What Dominic Cummings Gets Wrong

His ideas are worth engaging with but his worldview contains a fundamental flaw

Dominic continues to fascinate people. I get multiple requests every month to speak to journalists about him, because we worked together closely from 2009-2013. In the last month alone there have been half a dozen new profiles. And my two BBC documentary appearances to date have both been to talk about him.  

It's partly because he’s so different from the standard array of Westminster characters. He looks like a mad scientist dressed by a small child. He says things you’re not supposed to say, even about political opponents, let alone colleagues. His writing style is quirky to say the least, disregarding punctuation or grammar but mixing in memorable phrases (“there are no ninjas, there is no door”) with brutal putdowns and half-digested bits of science and philosophy.

There’s a running debate between people who see him as a total charlatan, a posing narcissist who should be completely ignored, and those who see him as eccentric, and sometimes arrogant, but also worth listening to. I’m definitely in the latter camp.

I learnt a huge amount from him about how politics works, the way to get media interested in a story, the kind of messaging that lands with low information voters and so on. And we had numerous interesting policy conversations about everything from the EU’s formation and the history of the internet to his well-rehearsed views on the failings of the British state. He’s also extremely funny in person, when in a good mood, and displays a kind of manic energy, which he tends to hide in public appearances.  

Over many conversations and reading pretty much everything he’s ever written (which takes a while) I concluded that while he is usually right about politics, explaining his success as a campaigner, he is fundamentally wrong about how to design and manage system reform.

He thinks about institutions in two ways. First in terms of individuals, whose weaknesses he is exceptionally good at analysing as long as they aren’t him. This can seem compelling – especially when the individuals are Boris Johnson or David Cameron. And can be a useful way of thinking about small organisations such as campaign groups or start-ups. For instance I would agree with Cummings’ view that had the Leave campaign been led by Eurosceptic MPs like Bernard Jenkin it would have likely lost.

His second lens for understanding organisational behaviour is bureaucratic obstruction. Essentially he assumes most people running existing institutions aim to frustrate change because they want an easy life and to maintain their powers. So, for instance, the civil service (and the EU) continually block reform because the status quo suits them.

This leads him to the conclusion that the problems with most institutions in Western democracies is that a) they are full of idiots – who in particular lack numeracy or the ability to get outside the bubble of trivia which dominates day-to-day debates – who are b) in organisations which benefit from blocking positive change. Having defined the problem in this way he then proposes solutions which involve finding a bunch of geniuses who are intrinsically motivated to do the right thing and then giving them maximum latitude, with no restrictions on their behaviour.

As I wrote in a previous piece:

“All of Cummings’ treasured examples of high-performance either come from the American military (Manhattan Project; DARPA) or single party states like Singapore or China. They are typically long-term, highly technical programmes, undertaken with no or minimal public transparency, and with the role of politician limited to signing cheques. The absence of any major social reforms from his analysis of success is something of a warning sign that what he wants is not in fact possible, certainly within the confines of British democracy.”

This worldview is neatly illustrated in a (free to read) post he wrote a few months ago on the next US election. The basic argument is:

1.       Both main parties and the media are run by fools:  “Many bad trends will deteriorate the longer [Biden] is there. If he dies the VP will be even worse. The Democratic Party is firmly in the grip of a generation of activists deranged by Ivy League insanity, BLM etc.”; “The GOP lost the plot and shows no sign of regaining it.”; “The political media is dominated by a subset of graduates who, like Oblonsky in Anna Karenina, largely cannot think for themselves and simply absorb and emit leftist political ‘views’ like clothes fashions.”

2.       US institutions are not run by politicians anyway but by self-interested institutions who have no incentive to change: “The permanent bureaucracies certainly don’t want anybody elected controlling the governments, and they don’t even run themselves themselves!... Imagine going into the Pentagon in January 2025 and starting discussions on ‘reforming the Pentagon’. How far do you think you will get? Nowhere. There will be a process. Powerpoints. Lots of smiles, flattery and civilised chats. Lots of trivial ways in which you seem to be getting your way. And lots and lots of leaks. And no progress on anything fundamental.”

3.       Therefore the only way to change anything is a small group of geniuses completely disconnected from the political process who then disappear: “A tiny and cheap (<$2-3 million) project — independent-for-now of any candidate — should start now to figure out …how they could actually control the government after they win, and who the candidate should be ….A project like this must be done away from DC...While there is a need for new institutions, this project should not be one. It should disappear in a puff of smoke, not try to justify permanent existence to donors.”

There’s nothing in his diagnosis that’s wrong exactly – though it is typically unnecessarily sweeping in its dismissal of everyone. The Democrats and, in particular, the GOP are not in a good way at the moment. US journalism does seem unhelpfully focused on identity politics rows. I can’t say I know much about the Pentagon but I don’t doubt reforming it would be hard and would meet with plenty of opposition.

The problem is he doesn’t take the next step – why are these things true? Is it just because everyone’s a fool and institutions hoard power? There is no context here at all; no analysis. He doesn’t talk about the US constitution, the design of checks and balances and the impact they’ve had. Or the problems of majoritarian voting systems and how they can lead to smart people feeling they need to say daft things. Race is only mentioned to have a pop at identity obsessed liberals, as if it there isn’t any particular reason as to why they might be obsessed with it.

These are all deep systemic issues that couldn’t be solved by a team of geniuses operating independently. If they tried they’d run into exactly the same set of problems. There is no simple, easy, revolutionary route here. A great President can make a difference, of course, but all of them have grappled with these same problems. Scrapping the Pentagon might allow you to start with a fresh organisation but it would still run into all the same issues about, e.g. intelligence collection in a democracy, as its predecessor. Moreover, there is a limited group of people with the necessary expertise on any given topic and they’ll find their way back into any new institution with the same predilections and beliefs they had before. Without a deeper analysis of the points of system failure and cultures you won’t get anywhere.

It reminded me of working with him at the DfE where most of our disputes were about removing regulations in the hope it would allow geniuses to do amazing things. I was much more interested in creating structures that allowed good schools (for instance) to support weaker schools. He just wanted to scrap everything that might get in their way.

For instance he wanted to remove pretty much all restrictions and regulation from academies. He didn’t really care if they were selective or for-profit or hired unqualified teachers. He just wanted some success metrics based on exams and if the schools passed then that’s fine. His starting point was that the system was so broken that it was worth burning down the whole thing to allow great teachers to build their own school groups free of Government restraint. As one school leader put it in a note to us at the time “if we had the head teachers required to run every school with the brilliance your reforms propose, we wouldn’t need reforms.”

It was the same with assessment. He wanted to scrap all GCSEs except for English and Maths as he could see the perverse incentives GCSEs create for schools and the amount of time they spend jumping through the hoops necessary to meet them. This was something he saw, with some justification, as holding back brilliant teaching and students who could perform high above GCSE level. But the system impact of this would be disastrous as you’d create an even stronger perverse incentives for schools to focus the entire secondary school curriculum on English and Maths. He couldn’t see the purpose of regulation was to constrain underperformers and bad actors. He only saw the negative impact it had on elites.

You can see the same tendency now in his many complaints about legal cases over PPE procurement. He knows (and I believe him) that he was acting in good faith and trying to override bureaucracy to make something critical happen faster. But the rules are there to stop people acting in bad faith (as Owen Paterson was, for instance, arguably doing on behalf of Randox).

Ultimately this is what Cummings gets wrong. Regulation, institutional norms, information transparency, processes, are more important than brilliant people. Because it is only those things that stand in the way of bad actors destroying systems. It is the current absence of these things causing America so many problems because Trump is a really bad actor. Indeed in winning the Presidency, Trump did more or less exactly what Cummings proposes in his blog (with a lot more money) because the institutional and cultural structures allowed him to do so. To a lesser degree Cummings facilitated Johnson doing the same here by encouraging the prorogation of Parliament and defenestration of Tory MPs who disagreed with his view. Johnson has always had a natural belief in his ability to get away with anything but these successes (in their own terms) can only have bolstered that belief.

Of course you can make regulation and processes sharper, and less likely to block great people from making a difference. You can change norms that lead to excess caution. It’s usually slow and boring and requires spending time analysing systems rather than burning them down. It doesn’t make for such dramatic blogs. But it works better.