At the end of last month Stephen Tall posted an essay entitled A liberal approach to evidence-based policy making on his blog. The essay had originally been submitted for last year’s CentreForum essay competition on The Challenges Facing Contemporary Liberalism but, for whatever reason, hadn’t appeared. I keep meaning to write a response to Stephen’s post, but I don’t seen to have managed it yet.
More to the point, I submitted an essay to the CentreForum competition but it wasn’t ever going to be published in that form. There is the possibility that I might develop a longer version of the argument. But I’m not sure when that’s going to happen. In the meantime I thought I’d follow Stephen’s lead and post it on the blog, rather than let it languish. So here it is.
Governments have traditionally relied on “shove”. When market failures result in inefficiency governments have used regulation to “shove” economic actors in directions perceived as socially beneficial. This is a relatively blunt instrument. More recently, policymakers have been persuaded of the power of fiscal incentives. Taxes, subsidies, and manufactured markets have increasingly been the preferred shoving mechanism.
Similarly, many liberals have robustly, but not uncritically, advocated choice and competition in the reform of public services. This is an attempt to capture the benefits of market mechanisms without wholesale privatisation. Some of the enthusiasm for this approach flows from a desire to combat the inefficiencies inherent in monopoly provision and the risks of government failure.
Concern about the cumulative burden of individually well-intentioned regulations is evident in the Orange Book. In addition, not only were market mechanisms within public services advocated but also manufactured markets to achieve environmental policy goals.(1) The subsequent decade has seen the embedding and extension of Better Regulation and the creation of substantial artificial markets such as the continent-wide European Emissions Trading Scheme.
Much can be said about the success, or otherwise, of these initiatives. But I want to explore the broadening of the policy agenda. Concern with market failure and government failure is longstanding; over the last decade it has been joined by a concern with “individual failure”. Indeed policy in some fields has become rather preoccupied with individual failure and “behaviour change”. The agenda poses potential challenges for liberals.
Most economic arguments in favour of the superiority of markets to allocate resources rest upon strong assumptions about consumers’ ability to make rational decisions. These assumptions have increasingly been challenged.
The 1970s saw a growing appreciation that information is vital in shaping markets. Imperfect or asymmetric information is a pervasive source of market failure. This can lead to, for example, regulations to mandate disclosure of accurate information on a standardised basis. This information puts buyer and seller on a more equal footing. Disclosure regulation may be all that is required to correct a market failure.
At the same time, however, economists and psychologists have documented an increasing range of decision-making anomalies. Consumers’ decisions regularly and systematically depart from the perfect rationality assumed by standard economic theory. A wide range of decision-making biases and heuristics have been identified. These include default bias, loss aversion, framing effects, anchoring and adjustment bias, peer group effects, and preferences that are only arbitrarily coherent. Dellavigna helpfully summarizes these departures from rationality as: nonstandard preferences, nonstandard beliefs, and nonstandard decision making.(2)
Behavioural economics, which brings economics and psychology together, is booming. A key message is that consumers are goal-directed and reasonable, but their rationality is bounded rather than perfect. Consumers economize on scarce decision-making resources by using rules of thumb and decision heuristics. This approach saves time and often works well. But sometimes it leads people to make mistakes.
The implications of these developments for policy have been recognised for quite a while.(3) But one particular approach captured the political imagination.
On econs and humans
In 2003 Richard Thaler and Cass Sunstein proposed “libertarian paternalism”.(4) They argued that social welfare would be enhanced if policy design integrated findings from behavioural economics. The approach is “paternalistic” because it aims to take advantage of decision-making biases and heuristics in order to guide people’s choices in ways that maximises their own welfare and social welfare. It is “libertarian” because it does not remove choice or alter incentives in the way that regulation or fiscal incentives do. Rather, by consciously taking advantage of biases and heuristics when offering choices, policy guides choosers to the option they would themselves have preferred if they’d spent more time thinking about it. But there is nothing to stop them choosing another option.
These ideas were picked up in policy circles almost immediately (5), but when Thaler and Sunstein published their best-selling book Nudge in 2008 interest in the idea exploded.(6) Nudge argued that much existing policy assumes people are “Econs” – the perfectly rational actors of economic theory – when it would be better to design policy that recognised people are fallibly human. Thaler and Sunstein’s key claim is that policymakers should act as “choice architects” who design choices to “nudge” people in the right direction. Importantly, they also argue that because framing effects are always at work, policy makers cannot choose between “nudging” and not influencing choice. Rather they have the choice between conscious nudging and unconsciously framing choices in ways that have unanticipated consequences.
The best-known example of nudge in action in the UK is the introduction of auto-enrolment into occupational pension schemes. This relies on the default bias. There is nothing to stop someone opting out of the scheme, but it is signalled as “the norm”, it is an effort to exit, and many won’t bother. The policy aim is for more people to reach retirement age with an occupational pension, and hence reduce the poverty rate among older people. Your retired-self will be grateful to your today-self for joining the scheme, even if today you might not have chosen to join without the nudge: an example of time-inconsistent preferences. The nudge would have no effect on those who would anyway have evaluated the benefits of joining a pension scheme and signed themselves up.
In 2010 the Coalition Government set up the Behavioural Insights Team at the heart of government. The team has examined the way in which nudge, and other behaviour change approaches, might be used in a number of policy areas including the payment of fines or taxes and engagement with job search processes.
Much of the discussion of behaviour change, including nudge, has focused on public health. Obesity and diabetes are major health challenges that are, it is argued, within the realm of individual control. Healthier food choices and increased physical activity can pay dividends for an individual’s quality of life and for society, as the usage of particular types of health service would be lower. This resonates with Thaler and Sunstein’s most frequently cited nudge example – the siting of fruit on a cafeteria counter. If fruit is positioned differently relative to chocolate bars then more customers will choose fruit and fewer will choose chocolate.
The BIT has gone beyond nudge and drawn on a range of thinking about behaviour change mechanisms, including its own research, to develop the MINDSPACE framework to assist policymakers when designing behaviour change interventions.(7) BIT was spun off from government in 2013 to become a mutual joint venture. One of its first acts in its new form was to publish the simplified EAST framework.(8)
Nudge focuses primarily on the act of deciding and the immediate context in which that decision occurs. But policy interest has been broader. In 2008 David Knott and colleagues published Achieving culture change, which recognises that as a result of broader social factors individuals can draw on differing levels of cultural capital when making decisions. If policy wishes to reshape behaviours then it may require more broad-based and long term interventions than simply tweaking the architecture of the immediate choice context.(9)
There are now several behaviour change frameworks circulating in the policy world. An even greater number of frameworks has been proposed across the academic literature.(10)
A liberal response to behaviour change
Quite a bit of the policy discussion of behaviour change takes what I describe as an ethical sidestep. It starts from the premise: given something needs to be done, how can evidence from the behaviour change literature help us design better interventions?
But for many liberals this starts in the wrong place. It rather begs the question. Should government be doing anything? Or is the whole behaviour change agenda ‘nannying’ on a grand scale?
The words of John Stuart Mill loom large:
… the only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant. He cannot be rightfully compelled to do or forbear because it will be better for him to do so, because it will make him happier, because, in the opinion of others, to do so would be wise or even right. Those are good reasons for remonstrating with him, or reasoning with him or persuading him, or entreating him, but not for compelling him …
Thaler and Sunstein are careful to argue that their approach is liberty-preserving and hence cannot have the charge of paternalism levelled against it. However, a substantial critical response has argued that their defence against the charge is inadequate. Some of the criticism is rather shrill libertarian accusations that nudge is the first step towards totalitarianism. But there is much more considered critique.(11) Nudge seeks to bypass the deliberative processes of a controlled behavioural response to achieve “socially desirable” objectives. It may not formally compel decision-makers, but because it seeks to work “with the grain” of automatic behavioural responses it is more insidious. In the long term it may undermine individual autonomy.
The argument that nudge is only aiming to help decision-makers arrive that the decision they would have made if they had deliberated more fully raises a whole host of questions. How are a decision-maker’s ‘deliberated’ preferences identified? How do we know they align with what is ‘socially desirable’? How do we know the choice architect is not subject to the same individual failures as those they are seeking to nudge?
A number of alternatives to nudge advocate for different behaviour change mechanisms. Two of the most prominent are “think” and “steer”.
Think, proposed by political scientist Peter John and his colleagues, rests upon deliberative democracy. When nudges may not deliver change, coming together in groups to understand issues and identify preferred outcomes may be effective. The approach sees goals emerging from debate, rather than the revelation of an individual’s pre-existing underlying ‘deliberated’ preferences. Think has been tested alongside nudge using randomised controlled trials, with encouraging results.(12) The challenge, though, is that think is more logistically complicated and effortful. It could only be used sparingly.
Steer emerged from the RSA’s Social Brain project. The aim is to encourage “thinking about thinking”. People are provided with information about how the brain works and makes decisions. The information takes the form of a small number of ‘principles’ – for example, “habit is king” or “when it’s difficult, just let it sit”. A better understanding of how decisions are made opens up a space for reflection and a revised approach. Steer was originally piloted on a relatively small scale. Again, it was found to have encouraging results. The RSA suggested that raising awareness of “thinking about thinking” could be a useful component of compulsory education.(13)
The framing of Nudge has a “negative liberty” feel to it – it stands in contrast to regulation as being about freedom-from constraint. This framing is itself disputed. Steer, in contrast, is much more in tune with the positive liberty of freedom-to. While some approaches to behaviour change policy may be dismissed as illiberal, others place the emphasis upon building capabilities and empowering individuals and social groups.
Much of the behaviour change literature can be criticised for acknowledging the importance of broader structural considerations in understanding policy problems but then parking them. For example, poor food choices may be one cause of obesity, but those choices need to be understood in the context of, among other things, an industrialised food production system that relies heavily on processed ingredients; a social environment saturated with aggressive marketing of foods of little nutritional value; and the declining prevalence of the skills needed to prepare food cheaply from scratch.
The rise of so-called “psychological governance”, preoccupied with behaviour change, risks being a distraction. Fixating on what happens at the cafeteria counter is to arrive at the problem too late in the day. Understanding behaviour is, of course, important, but it shouldn’t blind us to more profound structural issues nor prevent us from challenging more deeply engrained inequalities of power.(14)
(1) I am thinking in particular of chapter 5 ‘Liberal economics and social justice’ by Vince Cable and chapter 6 ‘Harnessing the market to achieve environmental goals’ by Susan Kramer.
(2) DellaVigna, S. (2009) Psychology and economics: Evidence from the field, Journal of Economic Literature, vol 47, no 2, 315-372.
(3) Jones, P. and Cullis, J. (2000) ‘Individual failure’ and the analytics of social policy, Journal of Social Policy, vol 29, no 1, 73-93.
(4) Thaler, R.H. and Sunstein, C.R. (2003) Libertarian paternalism, American Economic Review, vol 93, no 2, 175-79; Sunstein, C.R. and Thaler, R.H. (2003) Libertarian paternalism is not an oxymoron, University of Chicago Law Review, vol 70, 1159-1202.
(5) Halpern, D., Bates, C., Mulgan, G., Aldridge, S., Beales, G. and Heathfield, A. (2004) Personal responsibility and changing behaviour: The state of knowledge and its implications for public policy, London: Cabinet Office.
(6) Thaler, R.H. and Sunstein, C.R. (2008) Nudge: Improving decisions about health, wealth and happiness, New Haven, Yale.
(7) Dolan, P., Hallsworth, M., Halpern, D., King, D. and Vlaev, I. (2010) MINDSPACE: Influencing behaviour through public policy, London: Cabinet Office and Institute for Government.
(8) Behavioural Insights Team (2014) EAST: Four simple ways to apply behavioural insights, Behavioural Insights Ltd.
(9) Knott, D. with Muers, S. and Aldridge, S. (2008) Achieving culture change: A policy framework, London: Cabinet Office Strategy Unit.
(10) Michie, S., van Stralen, M.M. and West, R. (2011) The behaviour change wheel: A new method for characterising and designing behaviour change interventions, Implementation Science, vol 6, 42.
(11) The literature is extensive. It includes: Hausman, D.M. and Welch, B. (2010) Debate: to nudge or not to nudge, Journal of Political Philosophy, vol 18, 123-36; Rebonato, R. (2012) Taking liberties: a critical examination of libertarian paternalism, Basingstoke: Palgrave MacMillan; Grüne-Yanoff, T. (2012) Old wine in new casks: libertarian paternalism still violates liberal principles, Social Choice and Welfare, vol 38, 635-645.
(12) John, P., Cotterill, S., Moseley. A., Richardson, L., Smith, G., Stoker, G. and Wales, C. (2011) Nudge, nudge, think, think: Experimenting with ways to change civic behaviour, London: Bloomsbury.
(13) Grist, M. (2010) Steer: Mastering our behaviour through instinct, environment and reason, London: RSA Projects.
(14) See, for example, Jones, R., Pykett, J. and Whitehead, M. (2014) Changing behaviours: On the rise of the psychological state, Cheltenham: Edward Elgar.
Image: © studiostoks – Fotolia.com