July 20th, 2011
Editor’s Note: This is the second part of a two-part post discussing behavioral economics and how it is being used by British policymakers. Part 1 focused mostly on the development and general principles of behavioral economics. Part 2 below discusses some of the ways British policymakers are seeking to use insights from behavioral economics.
Behavioural economists thus recognise that people operate according to the heuristics and general observations outlined above; moreover, they have begun to design policy interventions that takes these insights into account, because, they hypothesise, this will yield more effective interventions. For instance, at the end of 2010 the Nudge Unit released a report – Applying Behavioural Insight to Health. The report, presumably the first of many that will embrace the whole range of public policy, lists a number of health policy interventions that the UK Government is either piloting or thinking to pilot.
The first of these interventions appeals to loss aversion in relation to smoking cessation. Those who wish to quit smoking are asked to voluntarily sign a contract where they lose ‘rewards’ if they fail a regular urine or blood-based smoking test. With rare exceptions, simply offering people rewards if they quit smoking has not been shown to be effective in any sustained sense; since people tend to respond to ‘losses’ more than ‘gains,’ to the extent that people view the loss of rewards as genuine losses, we might expect higher effectiveness from such a ‘commitment’ (often called a deposit contract).
The Government is collaborating with Boots, a British chain of drugstores, to develop a smoking cessation trial. A similar initiative has been developed by the Yale academics, Dean Karlan and Ian Ayres. Through a website, Karlan and Ayres offer people the opportunity to pre-commit themselves towards a behavioural target of their own choosing, and to voluntarily forgo a chosen amount of money (which they can, if they wish to strengthen the incentive, commit to a body that their find particularly obnoxious – e.g. the British National Party) if they fail to behave as they themselves wish.
Other initiatives include ‘prompted choice’ to be an organ donor, as a way of changing the reference point. In the UK, people currently have to ‘opt-in’ to be a donor, and roughly twenty percent of adults do so. In several other countries — France and Portugal, for example — people have to actively choose not to be a donor. That is to say, they are required to ‘opt out’, and, as a consequence, almost the whole of the adult population are listed as potential organ donors. Similarly, requiring employees to opt-out rather than opt-in to an employer-sponsored retirement plan has also been shown to significantly increase the likelihood that people save for their retirement).
Given the shortage of available organs in the UK, this simple change in the ‘default’ position would seem to offer a simple means by which to significantly improve the situation, although some are uncomfortable with the notion of ‘opting out’, seemingly maintaining the position that this would serve as an unnecessary obstacle to those who do not wish to be donors. As a watered down version of opting out, several US states have introduced with some success prompted choice: when people apply for a driving licence, their application will not be processed until they tick a box indicating whether they do or do not wish to be an organ donor. The UK Government will trial this system of prompted choice this year.
Using Behavioral Economic To Tackle Obesity
The authors of the report also propose a range of what they claim to be behavioural economics-informed initiatives intended to increase rates of exercise, healthy eating and medical self-monitoring. For instance, the report proposes extensive use of financial and non-financial incentives, presumably modified to have greatest expected effect, e.g. through the use of deposit contracts as discussed above for smoking cessation, or by using lottery payment mechanisms, whereby people may ‘anchor’ on the amount to be won rather than the probability of winning, or by simply making the incentive ‘fun.’ (This last strategy has no obvious behavioural economic explanation, but then, behavioural economics does not explain everything.)
For example, in the face of high rates of childhood obesity, a private sector collaboration between Bayer Healthcare and Nintendo DS offers redeemable Nintendo ‘points’ to diabetic children who agree to regular blood-sugar tests. In addition, the Nudge Unit has formed a partnership with LazyTown, a public-private initiative that has been operating nationally in Iceland since 1996, in which young children sign an ‘energy contract’ with their parents that rewards them for eating healthily, going to bed early, and being active.
Following the introduction of LazyTown, childhood levels of obesity in Iceland started to fall, but not all behavioural initiatives of this type are successful. In Hackney, for instance, researchers handed out pedometers to children in the hope that this would encourage them to walk around more. The researchers were surprised at how much walking obese children were apparently doing until they later discovered that the children had tied the pedometers to their dogs.
The authors of the report extend their ‘fun’ proposals to the idea of embedding sensors in stairs to encourage people to avoid elevators and make music with their feet (for an experiment in Stockholm, see here). Whether such an idea would make sense in, say, Holborn underground station is a question, but the general idea of trying to make exercise fun is unobjectionable, and, vis-à-vis hyperbolic discounting, is an attempt to tackle the immediate displeasure of exercising.
An additional series of proposals in relation to food intake and standards also appeal to incentives in some sense by providing people or organisations with visual prompts to encourage behaviour change. These prompts sometimes seek to alter the reference point that people attach to ‘good’ behaviour (by increasing the salience of relevant information, resonating with the successful ‘five a day’ campaign), and also sometimes appeal to the notion of public reporting of performance as a means to incentivise better practice, such as requiring restaurants to disclose their hygiene standards on their premises. This last practice, highlighted by Thaler and Sunstein, has reduced the rate of food-borne illnesses in some parts of California.
Moreover, in March 2011, the UK Government announced that restaurants and fast food outlets will be required to include calorie counts on their menus, and pubs will be asked to display alcohol units on beer mats and glasses. Further initiatives that the government is considering include calling for yellow tape to be placed across supermarket trolleys to designate a part of the trolley that the consumer might reserve for fruit and vegetables, in response to a pilot project in America that showed that this prompts consumers to buy (if not necessarily to eat) more of these items.
The current government, through the offices of the Nudge Unit, is therefore clearly considering a range of health policy-related initiatives that are broadly informed by the findings of behavioural economics. There is little doubt that more ideas, applied to other aspects of public policy, will soon be released, and if the Labour Party were back in Downing Street, I would speculate that the approach would not be abandoned.
The Ethical Considerations Of Behavioral Economics
At the same time, there is much concern around whether ‘nudging’ is manipulative — i.e. whether it facilitates the ‘nanny state.’ This charge was prominent in a November 2010 broadcast of the BBC Radio 4 programme, The Moral Maze, in which the journalist Melanie Phillips was a particularly misinformed (and somewhat shrill) participant. The charge of being potentially manipulative, however, is not entirely without foundation. There is a risk, for example, that the government might use knowledge of the findings of behavioural economics to steer people towards doing things that they do not really wish to do (e.g. local acceptance of a toxic waste dump near a poor community), and there is a broadly expressed concern, consistent with mainstream economics, that policy makers should not assume that they better know what people want than the people themselves.
However, if the aim of the government intervention is to help people behave in ways that, after a period of reflection, they genuinely want (e.g. quit smoking) or that they believe is genuinely good for society (e.g. again, to quit smoking), but face bounds on their rationality; if they are directed by heuristics that send them the wrong way, or find their decisions to be particularly sticky around pre-set default options, then designing policy to account for these latter considerations (such as prompted choice to register as a potential donor) appears legitimate and potentially beneficial. Default positions are, for instance, unavoidable, so if people are often inert in their actions, why not choose the default that it is likely to generate the most individual and societal benefit?
It is therefore important to emphasise that not all policies that can be classified under the heading of libertarian paternalism will necessarily be good (or bad) or (un)desirable. Some will surely be ineffective, whilst others, under close inspection, might be deemed to be politically or ethically questionable (e.g. the general public might well object strongly to rewarding people for doing things that most people might think they ought to be doing anyway). Each proposal should be assessed on its own merits, and we ought to guard against critics who might wish to throw the behavioural economics baby out with the bathwater.
It is also important to reemphasise that under any libertarian paternalistic proposal, people are not mandated to change their behaviour if they do not wish to do so. People can still choose not to be a donor, they do not have to quit smoking, and they can remain sedentary. The proposals ought not therefore be coercive in any strict sense of the word; rather, they should offer a gentle nudge to people to engage in behaviours that, on reflection, they would prefer to do.
This is not to say that behavioural economists oppose stricter forms of regulation, a common misunderstanding among those who criticise the potential effectiveness of the approach. No behavioural economist that I know would argue that voluntary behavioural interventions and nudges should replace, for example, compulsory seatbelt legislation, drink-drive laws, food safety legislation and the like. Rather, behavioural economics should be seen as an additional tool to complement regulation by moving society incrementally in a direction that might benefit all of us; it should only be seen as a substitute for regulation when additional enforced measures are perceived by the public as an expression of a government overstepping the mark. To quote one of the world’s leading behavioural economists, the American George Loewenstein: “For all of its insights, behavioral economics alone is not a viable alternative to the kinds of far-reaching policies we need to tackle our nation’s [indeed, any nation’s] challenges.” [text in brackets added]
In sum, then, we can hypothesise that if we want behavioural policy to have good effect, we should design it around how people actually do behave, rather than around the assumptions of standard economic theory. This is being recognised in policy circles, and behavioral economics could have an increasingly powerful effect in the formation of public and private decision-making, not just in the UK, but internationally. In the current financial climate, many of the potential policy proposals have the added advantage of being almost costless. For instance, altering a default rule through prompted choice requires imagination and a commitment to persuade sceptics that it is worth testing, but does not appear to require substantial additional financial resources for its implementation.
However, despite there being plenty of behavioral economic-informed policy ideas floating around, more ideas are needed, and far more evidence is required on their likely effectiveness.Email This Post Print This Post