Go back

Spending review deserves cheers, but tensions remain

Image: HM Treasury [CC BY-NC-ND 2.0], via Flickr

The chancellor’s largesse won’t relieve all the pressures on UK research policy, says James Wilsdon

Reflecting last week on the UK’s Net Zero Strategy, Rebecca Willis, professor in energy and climate governance at Lancaster University, described the “two-cheers problem” (that is, not three cheers) in environment policy.

This arises when ministers, civil servants and advisers feel that they have strained to deliver the best result that is politically possible, only to receive a tepid welcome from the green lobby that has been loudly pushing them hardest for change.

As chancellor Rishi Sunak presented his first spending review last week, the science lobby had its own two-cheers moment.

The uplift in R&D spending to £20 billion a year by 2025 falls short of the £22bn repeatedly pledged by the government. It makes reaching the medium-term goal of investing 2.4 per cent of GDP in R&D by 2027 harder. And it comes when the OECD average for R&D investment—on which the goal was based—has already risen to 2.5 per cent, with an impressive roll-call of countries exceeding this.

The £22bn target has been moved to 2027, but given the need to first navigate another general election, spending review, and perhaps chancellor, it’s effectively been kicked into the long grass. For now, £20bn it is.

This outcome is only disappointing, however, when compared with the high bar of expectation set by the government itself—particularly prime minister Boris Johnson, with his talk of the UK as a science superpower and predilection for white-coated, test-tube cosplay.

On any other measure, it’s a fantastic result. Set against flat or anaemic growth in R&D investment since 2010, and with intense post-pandemic pressure on all areas of public spending, securing a 25 per cent uplift in real terms over three years is a huge vote of confidence in the contribution that R&D can make to the government’s economic and social agenda.

Jitters and shrouds

It is also better than a jittery research community had feared. The weeks leading up to the review saw a barrage of letters and statements from the Russell Groupnational academies, Wellcome Trust, Campaign for Science and Engineering and others, warning ministers in ever starker terms of the dangers of retrenchment.

This aggravates the two-cheers problem. Having spent the summer shroud-waving, it can be hard to recalibrate in response to what is, on conventional benchmarks, extremely positive news. CaSE, for example, gave the chancellor a grudging three stars (on average) out of five.

Others were more enthusiastic, but the overall tenor of responses from research leaders was one of measured support. No one has given Sunak or science minister George Freeman a bouquet on behalf of the research community, in the way that William Cullerne Bown, founder of Research Professional News, did to then science minister David Willetts following the 2010 spending review.

Three cheers

I won’t be calling Interflora. But I am inclined to give HM Treasury the full three cheers. A couple of weeks ago, I summarised the possible outcomes of the spending review as ‘three Fs’: a fudge (more uncertainty); a fiddle (clever tricks with the public accounts); or a falling short against government targets.

We’ve ended up with the last of these, but not by much. As spending reviews go, this is about as good as it gets. Budgets will rise by £1.1bn next year, followed by a whopping £3.3bn in 2023-24, and the R&D system has received long-awaited clarity over the scale and speed of public investment.

Helpfully, the spending review also simplifies the R&D budget controlled by the Department for Business, Energy and Industrial Strategy (BEIS) into three parts. The first is core research, which covers quality-related funding, responsive-mode grants under UK Research and Innovation and the national academies. These receive modest but real year-on-year increases, shoring up the central pillars of university research on which so much else depends.

The second is Innovate UK, which receives a larger uplift to £1.1bn by 2025, as part of a suite of measures designed to boost business R&D investment.

Other positives include a near-doubling of research budgets for government departments beyond BEIS, some of which will go towards R&D in support of net zero; a relaxation of the rules for R&D tax credits, to make data science eligible; extra measures to attract talented researchers to the UK; and the unexpected restoration of the R&D elements of overseas aid, as the target to contribute 0.7 per cent of GDP kicks back in. This points towards a second coming of the Global Challenges Research Fund, after its sudden evisceration back in March sparked consternation across the research community.

Two horizons

Finally, there is a specified budget line for the UK’s participation in European research programmes, totalling £6.9bn from 2021 to 2025. This provides welcome assurance as to how the UK will meet the costs of Horizon Europe. But with final agreement on the UK’s association now snarled in the political impasse over the Northern Ireland Protocol, the path to the horizon is not yet clear.

UK researchers and businesses are being told to continue applying for funding as if we were formally associated, but the ongoing political problems will make UK partners less attractive to EU consortia and collaborators. This will exacerbate the downward trend in UK participation in EU research funding since 2016, with the exception of the European Research Council, to which researchers apply as individuals.

If this drags on into 2022, it may force a reappraisal of whether this £6.9bn is best spent through Horizon Europe or other domestic schemes—what science minister George Freeman calls his ‘Plan B’.

I spoke to a well-placed Treasury insider on the day of the spending review. He made it obvious where the government’s preferences lie.

“We’d be delighted,” he said, “if the research community turned around and told us they would now prefer us to invest that money directly in the UK system, and in extra support for international collaboration.” This, he suggested, would be more reliable and attractive than “inefficient, labyrinthine and unpredictable” EU schemes with “massive overheads”.

Many researchers would disagree, pointing to the collaborative opportunities and other benefits of EU programmes. My Treasury source stressed that there is no intention to row back on the UK’s commitment to association.

But the spending review casts the choice between this and Plan B more starkly. Sentiment in the UK research community may shift if delays persist and participation in EU programmes continues to fall.

Inputs and outcomes

Cheering the spending review doesn’t mean ignoring the pressures and dilemmas that lie ahead. Three tensions in research policy seem particularly acute.

The first is between a focus on inputs and institutions over outcomes and strategy. Since 2016, political and policy debate around research has been dominated by the size and scale of public investment and the institutional arrangements for spending it (including EU association).

Now that the spending review has settled some of this, attention needs to pivot towards the outcomes that the UK wants and needs from its R&D system, and the strategy to deliver these.

Here the signals remain mixed. The R&D Roadmap, announced in draft last summer as “the start of a big conversation”, looked as if it might become a point of strategic coherence, but seems to have been abandoned. And even as the system adapts to a flurry of institutional changes, including the creation of UK Research and Innovation in 2017-18 and the announcement of the not-yet-operational Advanced Research and Invention Agency in 2020-21, there is now the prospect of further tinkering through a review of the R&D landscape led by Paul Nurse, reprising his 2015 role.

There are valid arguments for greater institutional diversity in the UK R&D system. But there is also a pressing need to invest more and invest smarter in what is already working well.

On the biggest challenges, more work is needed to join the dots between ambition, investment and a fully formed strategy of the kind that then-chancellor Gordon Brown attempted with his 10-year framework for science and innovation in 2004.

How to ensure that extra public investment is matched by even larger increases in business R&D? How can R&D help to level up the UK economy and boost productivity? How can a policy discourse dominated by science and engineering be reconciled with an economy based largely in services?

The coming months may bring more clarity. UKRI’s long-awaited strategy is imminent, even as a fresh triennial review of how effectively it is working gets underway. There are some smart people working on R&D policy inside Number 10 and the Treasury, and Freeman is capable and experienced.

The new National Science and Technology Council met last week for the first time, chaired by the prime minister and attended by ministers across government, chief scientific adviser Patrick Vallance and UKRI chief executive Ottoline Leyser. Its focus will be cross-government and departmental R&D priorities, but it should act as a force for broader policy cohesion.

Hard and soft power

A second area of tension is between the soft-power benefits of investment in research and international collaboration, and a harder-edged view of the role of science and innovation in defence, security and foreign policy.

The dance between these is as old as science itself but, post-Brexit, the government is taking a harder line than its predecessors. This is reflected in March’s Integrated Review—perhaps the most significant science policy document of the Johnson government, and the clearest articulation of its science-superpower philosophy.

The tension is reflected in debates over Horizon Europe, and in the shift to a more hawkish stance on R&D collaboration with China. It also brings the Cabinet Office’s new Office for Science and Technology Strategy to the fore.

The OSTS, which is now up and running with a small team of civil servants and staff seconded from the military and security services, has a remit to identify opportunities that align with post-Brexit economic and foreign policy priorities. Its advice could determine the fate of a thick slice of the extra £4.4bn for departmental R&D announced in the spending review.

These may be sensible priorities. But as Ben Johnson, former BEIS adviser, described recently in Research Fortnight, there is a growing chasm between the research community’s “naive belief that the arc of history is bending towards a Star Trek future of global cooperation” and the government’s stance, which is “more Han Solo than Jean-Luc Picard”.

The smartest-guy problem

Early last month, the Commons inquiry into early failures in the government’s response to Covid-19 highlighted the pitfalls of groupthink and overreliance on narrow expertise. These lessons should lead to lasting improvements in the way the UK designs and manages its approach to science-for-policy, just as the failures over BSE and GM crops did a generation earlier.

Yet, on the other side of the fence, in the realm of policy for science and research, there is less sign of epistemic humility. For Nurse to be reviewing the research landscape for a second time in seven years—particularly when his last effort was methodologically weak and devoid of evidence—reflects a broader tendency in the UK for a relatively small group to exercise huge influence over the design and direction of the entire system.

Politicians and their advisers seem endlessly susceptible to what Stian Westlake, former adviser to several science ministers and now chief executive of the Royal Statistical Society, calls the “smartest guys in the room” approach to policy.

As he wrote in 2015: “The people who get the roundtable invites are typically alpha scientists: prestigious Nobel laureates, heads of learned societies and university vice-chancellors. They’re academic silverbacks (of both genders), the smartest guys (and gals) in the room. These are people who are not only ‘good at science’ but also extraverted and comfortable with dealing with committees. Ministers on the whole like to please them.” 

The longer I observe and occasionally participate in UK R&D policymaking, albeit from the cheap seats reserved for social scientists, the more I see the smartest guys in the room as an obstacle to progress.

I say this not to disparage the individuals; Paul Nurse is a brilliant scientist, research leader and genuinely lovely man. My objection is methodological. If only the smartest guys could apply the same standards of rigour and evidence they apply to their own research to their interventions in policy.

Call for evidence

It was heartening to hear John Kingman reflect in the summer about how much R&D policy “tends to turn on [the] gut feel of the individuals involved, [rather] than on hard evidence and analysis”. But it would have been more helpful if he had tackled this head-on during his five years as chair of UKRI, rather than mention it as he was heading out the door.

For now, in far too many areas of R&D policy, evidence remains optional: what matters more is the story about science, technology and research one tells, and the extent to which this reinforces wider myths of innovation and nationhood.

A huge effort has gone into winning the argument for extra R&D investment in the spending review. I hope that government and the research community can now allocate a tiny sliver of the same effort to improving the evidence and data that will enable us to use this extra money in effective ways.

Rather than leaping to redesign our institutions every six months, let’s take this as an opportunity to build a culture of testing, evaluation and experimentation with how we support, grow and sustain R&D. And let’s draw more openly and systematically on the collective intelligence, insight and expertise that researchers across the UK can bring to these tasks.

James Wilsdon is digital science professor of research policy at the University of Sheffield and director of the Research on Research Institute

A version of this article also appeared in Research Fortnight