HIGH FLYERS THINK TANK
National Research Priorities Strategic Forum
The Shine Dome, Canberra, 26-27 June 2002
Coordinating the roles of the federal and state governments: Current performance and future prospects
by Mark Matthews
Mark Matthews has recently been engaged as a Policy Advisor to the Australian Academy of Science. He is the founder and director of the Australian public policy research firm Policy Intelligence. He spent the last year working on science and innovation policy in the UK with SQW Ltd. His previous academic career spanned the Universities of Sussex (Science Policy Research Unit), Bath (School of Management) and Warwick (Warwick Manufacturing Group). He has carried out a wide range of science and innovation policy studies for Australian Government departments and for Whitehall. His current research focus is on comparing innovation and risk management in the public and private sectors.
There should perhaps be a question mark at the end of 'Coordinating the roles'. What I want to try and do in this talk is to set the scene for a discussion of this important area.
We are talking about national research priorities, and as we all know it is a complex R&D system. What I want to try and do is paint a picture of the division of labour in the federal-state involvement in the performing and funding of R&D, in order to emphasise that division of labour and put that discussion in context.
There is really an issue of examining how much more efficient and effective – we are talking about budget neutral ideas here, in the priority setting process – we can get our existing national R&D effort. I think most of you would agree that improving the degree of coordination between the federal government and state governments would be an important part of that. By the way, just to commend to you the submission made by the Institution of Engineers: there was a very strong emphasis, in their submission on the priority setting process, on bringing the states into the picture, which they felt was underrated.
Well, as many of you know but it is worth rehearsing, the states are a significant funder and performer of R&D, and this division of responsibilities that exists has really got to consider what the best natural roles for each tier of government are. I will just run through some overview data in terms of painting this picture.
The most recent comprehensive figures for R&D key sector of performance are for 1998, and that is itself an issue in terms of whether we have got the right data and policy tools for dealing with more prioritised funding of R&D. I will be returning to that later on. Basically what you see is a picture of sector of performance – business, Commonwealth government, state government, higher education and private non-profit, with the sources of funding highlighted. The key thing is to look for the reds, which is the state government, and compare that with the Commonwealth government sector. This is the intramurine expenditure, what is spent in organisations associated with those domains.
What is particularly striking is the relative importance of the state government vis-à-vis the Commonwealth government. As we all know, the Commonwealth government funds the bulk of the R&D that goes on in higher education, but when one thinks about research organisations, there is a tremendous amount of activity going on in state government organisations.
We can break that down and look at what that consists of. I mentioned the notion of division of labour, and there is a clear division of labour there. This is emphasising the R&D performed in state government versus Commonwealth government organisations. We see a very strong profile, not surprisingly, in agricultural sciences, the medical and health sciences, and the biological sciences.
Just to put that into context, throwing in higher education there does the calibration job and we see that in medical and health sciences the higher education effort is a lot stronger. But still in agricultural sciences the states feature very, very heavily.
So that is looking at the data in terms of field of research. By the way, these are all just standard published Australian Bureau of Statistics figures. Nothing fancy has been done with them.
Another way of looking at the R&D effort is by socio-economic objectives. These are the intended end uses, and the definition explicitly talks about the national benefits. Again looking to the red we see that plant and animal production and primary products are major emphases, health and the environment. Now, in the context of any priority setting exercise, you think about the dryland salinity problem, acidification, areas like that, environmental issues, agricultural research issues, the ageing population, health – these are all areas where likely priorities are going to encounter strong state involvement. So just to highlight the issue that we are talking about national priorities: getting effective coordination mechanisms is going to be really very important.
Just to draw out, because I think it is useful context information: although there are wide asymmetries between the states – this graph shows the aggregate R&D expenditure in each state, by sector of performance – when one normalises that for the population, apart from the anomalous case of the ACT, which is basically in ANU and CSIRO, we are looking at per capita R&D levels that do not vary nearly as much as the variations by state. Michael, in his lunch time address, mentioned a per capita figure in a this-is-not-much-to-pay context.
That per capita figure on a purchasing power $US basis is about 40 per cent of what is spent at a national level, which is the final column – 40 per cent of the United States and about 90 per cent in the United Kingdom. It was about 60 per cent of Canada. That is just to give you some kind of feel for what that means.
Now to move to how one might put into practice better coordination there: the policy context that I think is useful to think about this is in terms of linkage, capacity building. The ARC explicitly talks about linkage programs now. That is a feature of many OECD countries. There is an effort to fund partnership building, cooperative research, in the various national research systems. What seems to be going on there, although it is perhaps not as explicitly articulated as it might be, is this idea of what you might call launch aid. There is a learning curve involved in debugging and getting used to working with people in different sectors, in different universities, in different organisations. The idea is that if one can fund the process of familiarisation and debugging, moving along that learning curve, the process in the end should become self-sustaining.
Why is that an important feature of the policy framework in many OECD countries? Well, it is about better leverage of your existing R&D effort, and equally importantly it is about reduced duplication. By leverage we are talking about putting the right assets, tangible and intangible research assets, if you like – human capital – together in better bundles that cross sectoral boundaries. The cooperative research program is a very good example of that. The reduced duplication issue, as we all know, is really going to be a major issue, increasingly, in the coordination of state and federal government activities. I would suggest that one way of thinking about the best way to exploit potential synergies between Commonwealth and state R&D funding is to think about the leverage, the notion of improving the efficiency and the effectiveness of the national R&D effort.
So what might we do to improve the coordination of state and federal government support? There is what you might call the synchrotron factor: there is perhaps a role in research infrastructure funding. Some people refer to the pork-barrel process, particularly in the United States, where the physical location of major public sector investments is something that is a hard-won political battle. It is usually couched in a rather negative sense, but a lot of people argue that it works rather well in the United States. And it works well because when there are clear location advantages in a public sector investment, it is in the interest of the states, of the regional blocks, to bid and make sure that they are fully apprised of the benefits of attracting that investment to that region, that location, that state, and that they put up adequate funding.
Major research facilities is obviously one area where the location does matter, because of significant lumps of money, and as the synchrotron factor suggests, there might well be a growing role for states in research infrastructure funding, particularly when it comes to the commercialisation of research, because there are clear spin-off economic development advantages there that the system is going to be able to grapple with rather well.
The second issue is how we might actually achieve better coordination over SET priorities. Michael's suggestion at lunch time was to upgrade the Commonwealth, State and Territory Advisory Council on Innovation, CSTACI to Ministerial level. If that is to be done, the question is what kind of practical objectives for CSTACI could be defined that would really improve the coordination so that this division of labour between the states and the Commonwealth that I have tried to map out actually works – that we do not, in particular, get a lot of duplication. In the UK we have regional development authorities now, and every regional development authority has its biotech corridor and its aerospace cluster strategies, and there is a lot of, basically, the same strategic vision rolled out. There is not a division of labour working there.
What would this upgraded body do? One suggestion – this is ending on a bit of a dry note but I think it is quite important – is to think about the tools and methods we have at the moment for budget setting. We have the federal Science and Technology Budget Statement, which gives a very thorough account of portfolio, therefore agency, and program based priorities and gives a map of the overall allocation of funding. As I said earlier, when it comes to the R&D statistics, the detailed picture that I have been able to present – and that is just the top level – comes several years later. I mean, it is 2002 now and the data for 2000 is only appearing. So it is very difficult to capture the actual out-turns. If one is moving towards more prioritised elements of research funding, one might have more rapid feedback loops and, perhaps a bit like the Americans, injecting a bit more detail into the budget setting process – in the United States, the Office of Management and Budget conducts the hard dialogue with the different agencies that perform R&D and as part of that will say, 'Well, how much are you going to spend on basic research, how much are you going to spend on applied research, how much on experimental development, how much on infrastructure?' and degrees those as part of what you might call structural priorities, delivering upon the missions that they have been tasked with.
One of the tasks would be perhaps to make sure we have got the right tools for monitoring the resources that we are allocating, because if we are going to close the loop and measure outcomes, we might find that we do not have the right data to do that, and also, particularly importantly in the context of this session, to align those methodologies between the federal government and state governments. Now, as a pragmatic objective that would mean that we would be able to put together a comprehensive picture, along the lines that I have painted in terms of out-turns – what has happened ex post – on the ex ante phase. Well, if you look at the states and the Commonwealth, look at their budgetary plans – and that will be discussed as part of this Forum – this is what we would be looking at overall, which would give scope for negotiation and tweaking that.
So that does take us into the area of structural priorities – is our investment in research infrastructure adequate? I have mentioned this notion of being compliant with OECD classification guidelines. There is a set international way of measuring R&D. It doesn't include what happens after you do the experimental development, which is usually a bit of a hole in investment. If that were covered as well, while remaining compliant with the OECD, we would have a much richer framework for helping to negotiate and coordinate public sector investment priorities for science and technology, and work out, even if it is a three-year review process, 'Well, we are actually spending this.' What the Americans find is a very big gap between what people say they will spend and what is actually spent, and that becomes something that the National Science Foundation concerns itself with understanding in a policy context. So we would set up a dialogue which would give us feedback loops for monitoring what we are actually spending as against what we say we are going to spend, and how well we are doing.
