BY OLIVER BARTRUM, 27/11/2016
How can we encourage people to save more? How can we make sure that taxes are paid on time? How can we encourage lending in the sluggish period following a recession? Economists and policymakers have long wrestled with such questions. Recent research in behavioural economics claims that insights from psychology can, to some extent, help to answer them.
Behavioural economics draws largely on laboratory or field-experiments and individual reports of subjective well-being, to observe how people behave. Interestingly, the insights it generates often contradict mainstream economic analysis. For example, experiments suggest that people do not have well-defined preferences and emphasise the social and historical embeddedness of human behaviour (Hargreaves-Heap, 2013). In Nudge, Richard Thaler & Cass Sunstein (2009) argue that evidence of ‘irrational’ behaviour observed in such experiments can be used by governments to ‘nudge’ citizens toward making better decisions. Nudge was very well-received in both academic and policymaking circles, and has led many governments and public bodies to use behavioural insights in their work. The first unit devoted explicitly to the application of behavioural economics was the Behavioural Insights Team (BIT) in the UK. We spoke to the CEO, David Halpern, to find out more about the meaning of behavioural economics. The first half of this articles focusses on ‘nudging’ as a policy tool, i.e. how it is changing economic prescription. The second half of the article considers the more general implications of behavioural insights for academic economics, i.e. for economic models and explanation.
Economic Policy & ‘Nudging’
The (BIT) was set up in 2010 by David Cameron to apply insights from behavioural economics to policy, with the dual purpose of improving public services and saving money. To achieve this the unit applies the idea of ‘nudging’ advocated by Thaler & Sunstein, among others.
So what is a ‘nudge’? A nudge is a change in the design of a choice (“choice architecture”) that alters human behaviour, without actually coercing individuals down one path or another. The idea is abstract, but is simpler once illustrated with examples. Let us return to the question at the beginning of the article: how can we make people pay their tax faster? Halpern and his team turned to research in psychology that emphasises the role of social norms. By simply adding a sentence on the letters that went out to late payers, informing them that most people in their towns had already paid their tax, the payment rates increased by 15%. The ‘nudge’ appealed to our human desire to conform, but did not actually coerce the late payers into doing anything.
Halpern is keen to point out that nudges can be applied to bigger-picture questions as well. In 2010, in the aftermath of the financial crisis, the Bank of England had pursued an aggressive quantitative easing policy, yet cash & liquidity were still not reaching many small businesses, such as small builders and tradesmen. One of the psychological problems that the BIT identified was a lack of trust between small tradesmen and big banks. This lack of trust arguably flowed both ways. The solution offered up by Halpern and his team was to find someone that these builders did trust (and vice versa), so they joined forces with the Department for Business, Innovation & Skills to encourage and enable trade hire centres and lumber yards to lend to small tradesmen. In this case, the trust stemmed by the pre-existing relationship between the two parties led to a larger flow of money to reach small businesses. Schemes such as this show how behavioural economics can be applied to find solutions to problems that a neoclassical approach would not be able to deal with.
In general, the changes that Halpern and his team bring about are marginal, yet such changes are implemented in the hope that their sum will be substantial. This reflects a wider trend in policymaking that Dr. Halpern is keen to develop further – the idea of radical incrementalism. While often policymakers and businesses attempt to achieve great results through large-scale strategies, radical incrementalism recognises that there are a huge number of small changes available that – if you work them out on a compound basis – will run into millions of different options.
Dr. Halpern describes how many of us are unconsciously exposed to radical incrementalism in our everyday lives. When visiting Amazon, for example, what you are in fact seeing is one of many possible pages. Amazon are perpetually testing variations to see which leads to more clickthroughs or sales. Such an approach is incremental in that it favours continuously testing a large number of small variations. Conversely, similar strategies are rarely employed by public service. For instance, official communications from the government are equal for everyone, or do not display any systematic or meaningful variation – in the same way in which schools teach maths slightly differently but never really test which method is the best and for whom.
Why is this approach ‘radical’? Halpern thinks that “it’s radical and pretty exciting [because] once you acknowledge and embrace the idea that you don’t have to do just one thing, and you can test variations and implement this in every little aspect of what you do – over time it is utterly transformative.” Despite all this, there is still a long way to go in terms of implementing this approach across Whitehall, Halpern believes. One of the reasons why this may be the case is that policymakers are reluctant to apply methods that may not produce immediate results, as the best solution is often unknown.
If, however, Halpern succeeds in making radical incrementalism a widespread approach to policymaking in government, the potential for change is great. As well as producing more evidence-based policy, it introduces an element of humility (a recognition of what we don’t know) that has arguably been missing from conventional politics for too long. In his book (2015), Halpern claims that this wider experimental shift “will probably be the most important legacy of the quirky empiricism that BIT brought to the heart of British government in 2010, and is now spreading through the world”.
Halpern is clearly – and justifiably – proud of the BIT’s achievements in terms of its impact on society and the government’s budget, and there is in fact compelling evidence that the BIT has brought about significant positive change in many areas of public policy – we have only provided the two examples on tax & lending so far (see Halpern  for an exhaustive list). Furthermore, the changes that the BIT has enacted have been relatively inexpensive: the marginal cost of, say, adding a line of text to a government letter is essentially zero. While the costs are low, many of the projects have very large payoffs, as shown by the tax letters example. In fact, the sunset clause set at the creation of the BIT – that the initiative would have been shut down if there had not been a 10-fold return on cost in two years – was easily met. Furthermore, the innovative team at the BIT and its links with academia (especially Chicago University) mean that it can still produce cutting-edge work while government departments build their own in-house behavioural capabilities. The international impact of the BIT is also significant: many foreign governments (in Australia, for example) have worked to create their own behavioural teams, inspired by the BIT’s success.
Nevertheless, ‘nudgers’ are very aware of the potentially negative applications of behavioural insights. This line of criticism will be familiar to those who have engaged with public choice literature – how can we trust politicians to use this power wisely? Thaler outlines three principles for the use of ‘good’ nudges: their use should be transparent; they should be easy to opt out of; and there should be good reason to believe that they are welfare-enhancing (Thaler, 2015). Halpern (2015) sets out how the BIT does this in practice. First, they always tell the truth. So, in the tax letters example, if the BIT decided to appeal to social norms by saying “9 out of 10 in your town have already paid tax”, this 90% figure would have to be verifiably accurate. Second, the BIT is transparent and monitors public concern. The BIT regularly publishes open research, and hosts a blog on its website. This follows Thaler & Sunstein’s (2009) endorsement of Rawls’ (1999) publicity principle – the idea that government must be able to defend publicly any ‘nudge’ that it implements. The third principle is that robust checks & balances must be in place. For example, the BIT has set up a formal & independent ethics & trial protocol clearance process, assessed by independent academics. Inevitably, some will remain unconvinced. However, Halpern (2015) contends that we can ensure that ‘nudges’ are used for good so long as we follow a transparent and evidence-based approach (using trials). Moreover, he argues, in the presence of increasingly systematic use of behavioural insights by private businesses (some of which he terms ‘behavioural predators’), governments have a duty to respond.
Economic Theory & Explanation
What does the emergence of behavioural economics mean for the future of economics as an academic subject? The idea that the models on which a lot of policy is based are unrealistic is what Halpern says motivated him to apply his knowledge of social psychology to politics and policy. However, he doesn’t think that the standard models are completely defunct – he argues that “there are a fair amount of times when [standard models] work well. Just like Newtonian mechanics works pretty well in many situations, but it’s only in certain circumstances that you start to see that it’s not quite right”. He also points out that, while many economists today will acknowledge that people do not behave like Homo Economicus, it is generally accepted that standard models describe the behaviour of large groups and their interaction through markets relatively accurately, i.e. we behave rationally ‘as a population’. Sometimes this approach works but, crucially, sometimes it does not, leading to ‘behaviourally based market failures’. Halpern gives two examples. The first is the UK energy market. Why haven’t half the population switched their energy suppliers over the last decade? This is certainly not the kind of utility-maximising behaviour that standard microeconomics would predict. The second relates to the issue of impotent fiscal levers in relation to central banking. In Japan, for example, they have negative interest rates and are still not able to reach demand. In these situations, it is tempting to ask ‘what’s wrong with these people?!’. ‘Well’, Halpern says, ‘maybe they’re just people and the problem is with the models!’.
This kind of view is espoused by, among others, Akerlof and Shiller in their book Animal Spirits (2010), a term first coined by eminent economist J.M. Keynes (2007). Akerlof & Shiller argue that conventional economic theory assigns too much weight to the role of reason in economic decision making, and too little to the role of irrational emotional and psychological factors. These factors could explain the failure of fiscal levers in Japan. Following such a framework, they call upon economists and policymakers to drop the false theory of rational, efficient, and self-correcting markets. There is, however, still quite some way to go in terms of distilling the ‘animal spirits’ concepts and being able to incorporate them into formal macroeconomic models (Schwartz, 2010). Others take a different view. Shaun Hargreaves-Heap (2013), for example, argues that behavioural economics is indeed important for explaining individual behaviour, but its implications for the explanation of social phenomena such as market behaviour are far more ambiguous. His point is that, while behavioural insights are highly significant at the individual level because of the doubt they cast over the rational choice model, their implications for macroeconomic models are more slight. The argument goes that for something to constitute an ‘explanation’ of some social phenomenon, it must merely generate the behaviours that need to be explained. Deep recessions can be explained in this way by ‘models with suitable additions [or ‘frictions’] to rational choice behaviours’, i.e. without incorporating behavioural insights. While this is perhaps a methodologically controversial argument, Hargreaves-Heap makes the point that, if augmented rational-choice models can ‘explain’ behaviour in this sense, then ‘why complicate matters by requiring the model to be descriptively accurate?’. Indeed, for a model of a nation’s economy to be truly descriptively accurate down to the finest detail, this may well require a literally incomprehensible wealth of information.
It is clear, then, that there is no definite conclusion about what behavioural insights mean for the explanation of socio-economic phenomena such as unemployment and recessions. There is a long way to go before we have fully-functional behavioural macro models, if in fact this is possible or desirable. In our view, this is one of the most exciting aspects of behavioural insights. The experimental results are not normative in themselves. They feed into existing debates and create new ones, as we seek to understand and explain the world around us. Debates over behavioural insights and the closely related wellbeing agenda (discussed at length in our Dossier) mean that, at this moment in time, being an economics student is a genuinely exciting experience. In ten year’s time, will undergraduates still be taught rational choice theory and models such as the IS-LM-BP as fundamentals of micro- and macro-economics, respectively? Or will we witness a ‘behavioural revolution’, in which cognitive biases and the like inspire a total rethink of mainstream economics? Sure, behavioural economics currently has incredible momentum. But with such a lively debate still in its infancy, who can truly claim to be sure of the future?
Akerlof, G.A. and Shiller, R.J. (2010) Animal spirits: How human psychology drives the economy, and why it matters for global capitalism. Princeton, NJ, United States: Princeton University Press.
Halpern, D. (2015). Inside the Nudge Unit: How small changes can make a big difference. Random House.
Hargreaves-Heap, S.P. (2013). ‘What is the meaning of behavioural economics?’. Cambridge Journal of Economics, 37(5), pp. 985-1000.
Keynes, J.M. (2007) The general theory of employment, interest, and money. Basingstoke, Hampshire: Palgrave Macmillan for the Royal Economic Society.
Rawls, J. (1999). A Theory of Justice. 2nd edn. Cambridge, MA: Belknap Press of Harvard University Press.
Thaler, R.H. (2015). ‘The power of Nudges, for good and bad’. The New York Times. Available at: http://www.nytimes.com/2015/11/01/upshot/the-power-of-nudges-for-good-and-bad.html?_r=0 (Accessed: 3 September 2016).
Thaler, R.H. and Sunstein, C.R. (2009) Nudge: Improving decisions about health, wealth, and happiness. London: Penguin Group.
Schwartz, H. (2010). ‘Does Akerlof and Shiller’s Animal Spirits provide a helpful new approach for macroeconomics?’. The Journal of Socio-Economics, 39(2), pp.150-154.