Politics eats the evidence-base for breakfast

The idea and fetishizing of evidence-based policy is something we’re interested in at Arbitrary Constant. Here we wonder how many countries there are and what this might mean for Personal Health Budgets; here we explore what some of the biases and heuristics of evidence-based policy might be.

A news article about a lorry speed limit change (from 40mph to 50mph on single-track roads), an evidence-based impact assessment and competing interests was therefore bound to pique our interest.

The Daily Telegraph reported at the end of October:

The Government [has] pressed ahead with plans to raise the speed limit for lorries despite being warned of a likely increase in road deaths because it benefits the haulage industry

The information is generated by the government’s own impact assessment, the topline details of which are as follows:

  • There are between 60 and 80 fatal accidents involving HGVs on relevant roads, of which an estimated 18 per year take place at between 36mph and 44mph
  • Vehicles that travel between 36-44mph will be influenced by the increased speed limit – driving at an average of between 2.5-3.9mph faster
  • This increase would result in 2.6-3.5 more deaths per year
  • The potential benefit of reduced accidents from less overtaking is not included because there isn’t “sufficient confidence” it would happen
  • On the benefits side, hauliers will save time (worth £13.8m), reduce costs (£2.5m) and government will gain more fuel duty revenue (£2.1m).

Far from bringing certainty to the situation, the evidence base has put us in a precarious position, hasn’t it? We can see this in two main ways.

  1. An evidence base has been put together and a policy position derived from it. Whether the evidence base is robust I don’t know, but it clearly involves some assumptions, parameters and interpretations that could be used, if someone were so inclined, to question the conclusions drawn.
  1. The evidence base says the change in policy will be good for one group (hauliers and government) to the tune of around £18m. At the same time it also says the change in policy won’t be good for another group, i.e. the approx. 3 additional people who would die because of the speed limit increase.

Nevertheless, the relevant Minister has pressed ahead with the increase in the speed limit.

To me, this is a clear demonstration of how it matters to what end, i.e. policy, means, i.e. the evidence base, are put. It reaffirms not only that evidence-based policy isn’t rational, predictable or benefit-maximising, but that it also doesn’t happen in a vacuum.

Policy nearly always means politics, and – to adapt a phrase – politics eats the evidence-base for breakfast.


Published by


Man of letters & numbers; also occasionally of action. Husband to NTW. Dad of three. Friendly geek.

5 thoughts on “Politics eats the evidence-base for breakfast”

  1. Interesting criticism. But is it correct to describe this process as “evidence based policy”? Rather than “reaffirming…[that] evidence-based policy isn’t rational, predictable or benefit-maximising”, this example highlights for me, that ministers aren’t doing evidence based policy at all, since this would require the decision was based on the evidence, which this one does not seem to have been.

    1. Thanks Tom. For me it highlights the evidence swings both ways, and its politics the vast majority of the time that chooses which side it comes down on.

  2. In practical terms the challenge is in finding a balance between policy based evidence (let’s do a quick study to prove this thing does what we want it to) and genuine evidence based policy (let’s never do anything until there’s all the systematic reviews on all the things).

    In this case there’s a simple trade off; three lives are worth £18m, there’s maybe a Green Book algorithm somewhere which makes it a bit easier (I remember an old Home Office thing which put a murder at £5m, so that’d probably cost out OK.). The tools of policy making are blunt, but the processes are desperately complex. I don’t think there’s an easy trade off there, despite what many on both sides of the debate would like to claim.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s