Charlie Wilson, 15 Sep 2015
Behavioural science, meet public policy. The Behavioural Insights Team (BIT) was set up in 2010 within the Cabinet Office to design policies that help people make better decisions, and to use randomised controlled trials wherever possible to evaluate whether those policies work. By 2014, with a portfolio of trials and interventions under its belt, and increasing influence over policy thinking in the UK and abroad, BIT was spun off as a social-purpose mutual, owned by the Cabinet Office, NESTA, and its employees. It now works with a wide range of government and practitioner clients, and has spawned numerous analogues in Australia and Singapore, the World Bank and the European Commission, and elsewhere.
BIT recently celebrated its achievements and strategised for the future in a conference of the stars: Daniel Kahneman, Richard Thaler, Steven Pinker, Rachel Glennerster, Dan Ariely, Robert Cialdini, Hal Varian, and many more [www.bx2015.org]. Civil servants great and small were in attendance, from the Head of the Civil Service and the Head of the Government Economic Service down to social researchers and policymakers from multiple ministries.
The central argument of BIT is that public policy can shape decision contexts to enable (but not force) socially-beneficial choices that are also better for the individual. The scientific basis for BIT's work lies in four decades of behavioural economics, and cognitive and social psychology examining how decision contexts shape choices, and how those choices are 'predictably irrational' . A 2008 book, Nudge, by Richard Thaler and Cass Sunstein, brought this body of work to popular attention . It also lent the BIT its nickname, the ‘Nudge Unit’.
Nudge mantra number 1 is to make it easy for people to change their behaviour . Nudging works by setting the better choice as the default, by making social norms more visible, by ensuring the messenger is trusted, by asking for commitments in public to increase staying power, and so on .
BIT has applied these approaches to policy challenges in health, employment, growth, skills, consumer markets, tax collection, crime reduction, charitable giving, international development, equality … and energy (of which more later). Their headline success story is the reworded HMRC letter which recovered over £200m of unpaid tax. New results from the last two years were summarised in a report provided to all conference attendees . Refreshingly, this report included trials that did not produce significant effects, including one in which boiler engineers delivered energy efficiency advice.
The vibe at BIT’s recent conference was assuredly upbeat. The young, bright, uniformed staff of BIT evoked a Googlesque workforce tasked with developing public policy rather than technology. The conference organisation and the quality of the talks were closer to a slick, meticulously-prepared TED Conference than the more haphazard academic norm of over-running talks and scant attention to basic communication (storytelling) skills. (Steven Pinker gave a brilliant talk about the use and misuse of language by academics that should be required viewing). The conference bookshop overflowed with the popular science of happiness, effectiveness, good decision making, and health. It was hard not to be buoyed along on the current of can-do optimism and evidence-based discovery.
So what does all this have to do with climate change?
The first morning's session asked participants for the 'world's great challenges' that BIT should tackle. Climate change was on the list, alongside poverty, migration, health. A session on the second day of the conference was dedicated to climate change. Lord Stern's opening talk set the scene for the structural transformation needed this century to limit warming to 2oC. Talks that followed gave examples of low carbon behavioural insights in action. Andy Baynes, from NEST, explained how their 'learning' thermostat, built using Apple's design principles, had demonstrated 10% energy savings in a large-scale UK trial (whose results are to be released shortly). Simon Hill, from OPower, explained how smart phone alerts linked to normative information (what your neighbours were doing) shaved 5% of peak loads in Baltimore.
Perplexingly, the session was quite poorly attended. A session on altruism and wellbeing in the same room had been packed to standing; the climate change session had half empty seats. It must have been the topic not the lure of big names: Lord Stern is as big as it gets.
Certainly energy and climate has not claimed BIT's attention as much as other policy domains. There have been few 'wow' results from energy-related trials and interventions. Moreover the size of the problem is that much bigger. The proportionate potential of behavioural insights is correspondingly smaller. This sits uneasily with the BIT success story.
In my talk, I tried to bridge from the systemic challenge of climate change mitigation to the demonstrated but marginal effects of the best-available behaviour change trials. My simple conclusion: behavioural insights are a part of the story, but only a small part.
There's lots of reasons why.
First, climate change is a structural problem in that both hydrocarbons and profligate energy use are embedded within modern life and economies. Behaviour is the surface of a problem that reaches deep into what it means to be part of the contemporary social world. Mitigating climate change is inescapably linked to values, visions, and social organisation as much as it is to behaviour, technology, and innovation.
Second, the best available tests, trials, programmes, and policies targeting energy-related behaviours manage to reduce energy demand by less than 10%, and more often by only a few percent. Evidence that these effects persist beyond the duration of the trials is often weak. To paraphrase David MacKay, the former Chief Scientific Advisor to DECC, small changes do not add up to a big difference. Mitigating climate change needs a big difference.
Third, an emphasis on behavioural insights risks sidelining other structural approaches, not least the regulations and standards that have successfully addressed air pollution from industrial facilities, fuel inefficiencies in vehicles, heat loss in new buildings, as well as the notable successes in changing behaviour related to seat belts, drink driving, and smoking. The low-cost attractiveness of nudges reinforces a form of status quo bias to operate within existing regulatory frameworks and political institutions. Mitigating climate change means reimagining these institutions.
Fourth, small emission cuts over the near-term are not always consistent with large emission cuts over the long-term. 'Path dependence' describes how future change in the energy system is shaped by its past and current trajectory. This is most evident in long-lived energy infrastructure. While near-term investments in shale gas may help reduce emissions, the enduring consequences of those investments may delay the transition to a fully renewable energy supply. Just as technological systems can be inert once a direction of travel is established, so too can social systems. Behavioural change, particularly if it relies on incentives or other extrinsic rewards, can reinforce self-enhancing values. Mitigating climate change means harnessing self-transcendent values which are widely-held but suppressed.
BIT-informed policy makes it easy for people to change their behaviour. Mitigating climate change is not easy. That doesn't mean that behavioural insights aren't relevant, interesting, and applicable. It just means they're only a small part of the story. The evident influence of BIT on public policy is a welcome counterbalancing of axiomatic economics by evidence-based behavioural science. The enshrining of trials, experiments, tests ... and failing, learning, succeeding into the public policy process is also vitally important. The energy and climate portfolio of BIT interventions should expand.
But this should not be at the expense of regulatory and market-based approaches to climate change mitigation. Nor should it crowd out the wider cast of social and behavioural scientists with wholly different perspectives on what the problem is, how it can be addressed, and how efforts should be evaluated. It's important to do what works. But we also need to understand why.
*Charlie Wilson researches energy and climate change at the Tyndall Centre for Climate Change Research at the University of East Anglia.*
. Ariely, D. (2008). Predictably Irrational: The Hidden Forces that Shape our Decisions. HarperCollins: New York, USA.
. Thaler, R. and C. R. Sunstein (2008). Nudge: Improving Decisions About Health, Wealth, and Happiness. Yale University Press: New Haven, USA.
. Service, O., M. Hallsworth, D. Halpern, F. Algate, R. Gallagher, S. Nguyen, S. Ruda & M. Sanders (2014). EAST: Four Simple Ways to Apply Behavioural Insights. BIT: London, UK.
. Dolan, P., M. Hallsworth, D. Halpern, D. King and I. Vlaev (2010). MINDSPACE: Influencing behaviour through public policy. Cabinet Office & Institute for Government: London, UK.
. The Behavioural Insights Team (2015). Update report 2013-2015. BIT: London, UK.