CPP Lecture: Behavioural Economics and Public Policy

This report was prepared by Joe Gladstone

Behavioural Economics and Public Policy

Report on the lecture by Professor Cass Sunstein (Harvard) on 22 March 2013

A recording of the lecture can be viewed here

We were greeted with a stark choice. Imagine that you had kidney failure and needed a donor organ immediately. Would you prefer to live in Austria, where organ donors make up 99.98% of the population? Or instead would you prefer to live across the border in Germany, where the donor rate stands at just 12%?

It is not difficult to imagine in which country you would be more likely to receive the lifesaving kidney you needed, but what is the reason for this dramatic disparity in donation rates, between these two very similar countries? The answer is that Austria has a system whereby all citizens are by default registered as an organ donor, and people can choose easily to opt-out, while Germany has a different system, where people have to actively register to be a donor, which most people fail to do.

This comparison provides an example of the power of ‘default’ options, where people tend to accept whatever is the status quo choice provided for them. Academics in the field of Behavioural Economics have long argued that human decisions are strongly pulled towards whatever option is set as the default. Cambridge Public Policy was privileged to present Cass Sunstein, author of Nudge, and Harvard Professor, to discuss such default choices with us in greater detail.

The talk provided many pertinent examples of when defaults showed strong effects on behaviour, with clear implications for a wide range of public policy problems, from health care, to consumer protection, energy use, savings, and more. Importantly, the talk also addressed why defaults options worked as they do.

One such explanation was the general inertia which drives human decision making, with Professor Sunstein explaining that people have a bias for the status quo, as they choose to put off or reject making decisions which require effort. Illustrating this point, Professor Sunstein told the audience how in television scheduling, stations exploit this behaviour (and rely on it for 40% of their profits!), by placing less appealing programmes after more popular ones, relying on the fact that people will not incur the ‘cost’ of switching channels.

A second reason for the power of defaults was that they act as a kind of implicit endorsement for whatever is the default choice. Meaning that people believe that whichever choice has been set as the default is the one recommended by those with greater information, for example you may believe that experts have chosen the best default option for you when enrolling on a pension scheme at work.

The final reason given was the idea of “loss aversion” to a reference point. With much behavioural evidence suggesting that people dislike losses far more than they like corresponding gains. In an illustrative example Professor Sunstein outlined research on the role of economic incentives in teaching, where evidence has previously suggested money does not seem to improve pupil performance. However, in one recent study, loss aversion was employed with dramatic results. Teachers were given money up-front, and then told that if students did not show real improvements, the teachers would have to give the money back. The result was a significant increase in test scores. The underlying idea being that people will work hard to avoid losses, and what is perceived as a loss can be framed by the ‘choice architect’ (in this case, the researchers). It seems that often, deviations from the default choice may be seen as a loss which must be avoided.

Despite the many advantages of using defaults to influence human decisions for the better, there are problems with this approach. Critics argue that it takes away individual liberty, and that a better alternative would be to use an ‘active choosing’ method, where people are forced to make a binary decision, therefore ensuring that the choice is the right one for the individual in question. For example, rather than setting as the default for everyone to become an organ donor, instead they could be forced to opt in-or-out of this when applying for a driver’s licence or a bank account. Professor Sunstein acknowledged that this may be a superior approach in several situations, particularly where the ‘choice architect’ (e.g. the government) may be ill informed about what is best for the decision maker, or if the population has highly diverse needs. However, active choosing can also impose high costs as people may feel too busy to be bombarded with forced choices about seemingly irrelevant questions.

To overcome these problems, Professor Sunstein offered a powerful new extension to the default method which he suggested public policy makers could use to influence the choices and behaviours of citizens – the personalised default. A personalised default is where the choice architect bases the default upon personal information it holds about the decision maker e.g. their past choices, demographics, or similar information. So one group of people may be provided with one default, and another group a different default would be used. The great promise of personalised default rules is that they might eliminate the problems associated with standard, impersonal ones, without imposing the costs of active choosing. As default rules become more personalised, the advantages of active choosing begin to diminish, because personalised approaches can handle the problem of heterogeneity without requiring people to act at all.

Professor Sunstein’s central argument was that in many domains, personalized default rules could be the wave of the future, and offer the promise of large social benefits. As information accumulates about people’s actual choices, many private and public institutions will be in a position to provide personalized default rules in coming years.

Following the lecture, a lively discussion ensued in the little time that was left, where the audience questioned the speaker about several areas, including the difference between choosing as a citizen as opposed to as a consumer. One audience member also wondered about the accuracy of such ‘personalised information’, as providers which use similar methods currently, such as Google’s personalised adverts, had clearly designated her as a ‘teenage boy’ on her work computer, due to her unusual viewing habits. Such lively conversation and anecdotes continued after at the well-attended drinks reception.

Banner image from Magnus D on Flickr