How realistic is it to think that robust evidence drives policy?

What does the policy community look like from the other end of the telescope? In academia, assumptions abound: that evidence speaks for itself; that policy levers are linear and in abundance; and that political will is a given if the evidence is robust in itself. Here, Stephen Meek, formerly from Whitehall and now heading up a major policy institute, probes further and paints a highly nuanced picture of the practice of policymaking. His observations will help academics refine what the challenge of policy engagement actually is. Shamit Saggar

How realistic is it to think that robust evidence drives policy?

- Stephen Meek

Having recently moved from the UK civil service to a university, with the remit to connect its research better with policymakers and the public, here are a couple of reflections on the obstacles that get in the way of policy advisors and civil servants working together effectively.

The first is that policymakers need to open up the policymaking process so people can see what shapes decisions. What is second nature to a policymaker, and so obvious it is barely worth explaining, often looks like a black box to the outside world.

In particular, it is often unclear what the relationship is between the evidence that goes in one end and the decision that comes out the other. Did evidence inform the decision? Or was it just window dressing to give the illusion of rigour? If the connection isn’t visible, then people are likely to be cynical if the relationship isn’t obvious.

Good policymaking is a complex, synthetic activity which takes the relevant and often incomplete evidence, and combines it with other relevant factors to produce a recommended course of action. These might be how does it interact with other policies? What is the cost and benefit, to government and to others? How does the possible response fit with the elected government’s priorities and values? What about public attitudes to it? Political arithmetic?

Smoking is a useful example. When the evidence of the link to cancer first emerged, you had powerful evidence of harm. But you also had a population of smokers reluctant to believe in that harm, and an industry disputing the evidence, which was also a significant employer and generator of tax. The evidence pointed to drastic action, but that would have jeopardised your chances of being elected. And if you are not elected, you can’t deliver on anything else.

The policy choice, gradually to change attitudes and behaviours, was a judgement about what was the right thing to do in this particular set of circumstances. And while only a minority of policy questions are matters of life and death, most are ultimately complex, best-fit judgements rather than simply a linear process of evidence in, action out.

That is why I think opening up the policymaking process is so important if we are to get properly evidence informed policy. The more the factors that shape and constraint decisions are visible, the better academics can tailor the way they present and contextualise the evidence and the more likely the evidence will shape the policy judgement.

The second obstacle is the truism that the reason policymakers aren’t interested in the evidence is because they want a quick answer. This works both ways – policymakers will often say there is no point asking academics because they’ll say they need five years to do the research.

In fact, policy moves glacially slowly. The big policy issues – climate change, an ageing population, tackling productivity etc – have been with us for decades. What open and close quickly are the windows for action, driven by an event or crisis that creates a demand to do something.

Policymakers often have just weeks or months to capitalise on these windows. If they leave it too long, the window will close. They need good-quality insight on which to make sensible recommendations. But while academics can always say “more research is needed”, for a policymaker, not doing something can have real-world consequences.

While it is likely to be the case that precise questions being asked has not been researched before, the existing body of evidence will have something useful to help frame an answer.

So when a policymaker asks “what does your research say I should do about x – and I need an answer in weeks” they don’t mean (even if they say they do) “do you have a peer-reviewed answer on the stocks to precisely this question?”

What the dialogue should be about is “what does the existing evidence suggest is a plausible range of things to do?” Answering this challenge is where academics can really help policymakers.


Stephen Meek

Stephen joined the University as the Inaugural Director of the Institute for Policy and Engagement in September 2018, following a career in the UK Civil Service. He held board-level posts in the Ministry for Housing, Communities and Local Government and the Department for Education, and has also had senior roles in HM Treasury, Cabinet Office and, on secondment, in the Local Government Association and working with Cambridgeshire and Peterborough Local Authorities on public service reform and devolution.

Stephen is currently chair of the University Policy Engagement Network, which brings together policy facing functions from over 30 UK Universities.

Share this

Browse by Topic

Cookies help us improve your website experience.
By using our website, you agree to our use of cookies.