The Triple Helix Deficit and Australia’s Business R&D Problem

John H Howard, 7 July 2026

This Insight has been produced as part of a broader program of work at UTS on how ‘middle power’ economies can deploy artificial intelligence in the form of ‘industrial AI’ to grow productivity in manufacturing ecosystems.[1]

A theme running through the project findings is that productivity gains depend on mature innovation ecosystems more than on the technology alone, and that those ecosystems increasingly operate as networks of place-based hubs rather than as one national system.

Within each hub, the relationship between universities, industry and government, the interplay the Triple Helix describes, does much to determine whether value is created. That relationship, and its weakness in Australia, is what concerns me here.

Over 44 days of site visits across Europe, the United Kingdom and Scandinavia, one feature stood out in every productive innovation ecosystem. Industry, universities and government work together as a matter of habit. The collaboration is genuine, and in practice the institutional boundaries blur.

In Sweden, the Netherlands, Denmark and Germany, the three institutional pillars intertwine; in Australia they run more in parallel, rarely meeting and often wary of one another. This Innovation Insight asks what that difference might mean for business research and development, and what could be done about it.

Australia’s recent Strategic Examination of Research and Development (SERD) found a system once among the world’s strongest now broken and needing bold reform, with productivity growth at its weakest in decades and manufacturing and R&D intensity among the lowest in the OECD (Green, 2026).

The absence of a genuine Triple Helix mindset may help explain why business R&D in Australia remains stubbornly low. It may also help explain why global research-intensive firms anchor their advanced work elsewhere. And the remedy, if there is one, lies less in new programs than in patient investment in trust.

The Triple Helix and Its Institutional Foundations

The Triple Helix model, developed by Henry Etzkowitz and Loet Leydesdorff in the mid-1990s, describes innovation as the product of overlapping interactions between universities, industry and government (Etzkowitz & Leydesdorff, 2000). Each sphere keeps its primary identity while taking on roles once held by the others. The dynamism comes from the overlaps.

Universities become entrepreneurial, engaging in economic development and venture creation. Firms take on training and knowledge production. Government acts as a public entrepreneur, sharing risk and convening partners. Innovation emerges in the hybrid spaces between them, in joint institutes, competence centres, shared facilities and trilateral networks where the three pillars genuinely meet.

What the overlap produces is movement. People circulate between the spheres, carrying tacit knowledge with them. Researchers found companies, industry scientists teach, and public officials move into and out of both. Spin-offs and start-ups grow within reach of their parent laboratories. The ecosystem behaves as a single connected system rather than three separate ones.

Much of the early empirical work drew on Nordic experience, and Sweden became an early exemplar. The collaborative habit there predates innovation policy. It traces to the post-war social compact, and to the 1938 Saltsjöbaden Agreement between labour and employers, which built durable habits of negotiation and shared problem-solving (Etzkowitz, 2008). Those habits carried into the innovation system.

Germany rests on a comparable foundation. The Fraunhofer institutes provide a dense applied-research layer between universities and the Mittelstand. The dual vocational training system ties firms and educators together from the start. The Netherlands operates its consensus-based polder model and a top-sectors approach that convenes industry, knowledge institutes and government around shared agendas.

These are cultures of collaboration built over decades, supported by intermediary institutions that give the pillars a place to meet. Fraunhofer in Germany, RISE in Sweden and TNO in the Netherlands occupy the middle ground between discovery and commercial application. They translate, broker and reduce risk. Australia has nothing operating at comparable scale or with comparable permanence.

Australia’s Parallel Pillars

The Australian picture is one of separation and often distrust. A presentation to the Australian Innovation Research Group in 2019 documented the mutual suspicion (Howard, 2019). Business people described university technology transfer offices as ineffective, academics as unrealistic in commercial dealings, and incentives as skewed towards publication rather than results.

Universities returned the scepticism. Business could not be trusted because of the profit motive and focus on shareholder value. Firms could also compromise academic independence. Companies took a “K-Mart approach” to acquiring knowledge, treating it as a commodity to be bought cheaply rather than co-created through partnership.

Both pillars directed similar complaints at government. Public funding for research had declined. Innovation had lost its lustre. Government operated with a funding mindset rather than an investment mindset, churning out many small, short-term, competitive programs that built an industry of grant writers without building durable capability.

There is also a mismatch of interests. Australian business research concentrates in information and computing, engineering and health services, while university research concentrates in medical and health, biological sciences and engineering. National research output as a whole skews heavily towards medical research and psychology, the fields where public money flows. The overlap that should anchor collaboration is narrow.

This is the opposite of the Triple Helix. The pillars do not overlap; they guard their boundaries. Where trust is the currency of collaboration, Australia runs a deficit. As the 2019 analysis put it, people do business with people they trust, and trust requires sustained investment in social capital. That investment has been thin and episodic.

Can the Triple Helix Deficit Explain Low Business R&D?

The problem is long-standing. Writing in The Australian in November 2019, Green and Howard identified weak commercialisation, falling R&D expenditure and a mismatch between funded university research and the needs of future industries (Green & Howard, 2019). They reported R&D spending at 1.79 per cent of GDP against an OECD average of 2.4 per cent. The pattern has since deepened.

Australian business expenditure on research and development sat at 0.9 per cent of GDP in 2023-24, unchanged since 2017-18, having fallen from a peak near 1.37 per cent around 2008-09 (ABS, 2024; Department of Industry, Science and Resources, 2025). Gross R&D intensity is about 1.7 per cent, ranking Australia twenty-second in the OECD and below the OECD average of 2.7 per cent (OECD, 2025).

The contrast with the Triple Helix economies is sharp. Business R&D alone reaches roughly 2.6 per cent of GDP in Sweden, 2.1 per cent in Germany and 1.9 per cent in Denmark (Eurostat, 2025). Australian business invests less than half what its Nordic and German counterparts commit, and the gap has widened as those economies grew their commitment while Australia stalled.

Several structural factors clearly matter, and one runs deeper than the rest. The economies that perform strongly on business R&D tend to host large, research-intensive manufacturers that spend heavily on their own account, firms of the Bosch, Siemens and Ericsson kind.

Australia has few of them. Its economy weights heavily towards mining, agriculture, construction and domestically oriented services, sectors that perform less formal R&D, and many of its larger firms are subsidiaries of multinationals that locate research at headquarters.

Although the SERD made much the same point (2026), it did not make any reference to the triple helix or to innovation ecosystems in its analysis and recommendations. Seen this way, a strong Triple Helix could be seen as much of a consequence of that industrial structure than a cause of it. The deficit is one factor among several, and the evidence is associational rather than conclusive.

The mechanism, though, is plausible. R&D is most valuable to a firm when it can draw on complementary assets: accessible university research, a deep talent pool, trusted partners and government co-investment. These assets are complements, and the value of each rises when the others are present. Where the pillars run in parallel, those complements are hard to reach, and the private return to in-house R&D falls.

Absorptive capacity reinforces the effect. A firm needs internal research capability to recognise, value and apply knowledge generated outside it. Low business R&D both reflects and deepens weak absorptive capacity, which in turn lowers the value a firm places on external collaboration. The parallel-pillars structure and thin in-house R&D feed one another in a self-limiting loop.

This connects to the migration of value. Value tends to move towards ecosystems where complementarity is strong and away from those where it is weak. A firm weighing where to grow its research capability favours a location where it can plug into a dense, trusted network. Australia’s parallel pillars make it a less attractive place to anchor that capability, so activity, and the value it generates, settles elsewhere.

Policy choices have compounded the problem. The long decline in public R&D support since its peak around 2011-12 removed much of the co-investment that signals commitment and draws private money alongside it. When government steps back, the complementary public contribution weakens, and the case for private R&D weakens with it. The deficit is partly self-inflicted and partly the legacy of an innovation system in which the three pillars have rarely worked closely together.

Why Global Research-Intensive Firms Anchor Elsewhere

The locational evidence supports the argument. Global firms do conduct research in Australia, but the pattern is revealing. Multinational pharmaceutical companies collaborate actively in health and medical research, where the public health system, patient access and clinical trial infrastructure provide a strong anchor (Howard, 2019). The complementary assets exist, and firms use them.

In engineering, technology, advanced manufacturing, agriculture and mining, collaboration with global firms is far thinner. These are the domains where deep university-industry partnership drives frontier R&D in Sweden, Germany and the Netherlands, and where Australia’s thin intermediary layer and weaker collaborative habit leave little for firms to anchor to.

The European cases make the point. ASML, the leading supplier of advanced lithography equipment, sits at the centre of the Brainport ecosystem around Eindhoven, embedded in a web of suppliers, knowledge institutes and public support. Novo Nordisk in Denmark, and the engineering clusters of southern Germany, show the same anchoring effect. In each case the firm and the surrounding ecosystem have grown together.

A research-intensive firm chooses where to anchor advanced work for many reasons, including talent, tax treatment, intellectual property regimes and proximity to markets. The strength of the local Triple Helix is one input, and it may matter most precisely for the decision to anchor and grow capability rather than for marginal, short-lived projects. Australia competes adequately for some research and poorly for the kind that compounds.

Australia leans heavily on the R&D Tax Incentive to lift business R&D. A generous subsidy can raise the volume of qualifying expenditure, but it does little to build the trusted relationships and shared institutions that anchor frontier work. Lowering the price of R&D is a weaker draw than a place where the pillars already work together.

What a Remedy Would Require

A Triple Helix mindset cannot be legislated quickly. The Nordic compact took generations to build. The remedy lies in patient, coordinated investment across all three pillars, and in the intermediary institutions that let them meet. The responsibilities differ by pillar, and each could begin now.

Government

The central shift is from a funding mindset to an investment mindset, meaning fewer, larger and more durable commitments in place of a churn of small competitive programs. It means building intermediary institutions at scale, modelled on Fraunhofer, RISE and TNO, to occupy the translation layer between discovery and commercial application.

Vehicles such as the National Reconstruction Fund and the Future Made in Australia agenda could anchor this layer if designed for patient capital and genuine co-investment. Mission-oriented procurement, drawing on Mazzucato (2021), could pull private R&D through demand as well as pushing it through grants.

Universities

The university system is too uniform. All thirty-nine public universities look broadly alike, which is suboptimal for engagement (Howard, 2019). A more diverse system, with some institutions specialising as technology-facing and industry-facing partners, would give firms clearer points of contact. Reform of intellectual property terms, contracting and academic incentives would reward engagement rather than penalise it. Above all, universities would treat the building of trusted, long-term relationships as core mission rather than peripheral activity.

Industry

Firms carry responsibility too. Absorptive capacity, the internal capability to recognise, value and apply external knowledge, depends on doing some research in-house. The cycle of low business R&D and weak collaboration has to be broken from inside the firm. That means committing to long-term partnerships rather than transactional procurement, and investing in the relationships and shared facilities where genuine collaboration takes shape. Engagement might also move firms up the ladder from buying knowledge to conferring with and working alongside research partners.

Conclusion

Australia’s low business R&D has structural origins in industry composition and firm ownership, above all the scarcity of large research-intensive firms, that policy cannot quickly change. The Triple Helix deficit sits alongside those structural factors, both reflecting and deepening them. Where the pillars run in parallel, the complements that make private R&D worthwhile are hard to reach, and value migrates to ecosystems where trust and collaboration are dense.

The European and Nordic evidence suggests the deficit is real and consequential, though it might be one cause among several rather than the whole story. The encouraging implication is that the foundation of a stronger system is trust, and trust can be built. It would take patience, durable institutions and a shared sense of national purpose, sustained well beyond any single electoral cycle.

References

Australian Bureau of Statistics. (2024). Research and experimental development, businesses, Australia, 2023-24 financial year. ABS. https://www.abs.gov.au/statistics/industry/technology-and-innovation/research-and-experimental-development-businesses-australia/latest-release

Department of Industry, Science and Resources. (2025). Research and development expenditure. Australian Government. https://www.industry.gov.au/publications/australian-innovation-statistics/research-and-development-expenditure

Etzkowitz, H. (2008). The Triple Helix: University-industry-government innovation in action. Routledge.

Etzkowitz, H., & Leydesdorff, L. (2000). The dynamics of innovation: From national systems and “Mode 2” to a Triple Helix of university-industry-government relations. Research Policy, 29(2), 109-123.

Eurostat. (2025). R&D expenditure statistics. European Commission. https://ec.europa.eu/eurostat/statistics-explained/index.php?title=R%26D_expenditure

Green, R. (2026, March 18). Australia was once a world leader in innovation. A new report shows the system is now “broken”. The Conversation. https://theconversation.com/australia-was-once-a-world-leader-in-innovation-a-new-report-shows-the-system-is-now-broken-274012

Green, R., & Howard, J. H. (2019, November 6). Tech wave: We’re missing the boat. The Australian, p. 32.

Howard, J. H. (2019). Lessons from university-business engagement in Australia: Making collaboration more effective [Presentation]. AIRG Roundtable, Sydney.

Howard, J. H. (2026). Pillars in parallel: How to create genuine collaboration to achieve innovation ecosystem outcomes. Acton Institute for Innovation. https://www.actoninstitute.au/post/pillars-in-parallel-how-to-create-genuine-collaboration-to-achieve-innovation-ecosystem-outcomes

Mazzucato, M. (2021). Mission economy: A moonshot guide to changing capitalism. Allen Lane.

OECD. (2025). Main science and technology indicators. Organisation for Economic Co-operation and Development. https://www.oecd.org/sti/msti.htm

Strategic Examination of Research and Development. (2026). Ambitious Australia: Strategic examination of research and development, final report. Department of Industry, Science and Resources. https://www.industry.gov.au/publications/ambitious-australia-strategic-examination-research-and-development-final-report

[1] The project, Turning AI into Productivity: The Role of Innovation Ecosystems, is being undertaken by Emeritus Professor Roy Green AM and Dr John H. Howard, Executive Director of the Acton Institute for Policy Research and Innovation and Honorary Visiting Professor at UTS.