Some questions for British research policy

This piece is based on a summing-up I did at a meeting in London this March: A New Mandate? Research Policy in the 21st Century.

There seem to be two lurking worries that concern people in science policy in the UK at the moment. The first is the worry that, having built a case for state support of science on the basis that this will lead to innovation and economic growth, that innovation and economic growth may not be delivered. The second is that the scientific enterprise doesn’t have a sufficiently broad base of popular support. In short, are we suffering from an innovation deficit, and does our research effort have a democratic deficit?

An innovation deficit

The letter with the funding settlement from BIS to the Research Councils called for “even more impact” – the impact agenda in research councils and funding agencies really is accompanied by a sense of increased urgency of an argument that is by no means settled.

To many scientists the economic case for supporting science may seem self-evident, but the solid evidence in support of this is surprisingly slippery. There is certainly the feeling in some quarters – and not just the Guardian’s Simon Jenkins – that the economic impact of science has been oversold. The Royal Society’s “The Scientific Century” document was a serious attempt to assemble the evidence. What strikes me, though, is that it doesn’t make a great deal of sense to try and give an answer to the primary question – to what extent should the state support science – without considering the much broader question of how our political and economic system is set up to support innovation.

And it is in relation to innovation that there are some more general worries, both at a global level and in our own national circumstances:

  • Is the rate of innovation actually slowing – leaving aside the special case of information technology, have the easiest gains from new technology already been made? I discussed this in an earlier post Accelerating Change or Innovation Stagnation?
  • Is our UK innovation system broken? In the UK postwar settlement, universities were only one of a number of kinds of places where research – especially more applied research – was carried out. Major conglomerates like ICI and GEC had large corporate laboratories, there were major government laboratories associated with organisations like the Atomic Energy Authority, and the military supported laboratories like RSRE Malvern which combined quite basic research with more strategic research and development. In the post-Thatcher climate of privatisation, deregulation and the drive to “unlock shareholder value” most of these alternative research organisations have disappeared.
  • In their place, we see a new emphasis on the development of protectable intellectual property in Universities with a view to creating venture-capital backed spin-out companies. This gives rise to two questions – how effective is this as a mechanism for technology transfer, and does the new emphasis on protectable IP have any deleterious effects on innovation itself? Certainly, the experience of nano- and bio- technology does point to potential problems of patent thickets and an “anti-commons” effect in academia, where pre-existing IP positions inhibit other scientists from working in particular areas. It’s these worries, among other factors, that have driven a move to a more open-source approach, now spreading from IT to new areas like synthetic biology.
  • For the UK, the pharmaceutical industry has been particularly important, as an industry of genuinely international stature which has been politically very important in making the case for state-supported science (and influencing the shape of that support). So the fact that this industry is having innovation difficulties of its own – the closure of the Pfizer R&D site at Sandwich being a very visible signal of this – is worrying.
  • We’re seeing the introduction of a new kind of institution into the innovation landscape – the Technology and Innovation Centres. There’s still uncertainty about their role and some governance issues are still unclear, but what’s most significant is that there is a widely perceived gap that they are intended to fill.
  • A democratic deficit

    The idea that we’re in the midst of a popular crisis in trust in science is deeply embedded. I’m not convinced that the crisis in trust is with science itself, rather than the use of science in politics and commerce, which is something slightly different, but nonetheless this idea has been a driving force for much of the new enthusiasm for public engagement and dialogue, and for taking this public engagement upstream. While some people (including me) would want to set this move as part of a broader move to steer technology to meet widely shared societal goals, there is still a sense that for many, this is still seen as being about gaining acceptance for new technologies.

    On the face of it, these two worries – of an innovation deficit and of a democratic deficit – look to be in opposition. The idea of an innovation deficit suggests that our problem is that technology isn’t moving fast enough, and we have to work to remove obstacles in the way of innovation, while the negative perception of public engagement holds that its job is to put those obstacles back in the way. In fact, in times like now this perception is a real danger.

    But actually they’re quite closely connected. Underneath these dilemmas are two worries – a loss of confidence in the self-organising capability of the scientific enterprise, and a sense that something’s missing in our innovation system.

    Research councils – “from funder to sponsor”

    It’s these worries that underly current moves in the UK research councils, perhaps most explicitly defined by EPSRC, in their aim of “moving from funder to a sponsor” – i.e. moving from the position of responding to the agenda of the scientific community, towards commissioning research in support of national needs.

    The issues then are, how is national need defined, and how is the process of defining that national need given legitimacy?

    This is a big problem in our current system, where our political fashion is explicitly not to define such a need in anything other than rather general and vacuous terms (like saying we need to have a “knowledge economy”). To pose the question in its most pointed form, does it make sense to have a science policy if you don’t have an industrial policy?

    This situation puts research councils in a very difficult position. If governments are not prepared to develop such an industrial policy, how can the research councils do this – how can they do it practically, and how can their decisions acquire legitimacy?

    These legitimacy problems come in three directions:
    1. with the scientific community
    2. with the government
    3. with the population at large.

    The scientific community will see a potential clash with the Haldane principle (invented tradition though David Edgerton says this is), which could be interpreted as saying that the scientific community is the primary source, as an embodiment of the principle of autonomy of the scientific enterprise.

    With the government, a research council like EPSRC is in a very difficult position. They have to deliver the science in support of a national policy which does not, in fact, exist, but they will be judged by very instrumental measures of wealth creation.

    Can “challenge-led” research help?

    “Societal challenges” offer a new synthesis that can be considered a response to this. I find this attractive as a way of getting beyond a sterile dichotomy between applied and basic research, but the definitions of what might be meant by a societal challenge are contested, value-laden and full of interpretive flexibility.

    Societal challenges do have an advantage, in having a certain security in the face of political uncertainty and lack of direction, and a certain independence from political whims. Who can really disagree with the idea that sustainable energy will be a big deal on rather long timescales, for example?

    But there are problems – can governments genuinely take a long enough view? How can we avoid fads and the herd mentality? How can we be prepared for the inevitable unanticipated changes in direction in world events? how can we move from generalities to the particularities of real technologies?

    What is the place of public engagement? On the one hand, what better way of getting a direct view about what national need should be than consulting the public directly? Public engagement then presents itself as a partial solution to the problem of legitimacy, but one that isn’t necessarily going to make their relationship with government any easier.

    There is one other set of institutions that, strangely, don’t get mentioned very often. Those are the Universities. What’s their role? Can they be more than just a loose coalition of individual researchers responding to the incentives and demands of the research councils and other funders? Universities have their own considerable intellectual resources across the disciplines, and they have their own long history and independence, so one might hope that Universities themselves could be another focus for reasserting the public value of research. For a civic university like my own, Sheffield, surely the University should as a focus for the aspirations of the community it serves.

    Science and politics

    There is another driving force for public engagement; the sense that representative government is failing to provide a space for discussing big issues about our future choices and how people want to live their lives. Science and technology have to be a part of this discussion, and this is why discussions about science and technology must have a political dimension. There are those who assert the opposite – that science doesn’t have or shouldn’t have a political dimension, and that technology is autonomous, out of control, and can’t be directed. But these assertions are themselves profoundly political statements.

    7 Responses to “Some questions for British research policy”

    1. Innovation deficit? I think not. It seems to me that we have far too many academics who aren’t effective innovators and we are also starving our high-tech SMEs of funding for essential legal protection.

      We need to scale down the UK university system so that it provides really effective, high-level education and with the expectation that only 10% of the remaining academics will undertake (internationally leading) research. Of this, we may see only a tiny fraction leading to any commercial returns.

    2. Stephen Moss says:

      As a researcher in the biomedical sciences it is interesting to see that much of what you have written holds true in my own field. One distinction perhaps is that the MRC and BBSRC still fund largely according to the Haldane principle, albeit that both organisations set out strategic or priority areas for research (as does the Wellcome Trust). Nevertheless, any investigator with a little cunning can mould their proposal to meet the necessary fit.

      But it is the so-called ‘impact agenda’ that imposes the greatest strictures on innovation in biomedicine. On one hand, it is right that any grant writer should have to think about and comment on the impact their work may have, particularly when using public funds. But it is the tightly linked expectation that research must deliver economic benefits, implicit in the utterings of Willetts and Cable, that puts investigators in a straightjacket. Much of the most profitable research (in terms of return to a national economy) can take years to deliver. I can give two brief examples in an area with which I am familiar, therapeutic monoclonal antibodies that block new blood vessel growth, and thus have value in the treatment of cancer.

      The first concerns a small protein named VEGF that stimulates blood vessel growth. VEGF was discovered as the result of blue skies research, and the antibody that blocks its activity was patented some ten years before entering clinical use. Last year that antibody had worldwide sales of >$6bn, with a derivative used to treat blindness adding a further $3bn. Another more recently developed antibody against a different target, was patented in 2004, licensed in 2008 and has yet to go to market. However, the licensing deal was worth €500M in milestone payments, and if it goes to market double digit royalties will follow.

      What these examples illustrate is that if innovation in the UK is to yield similar returns, which is clearly what government and tax payer would like, the impact agenda becomes meaningless. The fundamental, basic, blue-skies research that primed these massive money-spinning drugs, would almost certainly not have been supported under the constraints imposed by the demands of short-term impact. A second point, related to your question of whether the UK is politically and economically receptive to innovation, is that when or if a UK investigator stumbles upon a therapeutic with such potential, will there be a UK company standing by ready to take up the license? Selling the rights to a US, Swiss or German company might make for one happy vice-chancellor and a handful of wealthy academics, but the lions share of the global revenues along with the manufacturing jobs, will remain outside the UK.

      For innovation to yield economic benefit, there needs to be proper investment in basic science (as in Germany), with investigators rather than research councils taking responsibility for the direction of their research. You mention that there is a feeling that the economic benefits of science funding have been over-sold. If this is true, it is not because funding is inadequate in the UK (though that is a handicap), it is because funding is being strategically and unwittingly manipulated in such a way as to ensure low returns.

    3. cdn lowest G8 corp taxes = low corp R+D says:

      I like challenge R+D but has to be good targets and flexible enough to abandon, which is hard for a rigid R+D schedule. Canada picked PEM fuel cells late 1990s and didn’t get good ROI due to technical hurdles. USA went to the Moon under W without any precursor R+D funding. I like R+Ding cheap mineral substrates. An Enceladus ocean core mission would cost $100Bs and might not succeed for under a trillion. The lab-on-a-chip sensors would be nanotech. Much soil has more than 25kg/m^2 of carbon, this dwarves 28g/m^2 annual peat growth, unless considering peat might impede drainage. At your Bangor U, C.Freeman is doing good research, but missing permafrost drainage. Need biogeology to grow peat quick.

    4. Stephen Moss says:

      I have attempted to post a comment but without success. Quite a long one. Is there a word limit?

    5. Richard Jones says:

      Sorry, Stephen, your earlier comment was trapped in the spam filter, which has a mind of its own. Now released.

    6. Richard Jones says:

      Patrick, I’m not sure how, if the problem is academic researchers being no good at innovation, having fewer of them helps, particularly if those that remain focus their attention on internationally leading research. I’m also not sure what you mean when you say “we” are starving our SMEs of funding. Who’s “we” here – do you mean the banks, venture capitalists, equity investors, or do you think the state should be stepping in more? And why do you single out their legal costs as the main problem?

      CDN, of course you must be right, the robustness of any “stage-gating” process has to be crucial for any R&D. And sometimes stuff just doesn’t work out, and you’ll have to accept that. But Ballard is still going as a fuel cell company, so Canada got some benefit from their PEM program, didn’t it? I suspect the problem there is one of where a particular technology (fuel cells) fits into a much bigger picture (getting a different infrastructure for powering automobiles). Not sure I would agree that the USA’s moon program didn’t have a precursor R&D program – surely the massive military rocket and satellite program was what that built on.

      Stephen, your examples of long-term benefits of research are good ones. But I think you are illustrating my point about the need to consider the wider innovation system when arguing what benefits you get from blue-sky research. In those cases, the blue sky research was able to provide economic benefits only because it existed in an environment in which inventions could be turned into products and taken to market. The worry is that that environment, now, doesn’t exist or is at least less favourable; perhaps it isn’t so much the impact agenda that makes it impossible to wait a long time for the pay-off from such blue sky research, as the demand of venture capitalists for a big return in 5-10 years.

    7. cdn lowest G8 corp taxes = low corp R+D says:

      I think (not sure) the Canadian gov of the late 90s invested about $4.5B in PEM technology with distilled water needs and a narrow temp range (bad in Canada) being the main hurdles, also was invested in a precious metals trough. They sold the vehicle cell technology they had for $260M, or something like that. Their forklift battery biz is good, but probably not worth much more than the big money passenger car arm they sold. That space application is an in situ sample analysis platform, that doesn’t exist yet. It would cut the time for sample analysis missions to Jovian and Saturn system in half. But needs automated SPMs, particle beams and other fossil findinf techniques. Nanotach obviously makes space cheaper via lighter computers and aerospace parts….I meant GWB funded the Moon speech out of nothing more than cutting NASA programmes needed to go to the Moon. The ISS was to be shortened, Cancelled CAM could’ve determined the Lunar plant composition.
      Forgot to mention for peat moss, it isn’t being bred like agri, or being extensively R+Ded; Freeman just figured out two years ago fuscum when wet retards decomposition enzymes.
      There are 50000 plants/m^2, so maybe a micron-tech gardening clipper could be devised to aid vegetative generation, something that looks like a tiny cheese grater. Existing farm implements traumatize the plant.