Science, Politics, and the Haldane Principle

The UK government published a new Science and Innovation Strategy just before Christmas, in circumstances that have led to a certain amount of comment (see, for example, here and here). There’s a lot to be said about this strategy, but here I want to discuss just one aspect – the document’s extended references to the Haldane Principle. This principle is widely believed to define, in UK science policy, a certain separation between politics and science, taking detailed decisions about what science to fund out of the hands of politicians and entrusting them to experts in the Research Councils, at arms’ length from the government. The new strategy reaffirms an adherence to the Haldane Principle, but it does this in a way that will make some people worry that an attempt is being made to redefine it, to allow more direct intervention in science funding decisions by politicians in Whitehall. No-one doubts that the government of the day has, not just a right, but a duty, to set strategic directions and priorities for the science the government funds. What’s at issue are how to make the best decisions, underpinned by the best evidence, for what by definition are the uncertain outcomes of research.

The key point to recognize about the Haldane Principle is that it is – as the historian David Edgerton pointed out – an invented tradition. It was only articulated as a principle half a century after it was purportedly created, and it has mainly been used, as are all invented traditions, to give a gloss of antique authority to current, more or less self-serving, arguments. Up to now, the Haldane Principle has usually been invoked to support the autonomy of science – the idea that the only people qualified to have any say in the direction that science is to take are scientists themselves. The argument here, though, is different, and made most clear in the response to the consultation on science capital: “The Government is committed to the Haldane Principle, which recognises the legitimate role of Ministers in taking long-term and large-scale strategic investment decisions.” These decisions concern investments in new research institutes, some already announced in the Autumn Statement, others promised to be announced in the March budget, a matter of weeks before the general election. These new capital investments of course carry implications for the general science budget, because their running costs have to come from somewhere.

David Edgerton’s 2009 paper “The Haldane Principle and other invented traditions in science policy” sets out the background. The supposed origin of the Haldane Principle was in recommendations made in 1918 by the former War Secretary, Lord Haldane. Edgerton’s view, however, is that this was only elevated to a principle in 1963, by the Conservative politician Lord Hailsham, who invoked it to criticize the then Labour government for introducing a Ministry of Technology: “Ever since 1915 it has been considered axiomatic that responsibility for industrial research and development is better exercised in conjunction with research in the medical, agricultural and other fields on what I have called the Haldane principle through an independent council of industrialists, scientists and other eminent persons and not directly by a Government Department itself.”

The point that Edgerton stresses is that while the principle of a free-standing, arms-length expert body making decisions on science spending has generally been adhered to for the kind of basic science funded by research councils and carried out, largely, in universities, for most of the twentieth century that has been only a fraction of the total science funded by the government, let alone that carried out in the country as a whole. Lord Hailsham himself was part of a government that maintained a massive, state-funded research establishment, largely driven by the needs of defense. As Edgerton says: ‘There was no Magna Carta, no Bill of Rights, no Human Rights Act, no constitution for government funded research; in 1918 or ever. There were lots of different arrangements and for different aspects. It was the case however, that in the realm of education, and non-departmental research, there was a powerful sense of the limits of appropriate government intervention.’

This sense of the limits of appropriate government intervention comes from something broader in our unwritten constitution, which I have found most clearly stated in Joseph Schumpeter’s great 1942 book “Capitalism, Socialism and Democracy”. In Schumpeter’s “realist” theory of democracy, for a democracy to be successful, some aspects of government need to be kept at some distance from politicians, focused as they are on the relentless day-to-day contest for votes. Schumpeter identifies the judiciary and criminal justice system as one such area, and the pre-1914 Bank of England as another case in which a major part of the state’s power – monetary policy – was entrusted to a body outside the control of politicians.

From today’s perspective, monetary policy is an interesting example, as it shows that these boundaries can shift over time. Shortly after Schumpeter was writing, the Bank of England was nationalized, becoming essentially an arm of the Treasury, which thus took over day-to-day control over interest rates. This changed in 1997, when the incoming Labour Government gave the Bank operational independence, to universal approval. It retains this independence under the current government, which has extended its powers even further.

In our system, Parliament can do what it likes, so there is no constitutional reason why the entire science budget should not be put under the direct control of a minister, just as the independence of the Bank of England could be reversed at any time, with the Chancellor taking back direct control of interest rates and monetary policy. Schumpeter’s view of democracy stresses to us that the correct question isn’t whether the government has the power to do this – it clearly has. The correct question is whether it would be a good idea. The trouble is that the Treasury wants control, and finds it difficult to countenance the idea that money can be spent wisely if the Treasury itself is not directing the money (despite the abundant empirical evidence for the contrary view). And what politicians want, above all, are announcements that will lead to some favourable headlines. As Schumpeter puts it, “the incessant competitive struggle to get into office or to stay in it imparts to every consideration of policies and measures the bias so admirably expressed by the phrase ‘dealing in votes’”. Both of these aims are difficult to reconcile with the long time-scales of major research projects and the expertise needed to ensure that decisions about such research projects are based on robust evidence. Nor can one very plausibly argue that politicians, right now, are particularly well trusted by the public at large.

Does this mean, then, that we really should leave it all to the experts? I don’t believe we should, and I do think that a Science and Innovation Strategy is the right place for a government to set out the big priorities, the priorities that do command the support of society at large. This is where I find the current Science and Innovation Strategy to be particularly disappointing. The will to exert more direct government control over the UK’s science enterprise is clear enough to see, but what seems much less clear is any vision of what the government wants to do with that control.

So, what should the overarching strategic goals of the government be for the science it supports, both directly and indirectly? To go back to Haldane, the question would hardly need to be asked in 1918. In the midst of a bitter European war, increasingly fought with the new technologies of submarines and aeroplanes, science needed to be applied to promote state military power, and medicine was needed to ameliorate the appalling human costs of the conflict. For Lord Hailsham in 1963, the military demands of the cold war would have been almost as prominent, though with his political opponents – the “White Heat of Technology” era Labour Party – there was an increasing sense that the UK should have been trying harder to extract economic benefit from its massive outlays on defense technology.

We should be pleased that, after the end of the cold war, the purposes of government sponsored science and technology no longer need to begin and end with defense. That said, the state does have a duty to defend its citizens and the current world presents many dangers, not least from new, widely available technologies. The strategy is almost completely silent on this issue.

In the current climate, the state of the economy should be a priority, and indeed the strategy opens with the slogan “Science and innovation are also at the heart of our long term economic plan”. But a strategy should be able to go beyond a slogan to some analysis and diagnosis, both of our wider economic situation and of the role science and innovation could play in improving that. Our economic position is characterized by the worst peace-time stagnation in productivity growth for more than a century, and a current account deficit running at 6% of GDP. We are becoming a low innovation economy, less and less able to pay our way in the world. The supporting Evidence Paper published with the strategy does a good job of summarising the UK’s overall competitive position in research and development: “overall R&D investment intensity in the UK remains below other leading knowledge-based countries and by some estimates may be sub-optimal for a developed country”. The Evidence Paper goes on to identify three particular sources for the UK’s poor R&D performance – heavy industrial concentration of research (with just one industry – pharmaceuticals – alone accounting for 22%), poor innovation performance in SMEs, and low R&D among UK-owned firms.

But the bleakness of the position that the evidence points to isn’t really reflected in the strategy. There are rather complacent statements such as “we are starting from a good base and have made excellent progress in recent years”, and “there is a strong industrial research and development landscape in large firms such as GlaxoSmithKline, AstraZeneca, BT, Unilever, Royal Dutch Shell and Rolls-Royce. Our innovative SMEs (small and medium enterprises) are wide ranging and growing.” All the companies mentioned are indeed good ones, but what is not mentioned is that the only two UK companies in the world’s top 100 R&D investors are the pharmaceutical companies GSK and AZ, and that UK business pharmaceutical R&D has fallen by around 18% over the last two years. Our business R&D landscape is too much a monoculture, over-dependent on a single sector which is itself in serious difficulties.

There are positive aspects to the strategy; the willingness to talk about industrial policy again is welcome – Innovate UK and the Catapult Centres have been successful, though arguably sub-scale. But in the context of this discussion, about the degree to which the government should directly control its spending on science and innovation, one thing is striking. One of the largest chunks of government spending on research is completely undirected. This is the £1.4 billion the government spends on the R&D tax credit, which subsidises research carried out in private companies. I understand the market failure argument for doing this, to compensate companies for the fact that they can’t capture the full social benefit of the research they do. And while the R&D tax credit undoubtedly carries a deadweight cost, subsidizing research that would have been done anyway, I’m sure it does lead to more industrial R&D being done than would otherwise happen, particularly in start-ups and SMEs. But it seems to me that there is a great deal of force in David Connell’s argument (in his paper Creating Markets for Things that Don’t Exist (PDF)) that directly contracting companies to do R&D in areas of strategic importance for the government is a much more cost-effective way of growing private sector R&D capacity than providing an indiscriminate subsidy through the R&D tax credit.

The final way of thinking about possible overarching strategic goals to underpin a Science and Innovation Strategy is in the language of grand challenges – those big, global problems that we need new technologies to solve. According to the strategy, “many of the ‘grand challenges’ for society, the ultimate customer for research, are obvious: developing cost effective low carbon power sources and storage solutions for energy-hungry economies; harnessing and managing scarce resources; improving human, animal and plant health.” I certainly agree with this. What the strategy is much less clear about is how we go about solving these grand challenges, what new institutional arrangements are needed, if any. Indeed, there’s some fundamental ambiguity or incoherence about what the strategy means by a “grand challenge” – elsewhere the document talks about allocating £900 million of the capital fund “to respond to grand challenges as they emerge”.

But it isn’t at all obvious to me that the solution to grand challenges like developing low carbon energy sources lies in creating new research facilities. What is surely more important is drawing together different disciplines (including some from arts, humanities and the social sciences – barely mentioned in the strategy). Mechanisms for connecting upstream research to translational work and scale-up, must be crucial, and this must involve the private sector – probably through the use of R&D contracts at a much larger scale. And work on grand challenges must be international – through the EU’s Horizon 2020 programmes (again, barely mentioned in the strategy), and with new partners who demonstrate an ability to scale-up and get things done – China, for example.

There is much more to say about this Science and Innovation Strategy, both positive and negative, but here I’ve concentrated on the issues surrounding the Haldane Principle, as I suspect these are likely to prove most controversial. To summarise my own personal position, I don’t see it as a problem that the government should want to steer the direction of the science it supports. Governments should set a long-term strategy for science; part of this will be about targeted research on the big issues we face. These issues should be chosen on the basis of a wide consensus – including real public engagement – and the government needs to be serious about following through upstream research with serious efforts to translate and scale up. But for part of the portfolio, the strategy should be not to have a strategy, but to maintain a diversity of funding for individual scientists and groups. It is through diversity that one maintains agility, being ready to respond to new, unexpected scientific developments, or maintaining the capacity to respond to new and unexpected threats. The strategy should be long-term, preferably to last more than one Parliament – and it’s probably prudent to write the strategy towards the beginning of the parliament rather than at its end, to avoid any perception that it is influenced by the short-term needs of electioneering. To implement the strategy, one should use arms-length bodies comprising experts of different kinds (not just natural scientists, not just academics), using transparent evidence-based processes such as peer review.

The danger one needs to avoid is ending up with the worst of all worlds – with politically motivated meddling at the operational level, but a failure of will when it comes to confronting the big problems of the economy and the environment.