Stepping out of Thatcher’s shadow

After 40 years of neoliberal failure we need more than superficial change in industrial strategy

On Boxing Day I got into an argument with the Tony Blair Institute on BlueSky. Could I justify my claim that its approach to industrial strategy doesn’t cross the neoliberal, no near-market support line drawn by Margaret Thatcher? Once we got into it, it turned out the TBI doesn’t know what the long-standing term “near-market” signifies. In a way, that’s to be expected. The TBI’s team has been set up to come up with new ideas and so they can’t spend too much time looking backwards. But the risk then is that the ideas the institute comes up with are superficial and repeat the failures of the past, which is what I think is happening.

I promised the TBI I would explain what near-market research is, and so here we are. In doing so, I will clarify some of the contours of the neoliberal, Thatcherite approach to industrial strategy that has shaped the country for the past 40 years. Then we’ll examine some examples of proposals from the TBI’s reports that it gave me to change my mind and see whether they break the mould.

Neoliberalism

Neoliberalism is an evolving project in which the idea of the market plays a central role; it posits itself as an alternative to central planning. According to Quinn Slobodian, neoliberals see “the intellectual project as finding the right state and the right law to serve the market order”. In this sense, it defines the EU as it was before the social chapter. In the story I’m telling, that is an important factor; all through our years in the EU, a hefty dose of neoliberalism was baked in – especially after Thatcher pushed through the Single Market Directive. However, at the practical level of drafting policy, Slobodian’s rather encompassing idea has been routinely boiled down to something much cruder and with much sharper edges: government intervention is justified only when market failure can be demonstrated.

For myself, I recognise the limits of central planning, the benefits of markets and the underlying logic of the market failure argument. The problem comes when that argument is applied in a slavish way that cleaves to an idea of market perfection and fails to take account of the many other factors that may complicate the picture. This dogmatism is what Thatcher brought to the party.

Since Thatcher, I think it is fair to say that we have had at core a neoliberal state. It has evolved, and this is why I talk of Thatcherism Mark I and Mark II, a distinction that will apply also to industrial strategy. You can make a case than politicians as diverse as Gordon Brown, Boris Johnson and Vince Cable were not neoliberal by instinct but, as I will show below, it didn’t matter very much. In industrial strategy, we’ve gone on living in Thatcher’s shadow.

Thatcherism Mark I

I remember a slightly misty night maybe 25 years ago, walking back along Carlton House Terrace towards Piccadilly with Robin Nicholson from a meeting at the Royal Academy of Engineering. Nicholson was a gifted metallurgist who rose to become the chief executive of the mining company Inco, a founding member of the Dow Jones Industrial Average. In 1981 he became Thatcher’s chief scientific adviser.

He told me that in, as I recall 1983*, he had gone to Thatcher with a pitch for increased government spending on R&D. To make the pitch, he divided R&D into two parts: research that was near the market, that directly led to products, typically in the labs of commercial companies; this was what he wanted more money for. And research that was far from the market, whose application was unclear, typically in universities. The pitch backfired. There and then, in the meeting, Thatcher said that near-market research was exactly the kind of thing the government should stop doing.

That established a shibboleth that took previous Conservative criticism of “picking winners” and transformed it into an ongoing campaign waged from within Whitehall. From that point on, the injunction against near-market research relentlessly tore away at government support for R&D. I started writing about this stuff in the late 80s and until 1997, cuts to government R&D justified by the shibboleth was a staple of the reporting. An example of the problem created, much discussed at the time, was the absence of money for demonstrators that scaled up new processes devised in the lab to show they were viable at scale.

As with so much Thatcherism, the shibboleth came with a neoliberal just so story rooted in the axiomatic perfection of markets. This was that cutting government support would allow corporation tax to be cut, which would allow the companies themselves to increase their spending on R&D. The chart below shows what did actually happen to spending on R&D in the UK.

As a percentage of GDP, total spending (top line in the chart) fell from more than 2 per cent to just over 1.6 per cent by 2013, when the comparable figures compiled by the OECD come to a halt. The vast bulk of that fall came from a fall in industrial R&D (middle line). Business Enterprise Research and Development, expenditure on R&D carried out by businesses regardless of the funding source, fell from about 1.4 per cent to just over 1 per cent of GDP. In defiance of the neoliberal just so story, private companies did not respond to the government’s withdrawal of support by increasing their own R&D expenditure, they responded by cutting it.

There are those who argue the idea of Thatcherism as a break with previous British approaches to our economic problems is overblown, but here in the realm of research, innovation and industrial strategy, it is impossible to overstate the impact of her uniquely strident interpretation of neoliberalism. As Mariana Mazzucato has been explaining for maybe 20 years, even the Americans under Ronald Reagan weren’t loopy enough to quit government support for near-market research. As Rachel Reeves said in her Mais Lecture, Thatcher’s policies were the source of “an unprecedented surge in inequality between places and people which endures today. The decline or disappearance of whole industries, leaving enduring social and economic costs and hollowing out our industrial strength. And – crucially – diminishing returns for growth and productivity.”

Today there is nothing specially confidential in the story Nicholson told me. In the research policy world, even the dogs on the street know it. However, there seems to be an attempt to rewrite history going on. Nicholson died last month and received an obituary in the Times that missed out the critical interview with Thatcher. In this way, the obituary managed to turn Thatcher’s effect on the UK’s high-tech efforts on its head by telling us he was appointed because, “It became clear that she would only countenance an adviser who could play a more dynamic role in making scientific research more commercially successful.”

Thatcherism Mark II

Once Thatcher was gone, it became possible to point out the damage that had been done and most governments since then have tried to do something about it. This is Thatcherism Mark II in industrial policy. The exceptions to this have been when a Mark I Thatcherite – Sajid Javid or Kwasi Kwarteng – got their hands on the industry department and started tearing things up again. So when people talk about a lack of stability in industrial strategy, intensifying the UK’s already chronic short-termism, what they are really referring to is an ideological dispute between two factions in the Parliamentary Conservative Party that successive Conservative Prime Ministers have allowed to play out in the heart of our economy as if it were a game of ping pong.

Thatcherism Mark II never repudiates the shibboleth; rather, it proceeds by inventing new, plausible-sounding ways to solve the problem the shibboleth has created without touching the shibboleth itself. These initiatives come in two varieties, commercial and academic, and the more trite they are, the more insistently they are dignified with the rhetorical robes of “industrial strategy”.

On the commercial side, under John Major the government started to exhort companies to spend more on R&D – for example Michael Heseltine set up an R&D Scoreboard to this end. And the Department of Trade and Industry began to re-engage with industry on questions of innovation and research, partly by organising a gigantic ‘foresight’ exercise involving hundreds of meetings across a dozen or so technology areas. But at the same time the EU budget for what is now called the Horizon programme was growing rapidly and the government’s response was to cut the DTI budget by the amount extra the UK had to pay. This meant that none of Heseltine’s more substantive ideas ever got funded, including his plan to create a network of technology transfer institutes on the German Fraunhofer model. Trapped in the neoliberal framework, the man who had resigned from the Cabinet when his plan for saving the Westland helicopter firm was rejected was impotent.

In New Labour, Blair largely left the questions we are dealing with to Brown. In the same vein, the TBI’s own examination of the question of whether New Labour was neoliberal or not doesn’t mention industrial strategy.

Brown announced a ratchet such that if private sector spending on R&D went up, so would government spending. It was never put into motion. He also set up the body now called Innovate UK, a body that operates in the same pre-competitive zone as the EU’s Horizon programme and has similarly struggled to be effective. As Prime Minister, he also established a new department under Peter Mandelson with responsibility for industrial strategy, but for the life of me I can’t remember a single thing that came out of it.

There was however one innovation that has demonstrably made a difference, tax credits for R&D. At roughly £10 billion a year, revenue foregone on R&D tax breaks is now comparable in scale with spending on universities through UK Research and Innovation and there is plenty of evidence that it is effective in increasing commercial spending on R&D.

Tax credits for R&D address a market failure. Firms can’t capture all the benefits of their R&D and thus invest too little left to their own devices. This makes the policy, like the others of Brown’s, comfortably neoliberal.

Under the Coalition, Vince Cable established some industry panels and implemented Heseltine’s idea for UK Fraunhofers, but with far too little funding from George Osborne to emulate their German inspiration. He called this industrial strategy. David Willetts picked some technologies for far-from market support and called this industrial strategy. David Cameron and Osborne got very excited about the idea that Chinese capital would rebuild manufacturing in the north of England but failed to make it happen.

Osborne also created a new set of fiscal rules, still in place under Rishi Sunak, to constrain the activity of departments. The debt rule counted the cost of an investment (for example in a fast-growing company) but put no value on the asset acquired. As with Thatcher’s original injunction against near-market research, this rule is heedless of the question of market failure. It thus baked her strident form of neoliberalism into the imagination of Whitehall. Note however that it required no withdrawal from the activities already established by Brown, which continued; it is the presence of such continuities that makes Thatcherism Mark II a useful analytic category.

Osborne also introduced the ‘Patent Box’, a novel form of tax incentive for R&D. This is notable for enriching pharma shareholders by creating a new form of international tax competition while doing nothing for the UK, just as the Institute for Fiscal Studies warned him it would.

Theresa May created an Industrial Strategy Council and announced an Industrial Strategy. As with Mandelson, as far as I can recall, on the ground nothing happened.

Johnson also got excited about Chinese investment in the North, this time in the green transition. It again never happened. Instead, in a very un-neoliberal way, he decided to invest hundreds of millions in the OneWeb satellite company. This novelty has not been repeated. In an equally un-neoliberal way, he gave subsidies of unknown extent to established manufacturing operations to stop them pulling out of the UK after Brexit. And he set up the Advanced Research and Invention Agency.

Since ARIA has been the focus of an enormous amount of hype, let’s look at what it is actually doing, for example at its Scaling Compute project. This is funding university researchers to look into technologies that might reduce the hardware costs required to train large AI models. This kind of programme is the bread and butter of the research councils; I have seen dozens of similar programmes over the years. Maybe the ARIA way will prove superior to the old way, but the gains can only be incremental advancement of an established and already oft-refined mechanism. The hype around ARIA is a classic case of the way Thatcherism Mark II repeatedly makes bogus claims of revolution in strategy.

As urged by the TBI, Sunak established the Department for Science, Innovation and Technology, which entrenched the division between inventing things and making them. Aside from that, he was another nothingburger. In particular, he failed to come up with a plan for how the UK could turn its stellar performance in AI academic citations into commercial value during the biggest gold rush in history.

On the academic side, the neoliberals have always funded basic research generously. This is reflected in the bottom line on the chart, expenditure by higher education institutions on R&D, which rose from about 0.3 per cent to just over 0.4 per cent of GDP. Time and again, when public spending was cut, academic research was not. However, this additional funding came with a series of innovations aimed at getting more commercial bang from the academic buck. To name a few: Major had the Realising Our Potential Awards for individual academics; New Labour introduced the University Challenge Fund to allow universities to invest in their own spinouts; Cameron’s time brought the formulaic Higher Education Innovation Fund. Throughout it all, the Research Excellence Framework (previously the Research Assessment Exercise) steadily allocated more weight (and hence funding) to departments that linked up with businesses.

Over the decades that Thatcherism has held sway over our industrial policy the country excelled at the thing the government supported, inventing things, and became much worse at the thing it didn’t, making things. One illustration of this is the gradual disappearance of the British firms that appeared on Heseltine’s R&D Scoreboard. And this is one reason why the policy focus in the UK has shifted from getting well-capitalised big firms to spend more on R&D to getting more capital into under-capitalised but R&D-intensive small firms.

This shift from big to small combined with the increased funding for academia and, given the near-market research shibboleth, naturally led to what Richard Jones calls over-reliance on supply side innovation policy and what I call ‘the goldfish syndrome’. In this economic disorder, the role of the state is conceived as providing for an environment, the goldfish bowl of academia, that spawns glittering goldfish – great ideas. For economic benefit, the state relies on external cats, private companies, to be attracted by the goldfish and scoop them up. The policy goal then is to make the goldfish more glittering (e.g. ARIA) or to make the glass in the bowl more transparent (e.g. increase university spinouts) or both (e.g. DSIT). The hard yards of building companies that win in the marketplace is left to the magic of the market and no heed is given to the question of whether the cats are British or not. The fact that in real life once the cat scoops out the goldfish it is dead is reflected in the fact that, in the neoliberal framework, those glittering ideas are dead to the policymakers once they have been acquired by a company.

Today, the prospect of a neoliberal world order has disappeared. Instead, China has shown that centralised planning can be successfully combined with markets and ushered in a new era of military tension that spills out into trade. The neoliberal approach in the UK, Thatcherism Marks I and II, has been exposed as shortsighted, ineffective and naïve in both the economic and geostrategic senses.

At the same time, two breaks with the past have liberated Starmer’s government from the neoliberal straitjacket. The first is Brexit. The UK is no longer constrained as it was by EU law to pre-competitive forms of subsidy. The second is the change in the fiscal rule on debt made by Rachel Reeves. Now that this rule is cashed out in terms of Public Sector Net Financial Liabilities, departments can make investments and the value of the asset acquired will be recognised. The central question for the coming industrial strategy therefore is how these new freedoms are to be exploited to meet our economic, political and strategic needs.

Now let’s look at the policy recommendations highlighted by the TBI and see whether they are a continuation of the neoliberal status quo or break out of it. These can be divided into two categories, policies that affect government support for R&D itself and policies that seek to increase the amount of private sector investment in R&D and fast-growing firms.

Policies that affect government spending on R&D itself

The TBI highlights one recommendation under this heading, a proposal for a Laboratory of Biodesign to focus on the invention of new biotechnology that is at too early a stage for commercial investors. “It would design and build new biotechnologies, biomolecules and therapeutics, helping to produce innovative ideas, build a strong pool of talent and bring biology together with computation,” says the proposal.

The creation of a big new biolab is not unknown in the UK. In our established research council system, scientists are perfectly capable of identifying exciting fields themselves and mobilising the necessary resources. Why then is the TBI involved? Logically, because the case for this lab is not purely scientific but also economic. The TBI included the following screenshot from the original proposal:

This makes clear that the primary mechanism through which the laboratory is to generate economic impact is via the commercial spinout of frontier biotechnologies, a classic instance of the goldfish syndrome. As the document itself states, this is something that other government-funded labs in the UK have tried and failed to do in the past. On its own account, the TBI here is repeating an approach that has previously failed multiple times. To persuade us that this time the TBI knows how to avoid the same result, it shared the following screenshot:

The first three bullet points have been in place at the LMB for more than half a century and the last two are unproven. At the end of the day, they are just renewed attempts to get more commercial bang from the academic buck and there is no reason to think their effect will be any greater than the old ideas. The key point is that the lab is not in fact a company; it is not trying to conquer markets; as in all instances of the goldfish syndrome, that is someone else’s job. Thus what we are looking at here is more of the same old same old. By choosing to concentrate limited resources on a new Laboratory of Biodesign, the TBI is making a choice about what area of science is likely to yield commercially valuable insights in the coming years, but the far-from market limitation it is imposing on the way those resources are organised implies the economic impact will be limited, in the UK at least. I can’t imagine a better example of both the TBI’s commitment to the status quo and of why the status quo is problematic.

An alternative would be to create not a lab but a company, to vest in this company all the facilities and funding that is envisaged for the lab, and to also arrange additional funding for the company itself and start-ups around it with the goal of establishing a dominant, monopolistic ecosystem in the new industry that is perhaps emerging from the new technologies. That would make sense of having projects run by commercial managers as the TBI advocates, cross the line drawn by Thatcher and break with the status quo. If you really believe there are huge commercial opportunities coming and that the UK has a powerful comparative advantage over other nations, why not?

Mobilisation of capital

When it comes to the mobilisation of capital, there are some basic points of agreement between my own position and that of the TBI. We both want to get more capital into fast-growing firms. That is what the meeting had been about all those years ago with Robin Nicholson. We both advocated the adoption of Public Sector Net Financial Liabilities as the yardstick for the revised fiscal rule on debt. We both want to retain in the UK more of the innovative firms we create. This being the case, how can I accuse the TBI of being a continuation of past failure while insisting that I am not?

It boils down to a superficiality that I discern on the part of the TBI. As I have described above, the defining feature of Thatcherism Mark II in industrial strategy is the presentation of plausible sounding policy innovations that try to work around the near-market shibboleth and end up delivering very little. What I see in the TBI’s policies is a repeat of that same pattern, but this time presented in a form that will appear plausible to a government that, for the first time in more than 40 years, is not committed to neoliberalism.

Here are the proposals highlighted by the TBI on BlueSky:

1. Make tax relief more generous for investors in Venture Capital Trusts that invest in tech

2. Increase the number of AIM tech firms covered by analysts so as to make trading their shares easier

3. Incentivise pension funds to consolidate and invest in the UK

4. Persuade the Pension Protection Fund to invest 25 per cent (£25bn) in UK infrastructure, equities and fast-growing companies

5. Invest £1bn in a fund-of-funds for venture capital funds to increase the size of investments that investees can raise.

Item 1 is another kind of tax break to encourage R&D. Item 2 aims to remedy a market failure in the provision of information in order to remedy a market failure in the provision of capital in order to remedy a market failure in the substantive development of tech firms in the UK. While both are arguably useful, they are also both also marginal, comfortably neoliberal and far-from market.

Items 3 and 4 feel like the last throw of the neoliberal dice. After failing for 40 years to get other people to invest their money in R&D, the idea now is to incentivise (item 3) or persuade (item 4) pension funds to do the investing. Since the proposal is not to enforce investment in the UK over the objections of the pension funds, in neither case is the investment decision in fact in the gift of the government. This is an example of what I mean by superficial; this is not actually an implementable plan to solve the problem.

Leaning on pension funds is a line of thinking originally developed by the Conservatives several years ago and reveals the enduring cross-party influence of Thatcher’s shibboleth. Presumably, the TBI does not see pensioners’ savings as a kind of free hit, a treasure trove that can be squandered on wild gambles. Rather, it believes that the UK investments it is proposing will yield a reasonable return for policyholders. That being so, why not simply invest the government’s own money in the same way? And if the government is not prepared to invest its own money in this way, why should pensioners embrace duff investments?

The now fashionable notion that pension funds will be the saviour of the UK’s industrial base overlooks the fact that the idea is fraught with risk. It is certainly possible to look at UK pension funds and notice that they are investing little in fast-growing domestic firms. But what they are doing is buying enormous quantities of gilts and in this way financing the national debt. Force them to buy UK shares and they may stop buying gilts. That is a fiscal reason for caution. At the same time, much of the public has been duped. The so-called ‘defined contribution pensions’ they have taken out are not pensions in the traditional sense of the word but investments, and many of them are among the worst investments on the planet. Consequently, these products are not going to provide for their retirement in the way many policyholders expect. That is a political reason for caution.

Item 5, published in October, is the idea of investing government money in a fund-of-funds that invests in the productive economy through venture capital funds focussed on tech. And yes, this does break the shibboleth against support for near-market research. It exploits the flourishing VC scene that has emerged in the UK in recent years and thus avoids the main problems associated with picking winners. In this scheme, neither civil servants nor ministers decide which companies are ultimately to receive investment; that’s left to the experts in the VC firms.

The trouble is that the TBI’s proposal is hopelessly under-powered. With a total budget of £1bn, it amounts to less than 1 per cent of existing government spending on industrial strategy over the course of Starmer’s first term. Consequently, it doesn’t raise the ceiling on the amount of funding scale-ups can access enough to stem the flow of these firms out of the country. It is aiming for funding rounds only measured in the tens of millions, a scale that is already achievable. This isn’t a break with the status quo, it’s a wrinkle.

Out of the shadow

In the hope of getting something constructive from this debate at the policy level, I have boiled down the difference in outlook between the TBI and me to concrete proposals for what should be in the innovation portion of the industrial strategy next year. These aren’t everything but they make explicit the core issues at stake in dealing with new rather than long-established firms, ideas and products. They crystallise what I think is required to step out of Thatcher’s shadow.

The industrial strategy is an opportunity to embed a long-term and co-ordinated approach in an appropriate institutional framework to enable learning, consistency and investment for the future. A suitably strategic response to this in the innovation portion would, I think, start with a clear statement that a top-level objective is to retain through to maturity more fast-growing firms in the UK so that we get the jobs that come with maturity. This needs to be accompanied by an analysis of the character and scale of the problem and proposals that are practical, within the gift of the government and commensurate in scale with the size of the problem. This opens up a dauntingly wide canvas in which the UK state has little experience, but that’s what happens when 40 years of a single dogmatic ideology comes crashing down in failure.

To this end, and others, the government needs two new investment funds backed with substantial capital. I will call these two funds Scale UK and Advance UK. The obvious people to run both funds are experienced venture capitalists, but that aside they are very different beasts.

The purpose of Scale UK is to allocate government capital to fast-growing firms through the mechanism of external venture capital funds. It would fund more start-ups and scale them faster and bigger, aiming in due course for funding rounds in the hundreds of millions. This is a completely generic project that aims to accelerate productivity and growth across the entire economy, wherever the VCs can find good prospects. Thus it can be compared to tax credits for R&D for small firms and grants from Innovate UK, and in a second term might substitute for both; tax credits have already been abandoned in the US. I have in the past advocated for both tax credits and the Technology Strategy Board (as Innovate UK was) and played a small part in bringing them into existence and keeping them going when they were threatened. This is intrinsically better than both; better than tax credits because it is discriminating; better than Innovate UK grants because it engages properly with the market questions. It is only possible now because of the growth of VC in the UK since Brown’s time and the presence of substantial spare capacity there; it is a uniquely British, uniquely 2020s option.

Instead of costing the Treasury billions a year, Scale UK should return billions in profit. In a time of straightened finances, this gives it a unique attraction. Another unique plus is that, because it relies on a tried and tested mechanism, it can be implemented and scaled quickly. Another is that the presence of private investors in the same funds is a kind of protection against bad investments. It is however an intrinsically hands-off kind of idea. There is plenty of evidence, as reflected in the difficulties encountered by the Northern Gritstone fund recently, that imposing additional criteria, for example geographic or sectoral, makes such an approach less effective because it obliges VCs to turn away from many of their best prospects.

Instead of working through external VCs, Advance UK invests in companies itself. It is intrinsically hands-on in that it is steered at a strategic level by the government’s priorities. It is thus targeted rather than generic and lacks all the uniquenesses of the first fund. It can be considered a light form of central planning and is needed to support near- and in-market activity, including research, in areas that the government identifies as important. This importance will arise from the desire to cultivate certain industries in the UK. That desire might stem, as with AI or the Laboratory of Biodesign, from technological opportunity, or from the drive for Net Zero, or the drive to advance sub-sectors in which the UK has a comparative advantage, or from geostrategic considerations, or a combination of all four.

Over to you, TBI.

---

*From a comment by the historian John Agar, it seems the year that Nicholson went to Thatcher may actually have been 1986.