The Rise in Private Sector Research: Adapting to a New Reality

DOI: https://doi.org/10.25453/plabs.33925117

Published on September 18th, 2026

Introduction

Private sector actors now perform by far the majority of Research and Development (R&D) activities in OECD economies, with US, Japanese and South Korean business R&D all exceeding 75 percent of total R&D since the early 2020s and with an even higher figure for China of 79.3%. While European business R&D remains below 60% of national R&D[1], here too over the last five to ten years, a consistently larger amount of total R&D was performed by businesses, illustrating the increased concentration of research in the enterprise sector.

The consequences of the shift towards private knowledge production extend well beyond changes in the funding share to the structural conditions under which scientific knowledge is produced, circulated, and used. The entry of private actors at scale into frontier scientific work, operating under disclosure codes that diverge sharply from the open-publication norms of public science, is reshaping the communication system of science in ways that existing public policy frameworks have yet to register, with consequences that are international in scope and not self-correcting through national policy alone.

Our argument proceeds in three steps.

1.    A “privatization” of research…

Drawing on the National Systems of Innovation tradition, the present configuration fits a gradual weakening of the institutional matching between the public and private knowledge components of advanced innovation systems: a dual crowding-out[2] in which the old, large incumbent firms under international competitive pressure, retreated in the previous century from privately funded fundamental research while universities simultaneously pulled toward academic basic research and away from more applied, market-driven work.

The current configuration extends this dynamic. Private R&D growth has been concentrated in sectors characterized by high research intensity and low publication intensity. Drawing on the sectoral typology of Camerani, Rotolo, and Grassano (2018)[3] and on the evidence presented in Arora, Belenzon, and Patacconi’s (2018)[4] of a long-run decline in corporate basic research, it can be shown that an increasing share of frontier scientific work does no longer enter the public knowledge commons, while corporate publishing’s share of total scientific output is itself contracting.

The traditional mechanisms by which industry-generated knowledge entered the cumulative scientific record--co-publication, conference participation, the “admission ticket” of disclosure required to access academic networks as highlighted in Nate Rosenberg’s seminal paper “Why do firms do basic research (with their own money)?”[5] have weakened or, in the dominant frontier sectors of computing and AI, inverted: in these sectors public science increasingly depends on infrastructure – including data as highlighted in Mons, Baak and Jonkers contribution[6] – controlled by private firms rather than the reverse.

2.    With inverted disclosure incentives…

At first sight such “privatization” trend might well imply a positive interpretation of what this configuration means for the productivity of science as a whole. The aggregate productivity slowdown reported in the “ideas are getting harder to find” literature[7] is, on our reading, better understood as a pathology of the scientific communication system than as exhaustion of the underlying field of inquiry. Following Leydesdorff’s framing of science as a self-organising system of communication[8], we argue that the entry of actors operating under proprietary disclosure codes is reshaping the system’s signaling structure and weakening the spillover and spin-off mechanisms on which cumulative science depends.

The diffusion-of-innovations framework developed in the cybernetic tradition of communication studies[9] is suggestive here: the channels through which corporate-generated ideas have historically diffused into the wider research community – change agents, opinion leaders and the open-publication record – are structurally less available where the disclosure incentives that sustained them have inverted. In contemporary configurations, these channels have become structurally constrained as disclosure incentives have inverted, creating asymmetric dependencies in the sectors where new knowledge is most consequential.

Private sector research is disclosed through a constellation of open and semi-open practices that do not fit neatly into traditional publication channels. Public repositories (for example, GitHub), notebooks and open data flows enable researchers to critique, reproduce and adapt work, yet access to core inputs such as data, models, pipelines remain guarded by licensing, data governance and platform constraints.

The result is that openness might look real in observable form but limited in usable substance - a regime in which surface openness coexists with strategic protections, making diffusion uneven across sectors and regions and challenging the applicability of traditional symmetry frameworks in the most pivotal contexts. This configuration also undermines the traditional triple-helix framing of university-industry-government symmetry, as dependencies have become asymmetric in the sectors where new knowledge matters most.

3.    Requiring public policies preserving autonomy

Two complementary theoretical traditions form the basis for the governance argument.

First, from Michael Polanyi’s defense of scientific autonomy in his ‘Republic of Science’[10], we take the position that the normative concern is not that science should be redirected toward social ends – a mission-orientation logic – but that the conditions under which science can follow its own internal logic are being eroded by private sector concentration, and that governance is required to restore those conditions, not to impose new ones.

Second, from Ostrom’s analysis of commons governance[11], we identify three dimensions of enclosure--knowledge, infrastructure, and human capital--each generating a distinct class of public goods failure.

The policy implications are straightforward.

They call for international action on a number of fronts: protecting and restoring the knowledge commons through disclosure conditions on publicly funded or subsidized private research; governing infrastructure concentration through shared and federated computational facilities and non-discriminatory access provisions; rebalancing agenda formation so that research priorities reflect global needs and knowledge frontiers and not only the strategic interests of leading private actors; building the data and policy infrastructure for scientific workforce mobility, including the global talent mapping that the present absence of harmonized data renders impossible; and reforming evaluation and recognition systems to capture the actual mix of contemporary scientific outputs, including software, datasets, and computational artefacts.

Conclusions

The science community is increasingly confronted with a new reality characterized by corporate dominance in research and development.

The historical record of paradigm transitions in the Schumpeterian sense – each requiring institutional constructions that appeared politically impossible during the early installation phase – provides nevertheless grounds for cautious optimism about the direction of travel, if not the pace.

Given the indeterminacy of the present configuration, we do not wish to prescribe fixed solutions but rather equip decision-makers with the analytical tools – typological, theoretical, and diagnostic – to navigate a configuration that will not stabilize on its own.

References

[1] See https://ec.europa.eu/eurostat/statistics-explained/index.php?title=R%26D_expenditure#R&D_expenditure_by_sector_of_performance

[2] In earlier publications this was described as a “Dutch knowledge disease”, see Luc Soete (2005), “On the dynamics of innovation policy: A Dutch perspective,” in Peter de Gijsel & Hans Schenk (Eds.), Multidisciplinary Economics. Dordrecht, (Springer), 2005. https://link.springer.com/article/10.1007/s10842-007-0019-5

[3] Roberto Camerani, Daniele Rotolo & Nicola Grassano (2018), Do firms publish? A multi-sectoral analysis, European Commission, 2018, JRC113261. https://publications.jrc.ec.europa.eu/repository/handle/JRC113261

[4] Ashish Arora, Sharon Belenzon & Andrea Patacconi, 2018. "The decline of science in corporate R&D," Strategic Management Journal, Wiley Blackwell, vol. 39(1), pages 3-32, January. DOI: 10.1002/smj.2693  

[5] Nathan Rosenberg (1990),“Why do firms do basic research (with their own money)?”, Research Policy, Volume 19, Issue 2, 1990, Pages 165-174, ISSN 0048-7333, https://doi.org/10.1016/0048-7333(90)90046-9

[6] See Frontiers Policy Lab’s commentaries’ piece of Barend Mons, Arie Baak & Koen Jonkers (2026), The Impact of data and AI on Scientific Production, https://doi.org/10.25453/plabs.32973416 

[7] Such as Bloom, Nicholas, Charles I. Jones, John Van Reenen & Michael Webb. (2020). "Are Ideas Getting Harder to Find?" American Economic Review 110 (4): 1104–44.  DOI: 10.1257/aer.20180338  

[8] See Caroline S. Wagner (2025), International science collaboration as a complex adaptive system in the work of Loet Leydesdorff. Scientometrics 130, 3183–3193 (2025). https://doi.org/10.1007/s11192-025-05309-y

[9] Such as in Everett M. Rogers (2003) Diffusion of Innovations. Free Press, New York.

[10] Michael Polanyi (1962). The Republic of Science. Minerva 1, 54–73. https://doi.org/10.1007/BF01101453

[11] Elinor Ostrom (1990), Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge University Press. https://doi.org/10.1017/CBO9781316423936

Disclaimer: The contents of this publication do not necessarily reflect the position or opinion of the European Commission.

Copyright: © 2026 [author(s)]. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in Frontiers Policy Labs is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

Please join the conversation and share your perspective.