Context counts

In our search for universal truths we should bear in mind the importance of the context

Nobel prizes don’t typically cause much commotion outside their respective scientific domains. Occasionally though, something about the award resonates more widely, perhaps because some people in the media believe it is of particular interest to their readers, viewers or listeners.

So it was when Richard Thaler was awarded the prize for economics on 9 October. Only people who were on a different planet last week will be unaware that Professor Thaler is a behavioural economist. Pretty much every self-respecting newspaper and news programme around the world ran a profile of the man. And local experts were wheeled out to explain what behavioural economics is all about, with stories of (mostly) boosted retirement savings and healthier eating.

You may have been left with the idea that there has been a revolution in the staid old dismal science of Economics. But we should keep a sense of proportion. Thaler himself is rather modest about the status of his niche. At the end of Misbehaving, his very readable autobiography, he sees Behavioural Economics eventually disappear: “All economics will be as behavioural as it needs to be.”

Yet that didn’t stop (and probably won’t stop for a long time) semi-informed journalists going on about the quasi miraculous insights of behavioural economics that have the potential to transform our lives and our businesses. Watch out for articles the title of which consists of a claim and the phrase “according to behavioural economics”.  Be critical of what they proclaim. Better still: just skip them. Such headlines are rarely the sign of a nuanced discussion.

A bit of a crisis

Uncritical treatment of behavioural economics (and more general behavioural science) findings is not confined to the popular press, though. One of the most widely discussed instances is that of the so called priming effect, referring to the phenomenon in which people’s behaviour is influenced by subconscious cues.

In 1996, John Bargh, a psychologist (then) at New York University, and colleagues conducted an experiment in which they purported to activate an ‘elderly stereotype’ in a group of undergraduate students. In the treatment condition the participants had to unscramble sentences which contained words associated with old age, such as worried, old, lonely, grey, retired, helpless and so on. In the control group, participants had the same task, but with sentences that contained no such words. What the experimenters found is that, after the experiment, participants in the treatment group walked down the hallway a second more slowly than those in the control group.

Fifteen years later, Daniel Kahneman called this study an “instant classic” in his book Thinking Fast and Slow. However, not long after, doubts were being raised by other researchers who failed to replicate the effect. (Kahneman acknowledges the problem in a note he wrote as the validity of Bargh’s and other priming studies was being challenged.)


And don’t forget to pay! (photo: Newcastle University)

The same fate has befallen other findings that caused a big stir at first. Do images of money make people act more selfishly? Does a poster with Big Brother eyes really make people behave more honestly? Does adopting a power pose really boost your confidence? Certainly not always. By 2015, a large replication study by Brian Nosek and 269 colleagues, found that the conclusions of over half of 100 recent psychology studies they investigated were dubious. A true replication crisis.

One explanation for the initial lack of critical evaluation is quite likely confirmation bias, which affects scientists as well as ordinary mortals. (In his email, Kahneman admits to being ‘a general believer’ in priming.)

Context matters

But another possibility is that the context is not fully taken into account. And context matters – also for nudges, the brainchild of Richard Thaler and his co-author Cass Sunstein.

In 2015, a group of Dutch scientists led by René van Wijk of the university of Wageningen, investigated whether they could nudge supermarket customers to buy (healthier) wholewheat bread instead of white bread by giving it a more prominent display position. However, they found that no more whole grain loaves were sold as a result of their manipulation. Making the healthy option more accessible had no discernible effect.

More recently, a field experiment in India by Ling Bai of Edinburgh University and colleagues, evaluated the use of so-called commitment devices to increase preventive doctor visits for patients with high blood pressure. The commitment contract consisted of prepayment for three appointments over six months, plus an additional (imposed or chosen) commitment amount that the patients would receive back in instalments when they kept to their appointments. The researchers found no effect on actual doctor visits or health outcomes: a substantial number of patients paid (and lost) money without experiencing any benefit.

And in the last week, Google trialled and quickly removed a nudge in their Google Maps service. They introduced a feature that showed not just the walking time between two points, but also the number of calories that would be burned in the process. The idea was ostensibly that this would remind people of the health benefits of walking, and encourage them to do so. But instead it was perceived as a crass attempt at ‘body policing’ and ‘guilt inducement’.


White bread sold out, wholemeal still available? (Photo: source)

Do these three examples prove or even suggest that nudges don’t work? Of course not. What they do show is that the context is of supreme importance, and that no amount of scientific studies can deliver a guaranteed way of achieving a certain result. In the article Nudges that fail, Cass Sunstein explores several reasons why behavioural interventions may be ineffective, from strong prior preferences, to time-limited influence. It is a plea for thoughtful analysis and intelligent experimentation, rather than blind belief.

Unfortunately this subtlety and complexity is not always mentioned by advisors – whether in business or in the popular media. When we are given simple tips, certainly when it is based on the authority of a Nobel Prize winner, it is tempting to accept it as gospel.

In our relentless quest to understand, describe, and predict the world around us, we are sometimes too quick to perceive universal, unconditional truths. Binary thinking is the enemy of progress, though. We should be careful not to accept too easily results that look cool, or that confirm our prior beliefs. We should also not dismiss insights because they fail to replicate, or because an implementation built on it is unsuccessful.

Humans are complicated beings, and if we should draw a single lesson from behavioural economics, it’s that there are no simple explanations for our behaviour.  Behavioural economics refines our understanding, but it is not a magic explanation of everything.

About koenfucius

Wisdom or koenfusion? Maybe the difference is not that big.
This entry was posted in Behavioural economics, Cognitive biases and fallacies, Economics and tagged , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s