Can a localist government change behaviours?

by Professor Peter John

16 May 2012

On 16 May 2012 the British Academy Policy Centre published Nudging citizens towards localism? One of the report’s authors, Peter John, talks here about the development of the report and how he reached his conclusions.

The starting point for this project was a question posed by Selina Chen, who originally commissioned the report Nudging Citizens Toward Localism? published on 16th May. Does the idea of behaviour change, which is associated with the central state using robust knowledge about what affects human behaviour, run counter to the aim of limiting the power of central government and letting citizens decide matters in any way they see fit?

This report presents a detached and neutral review of the arguments for and against the proposition, which is as it should be from an academic source and the British Academy. I did feel as if I had been set an essay question in a classic student assignment style, which I have answered with an ‘on the one hand’ and ‘on the other’, but my interviewees, to whom I am very grateful indeed, responded to this format openly, giving me many ideas on both sides of the argument.

I have spent a lot of time puzzling about this difficult, if interesting question. On the one hand I can see that the commitment to behaviour change needs a more concerted approach, which is the tenor of the recent House of Lords Select Committee report, Behaviour Change. The strong evidence that has emerged over recent decades on health, diet, the environment, energy use, volunteering and so on should be promoted strongly from the centre and could reasonably be thought to override local choice and variation.

And after all – this is England where we have generally favoured the universal principals of the welfare state, in particular equality, and efficiency, over the views of local government and other decentralised bodies. People may regret the loss of localism but accept the claimed benefits of a fairer and centrally-run state. So if government promotes a weaker central state, then if follows that the use of this evidence will not be a strong as it could be. Less will happen because of the hands off approach and the natural conservativism (small c) of many local areas.

The argument against the proposition is about promoting a virtuous circle linked to a programme of institutional reform: that decentralisation can go hand in hand with ideas and evidence about behaviour change as a form of local innovation, which happens from bottom up. Knowledge diffuses and there is a potential for learning and a variety of local practices, where people and organisations can imitate each other. There is no intellectual contradiction between behaviour change and localism because the two can assist each other.

Therefore I have been wrestling with these two conceptions. In spite of my equivocation, I became certain of several things. I came to disagree with the ‘slap on the hand’ approach of the House of Lords report – it is actually more difficult to use these ideas in the centre than might be thought. Most central policy-makers take behavioural research seriously, but in fact find much less evidence about what it is that government can do to implement reforms based on that research.  While we know that increasing the cost of alcohol will reduce consumption and thereby improve health, what we do not know is whether an intervention such as having a minimum price for a unit of alcohol will have this effect.  One point is about the general relationship between price and consumption, the other focuses on whether a particular initiative will work.  We have a lot of knowledge about the former, not the latter.

Government and other agencies need to know what will happen if they act differently: the counterfactual question. The best way to get this knowledge is from randomised controlled trials, but there just are not enough of them commissioned to make the difference. In this context, the approach of the current government seems sensible:  to have a champion at the heart of government, the Behavioural Insights Team, which works across and beyond government to promote the approach, uses the proofs from randomised controlled trials and then publicises the evidence. It simply is not true that the work of government is limited to soft nudges; the ideas of behavioural economics and the other behavioural sciences have been applied right across all activities of government, such as collecting debt or regulating energy use. I do think that there has been a shift of emphasis based on solid foundations, and this influenced my recommendation to extend the work of the Behavioural Insights Team. Of course, more should be done, and there is patchy emphasis so far, but I think there are moves to establish evidence more solidly through a NICE-like evaluation centre at the heart of government. I hope this will happen.

The other set of findings I am sure about is the state of play at the local level, which is particularly where I acknowledge the help of my co-author, Liz Richardson. It is here that many day-to-day decisions are rolled out and citizens have direct contact with public administration so can benefit from a behavioural approach. In that sense, what happens at Whitehall is far removed from local government and other local agencies and groups. And here we found lots of examples of the use of behavioural ideas – ranging across party control and happening in different sectors and parts of the country. But overall, the use of behavioural ideas has been patchy and one gets the sense that there are only a few innovators whose energy could easily fizzle out. The approach has not bedded down.

The other finding of the report is that many of the objectives of localism and the Big Society need behaviour changes to implement them. It is not obvious that citizens will do more without some set of encouragements and incentives, and there is not enough research and evidence to support initiatives that link local action to a wider reform movement. In other words, there is a danger that the potential virtuous circle of new ideas about behaviour change and local action will never happen because the links have not been thought through sufficiently.  I came to think that there was some possibility of a compromise between central policy on behaviour change and a local dimension, but that not all the moving parts of the machine had been linked together. By trusting solely in decentralisation, there is a real risk that the local action necessary for more behaviour change might not come about. This report is not just a plea for more knowledge (as is typical from an academic), but an argument for the better use of that knowledge in delivery mechanisms which both acknowledge local choice and rely on an active role for central government.

Here are my recommendations, with reference to the relevant pages in the report:

1. There is a need for better evidence on effective methods to increase citizen activity on a large scale, for example through the use of nudge-type techniques, and deliberative ‘thinks’. (See pages 17, 18-22).

2. More randomised controlled trials are needed to test out behaviour change interventions (26-7)

3. The work of the Behavioural Insights Team should be extended beyond its two-year life in 2012. (22-7)

4. The impact of the abolition of Central Office for Information needs to be evaluated. (24-5)

5. Local government policy-makers need to be nudged alongside citizens for a smooth and collaborative transfer of control to communities. (29-35)

6. The legitimacy of behaviour change policies and interventions should be based more on accountability to citizens. This requires large-scale ‘thinks’ to garner support. (10, 43-4)

7. Local implementation requires innovators and change agents to be nurtured and protected, with a higher acceptance of ‘failure’, and more space for creative experimentation than has traditionally been the case in central and local government. (29-35)

8. More power should be devolved from local government and to empower smaller organisations with lower levels of capacity than local authorities. (34-5)

Peter John is Professor of Public Policy at University College London and author of Making Policy Work and Nudge, Nudge, Think, Think.

Sign up to our email newsletters