Behavioural Science, Rationality and Public Policy

By Emmanuel Lee


Economists, like all other social scientists, study human behaviour. Their specialism is to examine behaviour in markets, where individuals exchange goods and services.

One key way they do this is through deductive reasoning; they take certain general assumptions about behaviour and from there make logical conclusions given different conditions and constraints.

For example: if we assume that an individual wants to keep as much money as possible, in a situation where the price of a product they want goes up, they will logically look to buy a cheaper alternative. This approach draws from rational choice theory; so-called ‘rational’ because it posits that individuals make decisions which maximise their ‘utility’ in line with their preferences.

The advantages and disadvantages of this approach have been subject to debate for as long as this form of economics has existed, but a significant pivot came in the mid-20th century when several academics felt that rational choice models failed to capture certain behaviours relevant to public policy.

These academics, such as recent Nobel laureate Richard Thaler, started drawing upon findings in psychology to demonstrate situations in which people do not behave in a way that is predicted by rational choice models. This still emerging area of study is known as ‘behavioural economics’ (a slightly confusing name if you consider that all economics is fundamentally about behaviour) and it is swiftly becoming a key part of the economics syllabus.

It appears intuitive to conclude that if an individual is not behaving in the way predicted by a rational choice model, they must therefore be behaving irrationally. Indeed, Dan Ariely, one of the key figures in behavioural economics, wrote a book called ‘Predictably Irrational’, exploring the circumstances in which individuals systematically behave in ways that appear foolish or unusual.

There are many well-established criticisms of rational choice theory; but rather than focus too closely on these, I want to draw attention to how the associated language of rationality and irrationality threatens to downplay, or even exclude, other academic disciplines from the realm of public policy.

When it comes to psychology specifically, the growth of behavioural economics has come as a doubled-edged sword. On the one hand, it has catapulted psychology to the forefront of public policy. Its students are increasingly able to apply their knowledge and skills working in the various ‘behavioural insight’ teams popping up across the public sector. It has also reinforced its importance in other areas such as business, respectively exemplified by the growth of behavioural science consultancies.

On the other hand, as its output is increasingly attributed to behavioural economics, there’s a risk that it will lose recognition as a discipline with its own application to public policy.

Daniel Kahneman himself has said that, for this reason, ‘behavioural economics’ is an inappropriate label for the type of work carried out by these organisations (he offers ‘applied behavioural science’ a catch-all alternative). In the same text, he also refers to the closely associated technique of ‘nudging’, as mainly an achievement of applied social and cognitive psychology, rather than behavioural economics.

Moreover, the behavioural economics moniker reflects a mistaken idea that psychology is simply a useful bolt-on to economics; there to fill in the occasional gaps where rational choice models are unable to explain behaviour. The language of rationality and irrationality perpetuates this idea: economics will help us explain common rational behaviour, psychology will help us to explain uncommon irrational behaviour and supposedly, as these disciplines develop, we will one day be able to explain all behaviour.

Not only is this problematic for psychology, which deserves to be treated as an important and applicable discipline in its own right, but it also leaves no room for the many other disciplines which examine human behaviour. Sociology, to use just one example, has made many important contributions to public policy, but rarely engages in the language of rationality.

The second point is that the rational and irrational labels carry implicit assumptions of normativity: in policy making terms they imply a distinction between behaviour that is desirable and behaviour that is undesirable. Identifying desired behaviours necessary to achieve policy aims is a fundamental part of policymaking, and if economics is the study of rational (and thus desired) behaviour, that automatically gives it an elevated status in the realm of public policy. However, you don’t need to scratch too far beneath the surface of these terms to see that these normative assumptions, in many cases, can be challenged.

The psychologist Gerd Gigerenzer is well known for his research looking at why decisions that appear logically incoherent should not necessarily be considered irrational. Indeed, many economists themselves have conducted research that demonstrates how irrationality, as it is conceived of here, does not necessarily equate to behaviour that is undesirable. Some of the best examples of this come from game theory experiments, where economists examine the behaviour of individuals in strategic settings. And there’s a popular (sort of) real-world version of one of these experiments that can help bring this point to life: the infamous ’split’ or ’steal’ game which is used in gameshows such as Goldenballs and Love Island.

As viewers of these shows will know, the premise of the game is as follows: Two contestants, usually at the final stage of the game show, have an option to ‘split’ or ‘steal’ the prize money, and they have to reveal their decision simultaneously. If the contestants both choose split they share the money. If one contestant steals and the other splits, the stealer takes all the money. If both steal neither take any money.

In economics, maximising utility is often equated with maximising material gain, in which case the economically rational decision in this game is to always steal. Given the fact that if the other person steals you get no prize money regardless, you may as well also steal in the prospect of maximising your potential winnings (in case they do decide to split).

Of course, as viewers of these shows will also know, the contestants often choose to split. And yet, it would seem counter-intuitive to label this behaviour is irrational. They don’t decide to split because of a lack of key information or some unconscious internal bias. They split because the outcome they desire isn’t the one which entails the most material gain.

In fact, through many different variations of these experiments, economists have attempted to show that we desire outcomes which reflect motivations of altruism, reciprocity, ethical commitments, aversion to inequity and so on. The economist Samuel Bowles considers this extensively in his book ‘The Moral Economy: Why Good Incentives Are No Substitute for Good Citizens’ and concludes that far from being undesirable, these are behaviours which policy makers, in many instances, should be encouraging.

But the crucial point that we return to here is that these other-regarding behaviours are the very behaviours that are often examined through the lens of the non-economic social sciences. And if we end up thinking of these behaviours as irrational, we risk downplaying their role in public policy. It’s important to qualify at this point that many economists dismiss the label of irrationality. In a recent tweet, Richard Thaler said he hated the word because it leads to pointless arguments. However, the discourse seeps beyond academic circles and the language of irrationality is picked up by practitioners and the media alike. Indeed, Thaler’s tweet was in response to a Business Insider feature on the ‘irrational’ behaviour surrounding the purchasing of lottery tickets. It’s clear that there is still a conceptual confusion in this space and this poses a risk to effective policy making, particularly where it involves assumptions of normativity that go unchecked.

Some may also contest that this is a small question of language; that the labels we ascribe to certain concepts are, in reality, not particularly important. But as behavioural economists and psychologists will no doubt counter, the framing of language can have a significant impact on the decisions we make. And whilst policymakers are increasingly thinking about how language influences the decisions of those for whom they make policy, perhaps there is value in thinking about how it influences their own decisions too.

Emmanuel Lee
Emmanuel Lee is a policy advisor working in the UK Civil Service. Part of his work involves looking at how behavioural science and other emerging fields can be used to increase the effectiveness of government policies.
Emmanuel Lee
Emmanuel Lee

Latest posts by Emmanuel Lee (see all)

By | 2018-02-09T11:39:52+00:00 November 24th, 2017|

One Comment

  1. John Brady December 5, 2017 at 4:42 pm - Reply

    This is very timely and apt. I hope it receives wide dissemination and attention.
    I invite your interest in a list of cognitive biases that I compiled? I made a poster of that list, and how and why we tend to rely on those biases, along with a brief description of each bias. It certainly gets conversations going…

    200+ Cognitive Biases

Leave A Comment

Send this to friend