Despite increases in business expenditure on R&D in recent years, Australia’s gross expenditure on R&D (GERD) remains low by OECD standards (1.77 per cent compared to the OECD average of 2.25 per cent in 2004).

However, as is often pointed out, this measure can be misleading as a nation’s industry structure can significantly skew the relevance of international comparisons. Some industry sectors such as computing and pharmaceuticals typically have high levels of R&D intensity but others, including agriculture and mining but particularly services, tend to have low R&D intensity.

In its recent study Public Support for Science and Innovation, the Productivity Commission argued that when industry structure is taken into account, Australia’s R&D intensity is much closer to the OECD average. Moreover, the existing structure of the Australian economy (comparatively high proportion of services, mining and agriculture and low proportion of advanced, high-technology manufacturing) was such that even if every industry sector in Australia did more R&D than the OECD average for that sector, Australia’s R&D intensity would be closer to, but still lower than, the OECD average.

This raises important national questions. What happens when the mining boom is over: does Australia have the right industry structure to ensure future prosperity? If not, how do we change it? Industry structure does change – as the rise of the services sector and the decline of manufacturing as a share of GDP in the past few decades show. But should industry structure be left entirely to the dynamics of global markets or is there a role for governments to influence structural change through procurement, public sector R&D or other targeted programs?

The idea that R&D alone can drive structural change seems pretty naïve, as global markets, price relativities and cultural factors around risk and entrepreneurship are likely to be far more important factors. Nevertheless, I think there are important strategic reasons to think deeply about what are the pathways from now to the future and how public R&D contributes to that.

Governments invest in R&D for a number of reasons. Perhaps the most important is that governments need to know about the problems, risks and opportunities
in areas they have responsibility for – including defence, public health and environmental management. They also need to invest to ensure sufficient capacity exists to ensure ‘preparedness’ to manage risks in an uncertain world. Governments also invest in to produce ‘spillovers’ to the private sector from basic and/or long-term strategic research, high-quality graduates, diffusion of ideas, access to high-quality research infrastructure and international linkages.

According to ABS R&D data, in 2004-05, $6.8 billion was spent on R&D in the public sector, including universities and Commonwealth and state agencies and departments. Of that, 37 per cent (about $2.5 billion) was classified as economic development with the balance being defence (five per cent); society, including health and education (33 per cent); environment (12 per cent); and non-oriented research (13 per cent).

Of the public investment in economic development, 34.5 per cent was spent on agriculture, 11 per cent on mining and 19.9 per cent on manufacturing in 2004-05. At first glance, this appears to be quite asymmetrical, given agriculture was 3.3 per cent of total industry gross value-add to the economy, mining 5.6 per cent and manufacturing 11.7 per cent in 2004/5.

Obviously you would not expect a linear relationship between structure and share of public sector R&D directed at economic development, given different R&D intensities. While there are very clear positive benefits from agricultural research, its prominence is also related to historical factors, the development of strong research teams and dissemination mechanisms, institutional inertia and significant political factors such as marginal regional seats. However, it is an interesting question to ask whether the distribution is related to past, current or future industry structure.

It is worth noting that even though R&D projects and programs tend to have longer horizons than political or commercial time frames (for example, three to five years for discovery grants, multiples of seven years for CRCs and 10 to 20+ years for some mission-oriented programs in CSIRO or major research groups in universities), there was quite marked reallocation of resources from just 2000-01 to 2004-05.

Research classified as ‘economic development’ declined from 42 to 37 per cent of total expenditure while ‘society’ increased from 28 to 33 per cent. In real terms, that translates as economic development remaining static at about $2.5 billion and society increasing by a little more than 40 per cent, from $1.6 billion to $2.3 billion.

However, aggregating public sector data masks important shifts, so while economic development remained static overall, government-sector R&D declined by $300 million in real terms and universities increased by roughly the same amount.

Curiously, these relative shifts away from economic development to health, society and environment coincided with the increasing focus on commercialisation of public sector R&D in Backing Australia’s Ability.

Significant change is also evident in the relativities between broad fields of research.

Between 1996-97 and 2004-05, gross expenditure on R&D in Australia increased in real terms by 42 per cent. However, the increase in expenditure in science (13.5 per cent) was well below health and medical research (82 per cent) and humanities, arts and social sciences (50 per cent). Moreover, science was the only broad field of R&D in Australia where gross expenditure declined as a percentage of GDP during this period.

It is not at all clear that changes in the profile of Australian research occurred by deliberate policy per se. While increases in health and medical research are deliberate policy choices, other changes appear to be unintended, including, for example, the consequences of the expansion and changing profile of the higher education system. The point here is that it is a good idea that there be informed public consideration of what social, environmental and economic outcomes we want from R&D, which in turn will guide investment patterns.

That brings me back to the relationship between industry structure and public R&D. It is probable that mining, energy and agriculture will remain very important features of the Australian economy over the next 25 years, although there are likely to be many different questions and issues than being thought of now. In agriculture for instance, we are likely to need R&D that deals with responses to climate change and water issues ranging from moving agriculture to northern Australia to different pest and disease vectors.

In my view, one of the weaknesses in our current policy environment is that we do not have a well-developed sense of where we might be in 25 years time. What are the possible problems? What are the geopolitical prospects? Are we building the right skills to be thinking about creating futures? Are we doing the right sort of use-oriented, basic research that will inform industry in the decades ahead?

The Productivity Commission report firmly established the rationale for public support of science and innovation and the fact that we get good value for our investment. That is, it discussed the why question.

I believe that this time of economic prosperity, but environmental and social uncertainty, is the time for a broad public discussion of the what, where, who and how much questions.

Professor Tom Spurling is the President of the Federation of Scientific and Technological Societies 2005-07 . He is a Professor at the Australian Centre for Emerging Technologies and Society at Swinburne University of Technology and Chief Executive Officer, CRC Wood Innovations. He is a former Chief of the CSIRO divisions of Molecular Science and Chemicals and Polymers.