It's hopefully uncontroversial to argue that government policy should be made based on evidence. That is to say, we want the people making important decisions to be informed on the issue they are deciding on, and to understand the impact of the different options they have. Working my way through Damian McBride's new book then, is a scary read. In among the (admittedly hugely entertaining) gossip and intrigue from the man in Gordon Brown's inner circle, there are some depressing examples of exactly the opposite of what evidence-based policy making should be all about.

The first took place during the fuel protests in 2000, when truckers blocked the roads in protest at the price of petrol. McBride, who was still working as a civil servant at HM Customs at the time, explains what Gordon Brown asked him to produce:

Now what we need for the day before the PBR is a twenty-page report, full of charts, making the principled scientific, environmental and economic case for cutting 3p off low-sulphur petrol and diesel, and explaining how everyone will benefit. I’m told you can do that for us.’ ‘Erm, yep, sure, I think so.’ He looked at me. It was the ‘This is what we’re actually doing’ moment. ‘This is very, very important. We’re not going to let people say we’re cutting duty because The Sun told us to or some truckers blocked the roads. Otherwise they’ll just do it again. We’re doing it because it’s the right thing to do and it’s good for the environment. And I’m asking you to write the report because I’m told you know all this stuff. Are you with me?

Location 499 in the Kindle Edition of the book

Obviously it is right that Brown should have wanted to end the protests as quickly as possible - and a cut in the tax on petrol may have been the best method of producing this, but what is worrying is that in the process of attempting to mitigate the political price of meeting the protest's demands, McBride was asked to find scientific, environmental and economic evidence to, er, sex-up (to use a phrase very much associated with the Labour government) the case for making the cut - leaving out the evidence against. This breaks one of the big rules of evidence-based politics: it can't be based on evidence if you're only cherry-picking the data that supports your argument.

Ultimately Gordon Brown had already decided what he wanted to do. Whatever this report says is true doesn't even matter - if the data has been cherry picked, then it doesn't follow that the the environment would in reality be any better off. Brown had picked the facts to suit his existing opinion.

Another worrying example of dodgy policy formation comes later in the book when McBride talks about needing a news story to give to the press at the end of a trip to India. Having already used up all of the pre-planned announcement, McBride is forced to improvise:

Some spin-doctors might have conjured up an apology for the Amritsar massacre, or a new policy on visas for Indian students, but I was not such a man. The Indian cricket team had just beaten the Aussies in Perth to end record-equalling run of consecutive Test wins and I sold Gordon on the idea that he should propose reviving the tradition whereby Commonwealth cricketers – such as Sachin Tendulkar – could be nominated for knighthoods, like Sir Don Bradman or Sir Gary Sobers. I say that I ‘sold’ Gordon on the idea. More accurately, I told him that I’d already briefed it, it had gone down pretty well and I’d written him a briefing note on the subject which he should read in case any of the hacks asked about it.

Location 3153 on the Kindle Edition

So let's think about this. Damian McBride, who in this context is not an expert on India, or the honours system, nor elected - but is basically just some bloke - made up a new government proposal and pre-briefed the press. While this is a fairly fluffy policy, and the proposal doesn't seem to have gone anywhere (though it was reported), imagine if instead he'd come up with a policy off-the-cuff about something that actually mattered? Say, the aforementioned Amritsar apology that would impact upon British diplomacy or the Indian student visas, which would impact immigration policy. Policy - or at very least the framing of the debates in which policy is decided - would be changed by just some bloke, regardless of evidence, expertise or even electoral accountability.

Finally, a third example of this scant regard for evidence comes from when McBride explains how he fought the Tories, in a tale that shows how easy it is to be consumed by the 'game' of politics, and not by the substance of the issues (y'know, the stuff that politicians always pretend to care more about).

Back in 2006, George Osborne took a trip to Japan and boldly announced a new Tory policy to build a futuristic Magnetic Levitation train here in Britain. Here's "Mad Dog" McBride explaining how he successfully managed to 'kill' the policy:

I spent the morning online researching and distributing to journalists the history of accidents and fires on mag-lev trains, and established the fact it wouldn’t even have time to get up to top speed on the route Osborne was proposing. One journalist told me that Osborne texted him and said: ‘What’s going on with this story? Why has everyone got so down on it?’ The journalist replied: ‘You’ve just met the dog'.

Location 2688 on the Kindle Edition

There are presumably lots of valid arguments on why MagLev trains are a bad idea. I imagine transport policy experts and engineers have some very strong feelings on whether or not magnets would be a better solution for the proposed High-Speed 2. But don't forget the important thing here: McBride isn't an engineer, an economist, an expert on transport policy, or MagLev trains. He's just some bloke. And rather than let the policy live or die by the scrutiny of people who actually know what they're talking about, what could for all we know have been a revolutionary and brilliant idea - was killed dead in exchange for a few less favourable column inches in the next morning's papers.

So why is this important? I think McBride's account of his time working in government sheds a light on some of the most damaging aspects of politics and the policy making process. As a whole, the book serves to underscore just how much of a stitch-up the lobby system is - with political journalists and McBride trading stories and getting rather too close to each other, and more broadly it shows just how obsessed modern politics is by the media, and what the next day's front pages will look like.

More depressingly, I worry that this all comes at the expense of good policy making. If we (as a society) aspire to live in a world where policy is made based on evidence, then the debate needs to happen in public, with expertise and data playing a key role. The culture that McBride describes surely makes this impossible - if the evidence is just another tool with which you can beat the opposition, then it erodes evidence as something that should be valued. And if politicians start calling an afternoon on Wikipedia cherry picking arguments to suit their agenda 'evidence', then this is only going to destroy trust in what 'evidence' actually is - and thus the ability to actually make things better a whole lot more difficult.

James O'Malley is editor of The Pod Delusion