-Khan et al., 2018), including using regulatory data protection while fully sharing all innovation artifacts, attracting significant industrial, governmental and philanthropic investment. The consortium's small world network structure positioned it to draw on a diverse set of skills to quickly validate and improve results. As a consequence, industry accepts the consortium's results without the need for further reproducibility studies, a rarity in the pharmaceutical sector.

There remain significant unanswered questions concerning the role of OSPs in overcoming the deficits in the existing innovation system. Many OSPs to date have been in the biomedical/biopharmaceutical field where innovation is complex, large data sets indispensable, and regulatory requirements high. Similar circumstances exist in other technological fields, such as green tech and artificial intelligence, but not in others. More informal open science relationships – outside of a formal partnership – exist in the information technology industry (West, 2003; Levine and Prietula, 2014). Further study of the use of OSPs in different technology fields is required to determine the broader application of these partnerships in overcoming declining productivity.

A second set of questions surrounds the long-term sustainability of OSPs. Without some form of intellectual property, venture capital is not likely to invest in them. Either OSPs will need to engage with some forms of intellectual property, even if not patents, or will need to find other sources of funding. In the biomedical field, OSPs can use regulatory data protection. While shorter in length than patents, they offer superior protection in many ways, including when they begin (at market approval rather than at invention) and the fact that they are difficult to invalidate. Spinouts of the Structural Genomics Consortium rely on this form of protection and have obtained private funding. Not all technological fields benefit from the same variety of intellectual property regimes as does biomedicine. How OSPs will attract funding in these areas remains an open question.

Third, governments interested in encouraging OSPs will almost certainly need to engage in policy reform (Ali-Khan, Jean, and Gold, 2018). This can take the form of advantages offered to OSPs, for example, research grants, regulatory regimes, purchasing and procurement, the provision of a sharing infrastructure, or direct funding. More research is needed to identify policy options and to assess their impact.

As a result of the COVID-19 pandemic, researchers and firms quickly adopted open science to accelerate research and innovation on both characterizing the SARS-CoV-2 virus and developing diagnostics and vaccines. The success of these efforts points to the important contribution that open science can make to productivity. OSPs provide a leading model of how to make this change permanent.