Slouching towards an industrial policy

The UK’s Science Minister, David Willetts, gave a speech last week on “Our High Tech Future”. The headlines about it were dominated by one somewhat odd policy announcement, which I’ll come to later, but what’s more interesting is the fact that he chose (apparently at quite short notice) to give the speech at all, only weeks after the publication of a strategy for “Innovation and Research for Growth”, that was widely regarded as, at best, a retrospective attempt to give coherence to a series of rather random acts of policy. I’m tempted to interpret the speech as a signal that a not completely formed government policy is still evolving in some quite interesting directions. In short, after 32 years, the Conservatives are rediscovering the need for industrial policy.

To recap the story so far, the Coalition government came to power in 2010 led by a party with very few views about science one way or another, but with determination above all quickly to reduce public spending. Whereas commitments had been made to protect budgets in areas such as health and education, the Department of Business, Innovation and Skills (BIS) – in which the budgets for science and higher education reside was “unprotected”, so had to take a disproportionately large cut. Given the large fraction of this budget taken up by Universities and science spending, something had to give. The outcome was that recurrent spending for science was protected in cash terms, but the price for this was to slash university teaching budgets by a near-tripling fees and to more or less halve science capital budgets. Political discourse in 2011 was dominated by the controversial student fees policy, and a summer 2011 White Paper on higher eduction barely mentioned research at all. Meanwhile, however, the rapid bounce-back in the economy that the coalition was counting on has failed to materialise, and the question of how the government was going to promote economic growth was, by the autumn of 2011, becoming very pressing. In this context, the preservation of the recurrent science budget in cash terms was given new political prominence as a central part of the government’s attempt to present a narrative about promoting economic growth, and the autumn saw some rather abrupt new spending initiatives, for example a research centre for graphene and money for high performance computing, which restored some, but not all, of the cuts to capital budgets.

Unsurprisingly, then, Willetts’s speech starts by emphasising growth as the central problem facing the government, and outlining some of the supply side measures in place to promote it. The dilemma that Willetts faces is that, while it’s clear that generic policies such as regulatory reform aren’t by themselves enough to promote growth, there’s a deep reluctance to support specific business sectors – that government can’t “pick winners” is now an article of faith. But Willetts acknowledges not only that government does need to make choices – for example in deciding where our energy is going to come from, but also that choices the government has made in the past have, in effect, amounted to a tacit industrial policy. The example Willetts uses to illustrate this is a particularly striking one – when the government funded the extension to the Jubilee line (at a cost estimated at £3.5 billion in 1999), this was justified as necessary for the development of the financial service industry in its new centres in East London. So if we do have an industrial policy, we might as well be open and thoughtful about what it is, and when the government talks, as it does, about rebalancing the economy away from financial services towards manufacturing we must recognise that what we are talking about is, indeed, an industrial policy.

So we soon get back to the familiar question of how we can turn a strong science base into “high tech growth”. The speech celebrates the strength of the research base across the board – specifically including arts, humanities and social sciences – and points to the recent report The International Comparative Performance of the UK Research Base as evidence of this strength. Willetts’s stated aim is to make the UK “the best place in the world to do science”, and this means it must be properly funded – hence the maintenance of the science budget (in cash terms, excluding capital, one needs to add here).

But how do we decide how this money is spent? Here Willetts makes a strong claim that one source of the strength of UK science is its independence from government – that it is “protected by the Haldane principle that Ministers do not decide on funding for particular research projects or particular university departments”. Many readers in academia will contrast this statement with the evident reality that funding agencies are being increasingly directive about what sort of science they will support, and increasingly active in pushing the scientific enterprise to support their strategic priorities – surely, they will ask, this tendency is running contrary to the Haldane principle? Like so many things in British public life, the Haldane principle is an invented tradition that is used by all sides in support of their own arguments; note that Willetts defines his version of it precisely here, in a way that is entirely consistent with research councils, or for that matter ministers, making judgements about which areas of science to prioritise. Within the research councils, there is a balance between “responsive mode” research and more directed programmes, though Willetts does make the point that this distinction doesn’t really map directly onto a division between “blue skies” and more applied research – some applied research is funded through responsive mode, while some directed programmes look pretty blue skies (the example he chose for the latter category – CERN – is important though hardly typical). But for those in the scientific community currently agitating against things like EPSRC’s “Shaping Capability” activity, it’s clear that he has only limited sympathy – “there comes a point when the Research Councils have to think about impact and priorities. I know this is controversial – and I do receive mass letters from aggrieved sections of academia who fear the Research Councils have failed to recognise their special significance. But it has to be done and the Councils try to do it in an open way that commands the consent of the research community they serve.”

One slightly unusual feature of the UK research system in comparison to much of the rest of the world is the dominance of universities in the publicly funded research landscape – in contrast, say, to the importance of Max Planck Institutes in Germany and National Laboratories in the USA. We do have some institutes with the characteristics of national labs in the UK – Harwell/Rutherford Lab, Daresbury, Norwich and Babraham. There does seem to be a growing emphasis on non-University research centres, with extra funding going to these campuses and the very large funding going into the new Crick Laboratory at St Pancras. But Willetts does recognise that universities do have a very large role in our research system, and that the number of UK universities in the top 100 and 200 of the various league tables is a source of strength. In fact, he wants to increase this number, saying “Today I set our ambition of aiming for the number of our universities in the top 100 to grow”.

At the moment, the number of UK top 100 universities is between 10 and 19 according to which ranking you use. This prompts two thoughts. For some years now, there’s been an explicit policy from government of increased concentration of research in a smaller number of universities. This has been enthusiastically supported by lobby groups of the most research-intensive universities, such as the Russell Group (my own university, Sheffield, is one of the twenty members of this group). But there’s been a certain amount of worry, even paranoia, that the current government’s research concentration agenda, together with a greater focus on national labs, might go further than that, with most research ending up in the top 5 or so. Since research strength is a major ingredient in these league tables, further research concentration is probably not consistent with pushing up our numbers in the world top 100.

But this discussion about league tables brings out another important point. In the evolution of the government’s strategy on higher education, there’s been an almost complete separation of research and teaching, and indeed a sense that the two missions are in tension. But the lesson of the league tables is that university reputation isn’t partitioned in this way. High league table places attract students, even when (as in the case of the Shanghai Jiao Tong table) the league table place depends entirely on measures of research performance. The students are not being irrational at all here – a substantial part of the value of the degree depends on the reputation of the university that grants it, and that reputation is driven largely by research. This separation of research and teaching in the thinking of government about universities has had another consequence – an almost complete neglect of the important role of postgraduate students in universities.
Postgraduates do, for a change, get a mention in Willetts’s speech, but in a rather puzzling context – the announcement that “the Coalition is inviting proposals for a new type of university with a focus on science and technology and on postgraduates.” It’s this announcement that gathered the headlines, with many people subsequently pointing out the difficulties of doing this in the absence of any government money. There’s not much I can add to other people’s analysis here (for example, this Nature editorial).

How do we turn excellent research into money? Not for the first time, Willetts expresses his scepticism about the emphasis on commercialising protectable IP through venture capital funded spin-outs – “the classic model expects venture capitalists to be following what is happening in universities and to invest in the IP after the university has spun out a company. This is the conventional sausage machine and it can work on some occasions but it is not widespread or straightforward. We have expected venture capital firms to finance early stage start-ups much further upstream than is realistic. Then we beat up on ourselves that our venture capitalists do not take risks they do in the US when even there the model is rather different.” But the reference to the USA does lead to an important new development – a growing realisation in the government that the USA, for all its free market rhetoric, is a state that is as shamelessly interventionist in technology and innovation as France or Korea. As Willetts says, the scales falling off his eyes: “The land of free enterprise has an innovation and research system which depends on federal and state government just like everywhere else in the Western world.”

In the UK, state intervention in nearer market technology is the role of the Technology Strategy Board, and our colleagues in TSB will be pleased to see the centrality of their role asserted in this way, especially since they had a few nervous months after the formation of the Coalition in which no-one in government seemed to mention them. Our TSB friends might also note that their budget wasn’t included in the science ring fence, even with flat cash, so at the same time as they’ve had reduced resources they’ve had an increasing number of responsibilities and new initiatives given to them to handle, such as the new “Catapult Centres” for translational research in collaboration with industry.

Once one has accepted the need for government to intervene, and has identified the agencies to deliver this intervention, the next question is how to decide what interventions to make. Here we do see more policy evolution from the government, with the idea of “sector leadership councils” set up by BIS to identify priorities. We already have such councils for space, automotive, aerospace, and life sciences, and further ones for e-Infrastructure and synthetic biology are being set up now. I think this is a positive development. I’ve written before of the difficulties research councils face (Some questions for British research policy) in being expected to deliver science in support of national priorities when there isn’t really a mechanism for deciding what those priorities are or should be, and when it isn’t obvious that the research councils have the capacity to make such judgements themselves. So a more thoughtful and informed way of deciding on those priorities should be a good thing.

But any such approach isn’t without dangers – to state the obvious, the quality of the advice these councils give depends on the quality of people on them. There’s the danger of incumbency bias and group-think, and a worry that the interests of existing industry have too strong a voice. This could have the effect of locking in place the existing UK industrial structure, and excluding the voices of those who might be able to deliver truly disruptive innovation. And there’s an even greater danger that the voice of ordinary people, those who will be profoundly affected by new technology without necessarily having a big stake in it, will be excluded as well. This will certainly be a very pointed issue for synthetic biology. The Minister has in the past spoken very approvingly of the public dialogue about synthetic biology that the research councils organised; the question must be, as the technology develops, will this involvement of the public continue to ensure that this technology is developed in a way that meets widely shared public goals and aspirations and recognises public worries?

Talking about “sector councils” makes clear that their are different ways of classifying and organising science and technology areas. Classifying by the technological needs of a particular industry sector – the automotive sector, say – is one extreme, while classifying by academic discipline and field – synthetic chemistry, for example – is another. A third classification which has been growing in prominence recently refers to societal grand challenges such as the ageing population, sustainable energy, or food security. Willetts finishes by talking about a fourth classification, with a nod to the idea of “general purpose technologies”, which have the potential to transform many different industries. The formulation Willetts uses is one with origins in the USA – the “bio-info-nano-cogno” technologies – though he is sceptical of the more overheated claims about the convergence of these technologies and wishes to add “carbo” for low carbon energy.

Using this as a checklist, we find bio covered via the government’s Life Sciences strategy, an emphasis on agricultural biotechnology, especially food security as envisaged by BBSRC, and a new roadmap for synthetic biology. For info and cogno, the East London TechCity development gets a mention, but the focus is on high performance computing and a new government report from Unilever’s Dominic Tildesley (which actually calls for quite a lot of government intervention). For nanotechnology, there’s the new investment in graphene research, but another new industry group has been set up to look at it. This is of particular interest to me, of course; it’s been a year or two since the last nanotechnology strategy and in the meantime the subject disappeared from UK science policy, prompting my post last summer Why has the UK given up on nanotechnology?

For low carbon, Willetts notes that “High on our agenda now is nuclear fission and fusion after a challenging report from the Science and Technology Committee of the House of Lords”. The report being referred to here – Nuclear Research and Development Capabilities – is “challenging” in the sense of offering withering criticism of the government’s belief that it is possible to have a significant new build of nuclear power without an underpinning strategic research effort (though of course the main culprit here is the Dept for Energy and Climate Change, which is not Willetts’s responsibility).

So, to summarise, what we’re seeing here is a small-government Conservative rather thoughtfully working through the implications of a realisation that more government intervention is needed in order to deliver on the promise of economic growth from the science base. For me, two issues arise. Firstly, the obvious problem seems to be that Willetts is willing the ends while being unable to provide the means, in terms of the resources needed. Secondly, what I don’t yet see is a connection between this important debate about where our growth is going to come from and the widespread dissatisfaction with the particular variety of capitalism we seem to have ended up with. This was the subject of my earlier post Good capitalism, bad capitalism and turning science into economic benefit. The argument here is that rebalancing the economy will have to involve discouraging bad capitalism as well as encouraging the innovation we’re all in favour of. Bad capitalism crowds out responsible innovation.

8 thoughts on “Slouching towards an industrial policy”

  1. Willetts’ speech covered too many areas to easily discuss in a single blog, but I agree with your overall view of where science policy is heading under the coalition. He does seem to be feeling his way and, in defiance of the laws of physics, heading in several directions at once. Some aspects appear solid and reasonably well thought through, like the leadership councils and catapult centres, but others such as new universities for postgraduates seem utterly fanciful. And there is a clear disconnect between blue skies research and commercialisation that I blogged briefly about here.

  2. Thank you for such a succinct and simple explanation of the state of play and how things might start get fixed (despite being longer than you intended). I hope it gets widespread attention.

  3. Stephen, of course I know that many people share your worries about the lack of support for blue skies research. But the more I think about it the more difficulty I have with the term “blue skies” itself, and what it might mean. You could argue that blue skies research is what you do if you are motivated purely by the desire to find out how the world works, and applied or goal-directed research is what you do if you want to make interventions in the world, through making new devices or guiding action or policy. But then the question comes, whose motivation are we talking about – the motivation of the scientist or the motivation of the funders? There seems to me to be a bit of a paradox in your position, which is to say that we will find out more useful and patentable things if scientists get more money to do blue skies research – you could argue that if the result of the funding is to produce patents, and that is its purpose, then its not blue skies research, but goal directed research. Then you’re arguing not about the best balance between goal directed research and blue skies research, but about the best way of funding goal directed research. In other words, your saying that if the goal is to produce more likely drug targets, the best way to do this is just to give talented scientists unrestricted money (i.e. from the point of view of the scientist, you’re funding blue skies research because their motivation is curiosity, but from the point of view of the funder this is goal-directed research because you’re expecting to get more patents out of it). This, of course, is quite a popular position amongst some influential people and can frequently be heard in the vicinity of Carlton House Terrace; and through things like the Wellcome and ERC schemes more funding seems to be going into this direction. I’m not sure that this is an unalloyed blessing. I wonder also though whether your complaint about not being able to get money for “fishing trips” is simply a reflection of a much more local problem, in that BBSRC’s view of what good science is seems occasionally to be one underpinned by philosophy of science that’s about 50 years out of date.

    Since you mention patents I should also say that I’m not at all convinced that counting patents is very relevant. The number of patents coming out of Sheffield’s research is controlled, above all, by the size of our patent budget. If we wanted more patents then we could increase the size of that budget. I don’t think there’s a strong case for doing that because I don’t think this would bring a return, given the current difficult conditions for raising venture capital and finding licensing opportunities.

  4. I almost agree with your comments on patenting. I am also not surprised that you are effectively rationing your patent budget – I think this is prevalent. However, I thought the strongest argument for university patenting and IP management is that it protects the interest of the exploiting company – not to raise institutional funds. Unfortunately current metrics (eg Hebci & HEIF) seem to reflect income generation and not consideration of potential market to exploit.

    The recent IPO report on Graphene implied that the UK was failing to patent, in comparison with overseas competitors. For an emerging technology where the UK started with first-move-advantage, this seems worrying. However, I think it says as much about the indiginous capacity to exploit, as the entreprenuerial inclinations of the universities.

  5. I agree with your point about the problem of defining blue skies research, and as I mentioned in my blog it appears that ministers and research council heads are just as unclear as the rest of us. In fact, it may well be that blue skies means different things to different disciplines. But I did not intend to give the impression that the purpose of blue skies research should be the generation of patents, and I’d stand even further from the suggestion that patent counting would serve any value.

    For many of us doing blue skies research the motivation is curiosity, and most of the basic research I have done has not led to patents. But I do argue that patents frequently emerge as a bi-product of blue skies research, and that if Willetts wants to see the UK share of global patents increase, then it would be constructive to devote as much attention to this end of the research spectrum as he is clearly giving to the more applied end.

  6. Mark, I do agree that generating a financial return isn’t the primary reason for Universities to protect IP (which is just as well as as an income stream it’s pretty much negligible even in the most successful institutions). But to do it to protect the interest of the exploiting company only makes sense if there is an exploiting company; if we can’t find a company with enough interest in it to license the IP, and we can’t put a plausible enough business plan to attract backers for a spin-out, then it’s not going to get exploited. Given the way patent costs escalate with time it makes sense to have a fairly clean and focused approach to deciding what’s worth sticking with and what isn’t.

    Stephen, I don’t doubt for a moment that all sorts of useful discoveries – patentable and otherwise – have arisen from blue skies research, in the sense of research motivated by the curiosity of the researcher. But how effectively this transition happens depends on the way an innovation system is arranged around that blue skies research. Many great and useful discoveries were made by physicists working in the 1950’s and 1960’s in environments that gave full expression to those scientists unbounded curiosity. But those environments were closely integrated into the wider military and industrial structures of that particular cold war world, and it was those very instrumental and utilitarian needs that ultimately provided the justification for giving those scientists that freedom.

  7. Dear Richard,

    having heard Lord Mandelson on the radio this morning it seems that there is now broad political consensus about having a more interventionist policy in the future of a high tech UK economy. He comments that he could have written a speech that David Willets gave the other day. His take is that we are just not spending enough on innovation as we should be spending.

  8. Relating to the closed CIGS thread….
    My government fought a carbon-tax, de facto subsidizing (themselves to hell) coal tar and oil for tens of billions annually. I begged them in 2007 to hit up a company, Gold Canyon, for a few tens of millions. They had a large Nevada CIGS deposit back when CIGS was cheap….any government that is formed out of the hatred of gov (Canada’s present arose paradoxically out of a lack of enough Great Depression gov), or unthinking maximizing of rich net worths…is inefficient.

    Given the lack of carbon prices or any tempering of emissions, until the Holocaust at least, I wonder if CVD furnaces will share this fate? If the hydrocarbon feedstocks, maybe CO2 is more plentiful, for CVD become too expensive, will that mean processes like laser ablation to create CNTs and nanohorns, will be favoured? Should laser ablation, where hydro or renewable utilities, be favoured now?

Comments are closed.