Hello. It looks like you’re using an ad blocker that may prevent our website from working properly. To receive the best Tortoise experience possible, please make sure any blockers are switched off and refresh the page.

If you have any questions or need help, let us know at memberhelp@tortoisemedia.com

Slow Views

Dominic Cummings is mostly right about the problems, wrong about the solutions

Wednesday 14 July 2021

The former adviser’s recent criticisms of Boris Johnson and the government aren’t without merit. But the answers are closer to home than to the moon…


When Jung wrote “everything that irritates us about others can lead us to an understanding of ourselves,” he wasn’t thinking about Dominic Cummings, but he couldn’t have found a better case study. In Cummings’ 18 May testimony to the select committee investigating the government’s Covid response, he railed against a failure to consider outside expertise, a lack of transparency, harsh treatment of staff, and poor management of delivery… all of which will have had former colleagues across Whitehall raising their eyebrows. And now, after years of laying into Westminster-obsessed political commentators, he has, in true Jungian style, become what he most hates: a pundit. And not just a pundit, but one whose subscription-only Substack newsletter has built a rapid reputation for feeding juicy bubble gossip to entertained hacks.

No doubt, though, there is a serious intent behind Cummings’ blogs. He has always burned with a genuine fury about the inadequacies of our political system. It’s unfortunate that, due to the Barnard Castle incident and his unique methods of communication, the wider public will largely ignore his more substantive claims. While his shouty nerd style can make him sound like a paranoid survivalist, sending tweets about robot takeovers from his bunker, what he says should be taken seriously. This is a man responsible for one of the most successful political campaigns in British history, and who persuaded a prime minister to give him almost unprecedented access and power for an unelected adviser. Moreover, his analysis of the government’s manifold failings during the pandemic are backed up by plenty of information in the public domain.

It is now clear that the much-lauded National Pandemic Plan that was in place when Covid hit was entirely unsuited to this particular pandemic. Yet the entire policy infrastructure continued to back this plan long after anyone armed with footage of Lombardy’s hospitals and a calculator could see it wasn’t viable. The groupthink that meant no one reassessed the situation, and the dismissal of critics outside the system, led to a delay in the initial lockdown that likely cost thousands of lives. Far more egregiously, the prime minister and the chancellor (though Cummings was unsubtly circumspect about implicating Sunak) refused to heed internal and external advice to lockdown again in both September and December, causing further deaths.

Throughout the pandemic, the government has refused to be transparent about its decision-making, instead choosing to brief newspapers with tantalising hints about re-openings or bad news about the impact of new variants. At key points, this has led to confusion about objectives and rules, and been highly damaging to public health messaging.

All of these failings are ones that Cummings has been highlighting, and occasionally contributing to, throughout his career. As someone who worked with him from 2009 to 2013 (and even read his magnum opus thesis, published after he resigned as Michael Gove’s aide), I can safely say that none of the criticisms of Whitehall that he raised during his testimony or on his blog are new. It is worth considering, then, how systemic these problems are and how one might go about changing things to improve governance across the board, as well as helping us to avoid the worst consequences of future crises. It’s also worth thinking about what Cummings doesn’t say, and where his sometimes simplistic faith in small teams of geniuses misses the mark.

His three big, connected themes were all on display during the select committee hearing. First, that the political class is, on the whole, shallow, innumerate, and obsessed with appearances rather than the details of policy. Second, and partly as a result of this, politicians have not ensured that they have the requisite levers of control to drive change, and those that try are hamstrung by a Whitehall system that actively tries to maintain the status quo and is hopelessly lacking in transparency. Third, and due to both of these factors, the civil service machinery is uniquely bad at delivering large-scale programmes.

It is hard to dispute the first of these arguments. While there are, of course, plenty of examples over the years of excellent ministers and astute commentators, the general trend is not a positive one. As party political engagement has gradually decreased from its post-war peak, the small handfuls of activists who now tend to pick MPs are not the group you would want to select elite talent. The conventions of our system then require prime ministers to pick their executive from this limited pool. Cummings has said that he tried to get Johnson to appoint expert ministers and put them in the Lords, something which would have been seen as highly undemocratic (not least since we don’t directly elect prime ministers). It would also be actively unhelpful to a prime minister who is highly reliant on loyalty and the ability to hand out patronage.

On top of this talent deficit, the dynamics of 24-hour political coverage reward ministers who are good at defending a line on television and coming up with neat soundbites and gimmicks. Likewise, commentators are rewarded for being entertaining and vociferous rather than right.

There is also plenty of truth in Cummings’ second claim that Whitehall is not well designed for paradigmatic reform. Most civil servants would, rightly, take exception to the idea that they are deliberately trying to foil government initiatives. Nevertheless, they are also typically quick to acknowledge systemic flaws such as the need to move to a new policy area, and often a new ministry, in order to get promoted, or challenges with engaging outside expertise. There have been multiple attempts over the years, some more successful than others, to strengthen the quality of advice and analysis, though nothing as radical as Cummings’ suggestion that all appointments be thrown open to external applicants.

But, ultimately, the first problem – weak incentives to prioritise good policy – leads to the second. The reason the civil service operates in the way that it does is because it suits politicians. If they really wanted civil servants to have deep knowledge in a subject area, or to engage with outside expertise, I suspect that many permanent secretaries would be keen to support that. But that’s not what they want. They’re usually looking for civil servants who are good at finding ways to incorporate existing manifesto commitments and headline-grabbing stunts into something that vaguely resembles an implementable policy. Genuine evidence-based critique, either internal or external, is seen as disloyal and unhelpful.

Take, for instance, the recent Department of Education policy on banning mobile phones in schools. In order to give this some semblance of reality, civil servants have had to produce a consultation document proposing some non-statutory changes to guidance on phones. The purpose of the exercise was not to change anything in the real world – most schools already ban the use of phones during the school day, and have for years – but for Gavin Williamson and his spads to generate tough-sounding headlines. Why on earth would you want to hire top policy experts or expand capacity for external advice when that’s your objective? The last thing you want is people pointing to the evidence.

Of course Cummings has engaged in this behaviour when it has suited him to do so. Back in 2019, Tory critics of the aggressive approach to Brexit negotiations were brutally expelled from the party. No space for thoughtful and informed debate there. And the many organisations who raised concerns about various elements of the proposed deal were dismissed as remoaners and anti-democratic traitors. Cummings is at his best when he’s campaigning, and that requires a ruthless ability to polarise regardless of the truth. It’s not hard to see the lessons that politicians have learnt from the Vote Leave campaign and how they actively work against the expert-based politics Cummings says he wants.

The same applies to the government’s lack of transparency. Cummings forced the DfE to go to court to prevent details of the first new round of free schools being published, and was prepared to do the same to stop his own private work-related messages being subject to Freedom of Information requests (he seems to have unilaterally decided that Boris Johnson’s don’t require any protection). And he did this for the same reason the rest of government prefers opacity – being transparent requires ceding control and giving information to those who might wish to use it against you. This is why the government has consistently refused, despite claiming Covid policy is driven by data not dates, to give any targets or thresholds at which restrictions would be reduced or increased.

So it’s undoubtedly true that the centre of government isn’t designed for maximum effectiveness, but it is designed for what most politicians want: control over the narrative and information. It is nevertheless possible for ministers and advisers who do want to make a serious contribution to do so, as long as it stays largely under the radar. Where I think Cummings is wrong – or at least much too simplistic – is in his third claim that the inadequacies of Whitehall are entirely responsible for our structural problems in delivery and implementation. After all, weak executives with bad incentives exist in many countries. The UK government is unique in the way we try, and often fail, to deliver large-scale programmes.

Essentially, there are three ways in which central government delivers programmes. It can ask some form of locally elected body to deliver it; contract a private company; or set up a central agency that reports to them. In practice, those central bodies tend to end up using private companies a lot of the time anyway.

Over the past few decades, we have moved from being fairly centralised in the way we deliver public services to being exceptionally centralised. 91p of every £1 collected in tax is controlled by central government. The vast majority of local council spending, like school budgets, adult social care and housing benefit, is tightly controlled by central government. The less directed element of central funding for councils has been cut to ribbons in the last decade, falling almost 40 per cent in real terms. Moreover, councils can only raise council tax, their other main source of revenue, by two per cent a year – and unlike, say, NHS trusts, they are not allowed to run deficits.

With such restrictions on action, and with so little money to spend, it’s hardly surprising that working in local government no longer feels high-status. Over the decades, its capacity and expertise has reduced to the point where it no longer represents an attractive option for the delivery of new or redesigned policy, even if central government overcame its predilection for control. This makes our government and its central agencies very dependent on contracting out to deliver services.

We now spend £284 billion a year, about a third of all public expenditure, with external suppliers. Far more than is spent by local government. Four ministries spend more than half their entire budgets with private providers, with the Ministry of Justice leading the way. This is not intrinsically a bad thing: some contracting-out has led to more efficient services at lower cost. But the government tends to automatically opt for contracting, without considering whether it’s the best course of action. Where there’s a strong competitive market in a particular area and it’s easy to measure whether they’ve been successful – rubbish collection, or producing vaccines, being obvious examples – then contracting out can work well. But where the delivery question is a complex one, with no simple metrics and a market consisting of the euphemistically named “strategic suppliers” like Serco and Capita, whose business model is simply to collect government contracts across a wide range of sectors, it can fail disastrously.

Most of the biggest public policy messes of the last few decades have come via contracting failures, from the London underground Metronet fiasco through various mad PFI contracts to the total car crash of the probation service privatisation, reversed last year. We saw it again during the pandemic, with the failure of Serco’s central test and trace programme, which has remarkably been re-contracted to the same company, and future disasters are still being seeded. To take one recent example, the DfE decided in May to contract out their flagship national tutoring service to a Dutch HR company that has no record of education in the UK.

Cummings regularly rails against procurement rules, but it’s not the rules that are the problem. It’s the fact that we’re so dependent on procurement in the first place and that, in many situations, the only bidders are a small number of substandard organisations.

A serious attempt to fix public sector delivery would start with a new set of principles about when to use the private sector and when to use public agencies or local government. And an overhaul of the government’s contract management practices. It would also look to create more options for delivery by reforming local government. The current patchwork of upper and lower tier authorities, mayoral combined authorities and local enterprise partnerships badly need to be rationalised into a coherent structure that has genuine responsibility for the economy in their area and the capacity to deliver complex and integrated programmes. This is also a prerequisite for any serious attempt to “level up” the country outside of London.

Sadly, this is much less exciting than Dom’s desire to recreate the moon landings. But these are the core questions that define how well the British state can deliver. Indeed the Cummings vision – small groups of eccentrics and Nasa-style oddballs running everything from a Downing Street control centre – would only exacerbate the current problems caused by over-centralisation and refusal to cede power. From speaking to civil services, there’s a widespread view that his own determination to keep departments and agencies, let alone local government, out of the pandemic decision-making loop contributed to the problems they faced.

He’s right, though, in his contempt for the current political class – and it’s hard to see where the will comes from for a serious rethink of how government works. We can’t, as Cummings wants, take it out of the control of elected ministers, but we do need more voices like his revealing the poverty of ambition of the ones we have. Revolutions can only start when a light is shone on the rottenness of the political class.