The Chilcot Report and Brexit provided stark reminders of how politicians and civil servants will sometimes use whatever evidence they can find to justify decisions that have already been made. We need a genuine commitment to evidence-based decision-making, from top to bottom.

For the last 20 years, the UK government has advocated for evidence-based policy-making at home and abroad. It has developed systems to ensure that civil servants commission and assess research-based evidence, as well as consulting with relevant stakeholders, as part of the process of developing policies.

But even with these systems in place, we’re still not always getting it right.

The recent Chilcot Report into the Iraq war identified a deliberate exaggeration of the threat posed by the Iraqi regime, based on flawed information produced by the intelligence services, which was not sufficiently challenged by the Ministry of Defence and the then Prime Minister

And the so-called ‘dodgy dossier’ used to justify the UK’s position on Iraq came just five years after the UK government made a commitment to more evidence-based policy-making in its Modernising Government White Paper.

Brexit gave yet more examples of how evidence can be manipulated. Campaigners on both sides presented what they claimed was rigorous evidence supporting their arguments, some of which was later discredited and abandoned.

RAPID’s research shows that the use of research-based evidence generally isn’t driven by how rigorous that evidence is. The key factors are nearly always people and politics and the incentives that drive them.

Embedding evidence-based approaches in organisations, government departments and political parties takes a long time. It needs long-term commitment from managers and staff throughout the system.

In our advisory work, we’re frequently called in by organisations that want to add a veneer of evidence-based policy-making but not change the fundamental systems that make it difficult for their staff to make systematic use of existing evidence and commission additional research.

Where organisations have made substantial improvements in the use of evidence in policies and policy processes, it’s usually down to highly committed individuals who have risen to positions in their organisations or governments where they can really make a difference, and have established systems, incentives and recruited committed staff to make the systems work.

So the challenge now isn’t to identify approaches and tools that can help decision makers make better use of evidence; we’ve already done this. It’s to build a real commitment to evidence-based policy-making throughout the system.


The Chilcot Report showed that it’s not just the quality of the evidence that counts; the quality of the processes through which evidence is interpreted and used is equally important – especially when dealing with uncertainty.

In the evidence community, we talk about ensuring that evidence is credible (technically adequate), legitimate (recognising divergent views and beliefs) and salient (relevant to the decisions in hand). I add a fourth characteristic: evidence needs to be reliable, so that it can contribute to future decisions and not just focus on the specifics of the here and now. 

But the recent Chilcot Report showed that it’s not just the quality of the evidence that counts. You can have the best evidence in the world, but someone, somewhere has to exercise judgement about how that evidence should be used – what should be done. Therefore, the quality of the processes through which evidence is interpreted and used is equally important. 

In the run-up to the Iraq War, Chilcot found that the UK’s cabinet and ministerial committees were bypassed, ‘reducing the scope for informed collective political judgement’. The fact that some meetings were not recorded meant that there was no opportunity for others to digest the basis for the decisions. ‘Huge weight was placed on intelligence assessments that went unchallenged’, says The Economist, because key people were left out of the decision process.   

Credible policy-making processes consider evidence about people’s beliefs and values, not just statistics, scientific or economic data that many people don’t understand or engage with. Legitimate processes understand whose voices count in debates, and put in place the checks and balances to ensure that weaker voices are heard too. Salient processes actively seek out all the evidence relevant to a decision, not just what technical experts provide. And reliable decision-making processes ensure that decisions are taken with regard to evidence around long-term sustainability, not just short-term political gains.

What’s galling about the Chilcot Report is that the UK government knows how to do this. Government responses to other major inquiries – such as into the devastation caused by the Foot and Mouth epidemic – set up structures and processes to improve evidence-informed policy-making.  For example, networks of Chief Scientific Advisers now advise ministers through cabinet committees. 

The UK’s next big decisions will be around Brexit, which has global ramifications. As negotiations begin, the government faces an extraordinarily complex set of political pressures. And, as Brexit plays out, political judgements will be scrutinised like never before.  In holding the government to account, we need to pay careful attention not just to the evidence used in making those judgements, but to the quality of the decision-making processes put in place.

After Brexit, it seems clear that there are two kinds of facts. 

There are policy facts: when a government wants to achieve a particular objective, and facts and evidence can suggest different ways to do it. Many of us here at ODI spend most of our time on this kind of fact.

When providing advice and recommendations to governments, we assume that politicians care about the truth, and will listen to those who have it. If you know you want to keep dry in the rain, then advice about the relative costs and benefits of anoraks and umbrellas is welcome and useful.

Then there are political facts. The notorious  – and false – ‘£350 million a week’ touted in the Britain’s EU referendum campaign was one of these. So is the Oxfam figure about the people on the bus controlling half the world’s wealth. So are the figures about the amount of money that Africa loses to tax havens. They are facts that change how people see the world and what they think is important; facts that speak to emotion, not reason. 

They are often less certain and reliable than policy facts – more dependent on the assumptions of the people who put them together, usually with an explicitly political purpose. They are frequently challenged by the opposing side in the debate. 

But politicians use them because they work, and people believe them because they resonate with their beliefs and values, and they make a difference.  This is not about raincoats and anoraks – this is about making the political weather. 

In a debate around Brexit, or an issue like migration, the kind of policy facts ODI deals in don’t seem to change things.  We hope they will – we talk about the ‘weight of evidence’ as if, by piling up an ever larger number of policy facts, we can somehow convert them into the kind of political facts that change the debate. 

But hundreds of pages of excellent evidence on the risks of leaving the EU didn’t, in the end, outweigh the single ‘£350 million’ number. Study after study on the benefits of migration can’t cancel out the entrenched view that there are too many people coming in. 

Partly this is about the audiences. Policy makers and experts hear policy facts; political facts are for the general public – the people whose votes determine the political context for the experts to function. And, as we established so clearly during the referendum campaign, the people who want political facts are pretty unimpressed by policy facts.

Here at ODI we care about facts, of all kinds, but we also care about influence. Staying in our comfort zone of policy facts is easy and safe, but if we only talk to the policy making elites who use them, it limits our influence on some of the things we care about most. But moving over to political fact territory – and widening our audience – risks losing our hard-won reputation for even-handedness and rigour.

What we do about that is a live issue in ODI right now. What would you do?