Support 100 years of independent journalism.

  1. Politics
20 March 2015

Politicians don’t know what works, so we need to experiment

Politicians have feet of clay, too, and policies don't always work like they should. We need a political culture that allows experimentation - and the humility to accept when we get things wrong, too.

By jonathan Breckon

In case you hadn’t notice it’s election time.    Politicians are making big promises in speeches, manifestos, and on Twitter. A future is being mapped out based on certainty with claims that they know how to fix Britain. But here’s the thing: they don’t really know what will work. There’s a dishonesty in us all colluding into thinking we have the answers. A better approach would be to commit to experimentation so that we try out new policies first. This is just what has happened in Finland, which also has a general election coming up in April, where three of their main political parties have formally committed to experiments in their manifestos.

If we want effective public services, we need an experimental, learning government – robustly and systematically testing things out, measuring them and growing what works.  And, importantly, dropping policies that fail.

Without experimentation we will stagnate. And we can’t afford business as usual in the face of the immense challenges to adapt to ageing populations, climate change and fiscal pressures. In business, experimentation is commonplace. Google claims to have run 12,000 randomised experiments in just one year, with about 10 percent of these leading to business changes.

The advantage of an experimental approach is that you can be ‘innovative and cautious at the same time…try things out in an overly tentative manner’, according to an influential 1980s US position statement on social policy experimentation, funded by the MacArthur Foundation.  You can test your new idea and, if it fails, it’s only on a limited scale that limits damage. But if the innovation succeeds, you can extend it to the wider world.   Trials by HMRC and the Behavioural Insights Team on the most successful ways to encourage people to submit their tax returns, brought forward over £200 million in additional tax revenue to HMRC, the UK’s tax authority.

But experimentation should not be haphazard. Too often new public policies are rolled out with little evaluation.  In effect, governments experiment on the whole population at once – but without learning. In the UK, £66 billion worth of government projects have no plans to evaluate their impact, according to a National Audit Office report.  It is unethical to experiment on us in this arbitrary way, where we will not learn if they are doing more harm than good.  We need the best of social science to understand if – and how – policies are working, from Randomised Controlled Trials to other research, as set out in a new report by the Alliance for Useful Evidence today.

Sign up for The New Statesman’s newsletters Tick the boxes of the newsletters you would like to receive. Quick and essential guide to domestic and global politics from the New Statesman's politics team. The best of the New Statesman, delivered to your inbox every weekday morning. The New Statesman’s global affairs newsletter, every Monday and Friday. A handy, three-minute glance at the week ahead in companies, markets, regulation and investment, landing in your inbox every Monday morning. Our weekly culture newsletter – from books and art to pop culture and memes – sent every Friday. A weekly round-up of some of the best articles featured in the most recent issue of the New Statesman, sent each Saturday. A weekly dig into the New Statesman’s archive of over 100 years of stellar and influential journalism, sent each Wednesday. Sign up to receive information regarding NS events, subscription offers & product updates.
I consent to New Statesman Media Group collecting my details provided via this form in accordance with the Privacy Policy

In some situations, experimentation is unavoidable. During the post-Saddam counter-insurgency in Iraq, General David Petraeus and Major HR McMaster allowed troops on the ground to experiment.  Partly it was because they had no choice: Donald Rumsfeld’s command-and-control approach was spectacularly failing. Deaths of Iraqis and Americans were mounting on a weekly basis. But in Mosul or Tal Afar, junior US army personnel would come up with a hypothesis for tackling the conflict in areas. Test it out.  Drop it if failed, or expand it if it worked.

Of course it didn’t solve anything long term, and Iraq remains a mess. But it was dramatically more effective in limiting the carnage at the time.  And a much better approach than central-government-knows-best. It’s always a good idea to listen to the voices on the ground – be they soldiers, teachers, nurses, social workers.

This sort of promising approach can be seen with a government-backed scheme asking schools to suggest innovative approaches to teaching disadvantaged pupils.   Schools and teachers can apply to a £4 million research scheme called ‘Closing the gap: test and learn’ and be part of a randomized controlled trial. The best approaches come ‘bottom up’ from teachers.  Around 700 schools are taking part, testing out specialist maths support by teaching assistants, or allowing other teachers to review how you are teaching.  The ones that look effective from the experiment will be replicated across the country.

We also now have institutional support for experimental government, through the What Works Centres co–funded by the Economic and Social Research Council and the UK, Welsh and Scottish governments, covering everything from local economic growth to crime reduction. One of these centres, the Education Endowment Foundation has funded 77 Randomised Controlled Trials, working in over 4,000 schools, such as a trial on benefit of short bursts of physical activity on academic outcomes in English and maths, or using text messages to increase the involvement of parents in their child’s education.

Experimentation and piloting must be about being open to change.  Many governments pilot new approaches, but with little intention of changing course. Pilots can be a way of pushing controversial policies ‘into the long grass’.  Or simply part of narrative of a successful policy roll-out.  The Coalition’s troubled Universal Credit programme was piloted in 12 local authorities, and the welfare reforms developed an admirable ‘test and learn’ approach . But the Whitehall reformers  were not really open to admitting to failures, such as writing off  tens of millions of pounds of IT investment.  As the ex-head of the civil service, Sir Bob Kerslake, said there was a “culture of good news”, whatever the data was telling them. The man that fronted the flagship reforms,  Iain Duncan Smith said in an interview on Radio 4 Today programme, ‘I have a belief I am right’, despite the UK Statistics Authority rebuking him for claims made about the numbers of people getting back into work.  To learn and adapt, we must be open about failures.  We need a ‘spirit of experimentation, unburdened by promises of success’, according to the late Professor Sir Roger Jowell, author of a 2003 Cabinet Office report ‘Trying it Out’.
But it’s not just the politicians who are to blame. We as voters unrealistically demand perfect 20:20 policy vision from our political masters.  Government is not allowed to fail, or do U-turns, or to test out new ideas. This makes it hard for adjustments, reversals and tweaks.  When the dust has settled after the election, we would all do well to have a bit of humility.  Allow those in power to try new things out, and not destroy with media vitriol if their best laid plans fail, as long as they learn and move on to make the best public services, to benefit us all.