Keeping behavioural science on the straight and narrow

As we see behavioural science being applied all around us, we are left with more questions than answers

Behavioural science has well and truly left obscurity and entered the mainstream. Governments – if they haven’t already done so – are following in the footsteps of the UK and setting up nudge units, and companies are using all the behavioural tricks in the box to market their wares to consumers. But is all that done responsibly? And if not, what should we do to make sure it is?

Those questions formed the backdrop of the March meeting of the London Behavioural Economics Network (LBEN). To celebrate its fifth anniversary, they had a panel of seasoned practitioners pondering the ethics and the responsible application of behavioural science.

Behavioural economics and its close cousin, behavioural science, have of course been in existence for much longer than five years. Amos Tversky and Daniel Kahneman published their famous Econometrica paper on the Prospect Theory more than 35 years ago. But for much of the time since then it all remained well outside the mainstream.

Even ten years ago, as people began to apply concepts from behavioural science, it was pretty obscure – you needed to convince your boss that you were not a lunatic, as Greg Davies, one of the panellists, remarked. But things are very different now. The popularity of the domain is rising, and with it, the suspicion that it is being abused.

Richard Thaler, another pioneer of behavioural economics and co-author of Nudge, is well aware of that:


Some people are distrustful of any nudges – good or bad – because nudges are by definition manipulative. And they have a point: nudges change people’s behaviour without their being aware of this. Yet ‘libertarian paternalists *’ maintain that as long as nudges are not sneaky, don’t restrict the choices of the people being nudged and are in their interest, things are OK.

The behaviouralists’ angst

Nonetheless the ongoing suspicion and criticism causes behavioural economists a fair amount of angst. Cass Sunstein has taken up the defence of nudges, and Pelle Guldborg Hansen has explored the criticism of nudges and attempted to nail down the definition of a nudge.

But as Leigh Caldwell, another panellist observed, one of the questions that this debate sidesteps is of course: “what do people really want?” Of course, you can ask people directly. For instance, people who struggle to manage their finances may state quite explicitly that they need help. Nudges to help them keep their spending under control could hardly be viewed as unethical.

Still, many assumptions are made that don’t stand up all that well to scrutiny. “Is it in someone’s interest to be nudged to save for their retirement rather than buy a new pair of shoes?” wondered panellist Emily Haisley. Usually the choice we would make in a ‘cold’ state of mind, free from immediate temptation, is regarded as the right one. Decisions made in the ‘hot’ state, when our impulsive System 1 is in control, are generally assumed to be the ones that we’ll regret later on. But is that really so? After several minutes of debate, the panel didn’t manage to come to a conclusion.


…and soon the chairs would be engulfed in engrossing debate!

What about the criticism that behavioural science is all about manipulation? Panellist Oli Payne dug into his own experience and responded by pointing at an inconvenient truth. Manipulation may not be in the eye of the beholder, but we certainly don’t consider all instances as equal, it seems. A prime example of consumer manipulation was the launch of Procter & Gamble’s soft drink, Sunny Delight. That was actually a distinctly downmarket product, with just 5% citrus juice in a concoction of water, sugar, vegetable oil, thickeners and additives. Nonetheless, P&G’s marketers ensured it was sold in the chiller cabinets, next to the bona fide fruit juices. It worked: consumers paid the premium price, in the belief that they were buying something healthy and worthwhile. (It didn’t last: three years later, the subterfuge had been well and truly revealed and Sunny D had fallen from grace…)


But when similar tricks are used to manipulate people to recycle more, there is hardly any moral indignation. That makes objective evaluation of what is (and is not) responsible use of behavioural ploys really quite hard, if not impossible.

Perhaps educating people better, so they are more aware of our cognitive biases, and the techniques used by others, well-meaning or not, to influence us might help? Unfortunately there is little or no evidence that this works. Even Daniel Kahneman admits to being just as prone to biases as the next person.

More questions than answers

The concern about unethical use of instruments that affect our choices and our behaviour without us realizing is real and justified. And the desire to set and enforce new ethical standards for behavioural practitioners is understandable.

Is that possible? Could we come up with solid definitions of what is, and what is not acceptable? Could we establish a body whose judgement would guard us from evil nudges? Who would guard these guards themselves?


Vintage manipulation

The panel discussion at the LBEN left the audience with more questions than answers. Yet, enforceable standards and rules rely on clear-cut, unambiguous definitions, and a pile of difficult questions is not a good starting point. With a bad regulatory framework, we risk throwing out the good nudges with the evil bathwater.

Perhaps new regulations are not really needed, anyway.

Attempts to manipulate and mislead people are not new. Yet society has managed to curtail the most egregious excesses. Yes, we have laws to tackle outright fraud, but for the fuzzy stuff we rely on a system of much more informal checks and balances – consumer organizations, the media, the social media. If we could deal with the snake oil of the late 19th century, and with the pretend orange juice of the late 20th, we should be able to deal with the ‘evil’ nudges that the 21st century will throw at us.

For the time being at least, it seems the best we can (and should) do is to remain vigilant and critical of what we see, and shout out when we observe something dubious – whatever side we are on, practitioners or simple nudgees.

All we need now is a clever nudge to help us do so.

Posted in Behavioural economics, Cognitive biases and fallacies, Ethics | Tagged , | Leave a comment

Nudging people on the escalator

(featured image: Gabriel Jorby)

How to make people stand still to go faster

My very first visit to London left such a big impression on me that I can still recall it decades later. By far the most evocative memories are those of the London Underground: the scent of ozone and lubricant, the sounds of the sliding doors, the labyrinthine access tunnels and, of course, the escalators. The stoic discipline of the Londoners, standing on the right, walking on the left – as the signs everywhere urged travellers to do – was almost as baffling to a continental kid as the neat queues at every bus stop.

That segregation goes back nearly a century. In 1921, ten years after they were first introduced, passengers were first instructed by a recorded voice to stand on the right on escalators. Since then, stand on the right has become embedded in the collective psyche. It has guided generations of passengers smoothly through the underground system.

This is a great example of how a modest heuristic is able to provide both societal and individual value. If you’re in a hurry, you’re free to rush up on the left hand side and get to the top (or the bottom) a few seconds earlier. If not, just relax on the right and let the steps take the strain.


Will this soon be history? (Photo:

Except it seems it’s not that simple. The separation between walkers and standers may be on its last legs, so to speak. Earlier this week, an article on the gadget website Gizmodo reported on a study by public transport operator Transport for London (TfL). For six months, they had been monitoring the traffic flow on the escalators of Holborn tube station – one of London’s busiest – with passengers instructed to stand on both sides. This followed a 3-week experiment in November 2015.

In both cases, the evidence suggests that there are situations where it is actually better to stand still on both sides. The Independent’s headline read “It *is* faster to stand on the escalator rather than walk”, and in a slightly tongue-in-cheek way, they were right. The data showed that, when 40-60% of the people were walking, the escalator could handle maximum of 115 passengers per minute. But when everyone was standing, that went up to 151 passengers per minute.

More space

It is not hard to see why this is the case. The amount of space a walker needs is bigger than the requirement of a stander, so the capacity is lower when people are walking. You see this also on congested motorways: the safe distance between vehicles (and so the amount of road space each vehicle needs) is larger when the speed is higher. In many countries there are signs above each lane to restrict the speed when congestion looms. When everybody moves at the same, lower speed, more vehicles get through every minute.

In addition, keeping everyone driving at the same speed means there is little point in changing lanes. And that is good too, because lane-hopping has what economists would call a significant transaction cost. If someone squeezes in front of you, that forces you to brake, and the same for the car behind you, and so on. The resulting concertina effect causes everyone else to slow down – in economics this is called an externality.

Such externalities are one reason why markets are rarely perfect – including the markets for roadspace or for transportation by escalator. Lane hoppers impose a delay on the other road users, and very similarly, people who choose to walk on a congested escalator impose a delay on everyone else.

The problem for TfL and its users is that it is only in certain situations that it’s better for everyone to stand. Not only is the old stand on the right heuristic still perfectly fine outside the peak hours, things also change according to the height of the escalator. An earlier trial at Canary Wharf station, where the height of the escalator was just 10m, found that with everybody standing capacity was reduced by 10%. (The Holborn escalator has a height of 24m.)

Heuristic versus nudge

Replacing a century-old heuristic with a new one would be quite a challenge. But even a new heuristic wouldn’t help here: a rule of thumb couldn’t really capture the complexity.

Is there another way to guide people guided towards the behaviour that is right for the location, and for the situation? How can passengers in a hurry be prevented from mindlessly pursuing the old heuristic, or from following their counterproductive instincts that cause others (and indeed themselves) delay?

The approach used to control traffic flow on motorways is not really suitable. It depends on enforcement, and fining passengers for walking when they shouldn’t would surely meet with protest. Would just telling people help?

Simply advertising what you want people to do is easy and straightforward, but it’s unlikely to have much effect here. Imagine posters explaining the conditions when walking is allowed, and when standing is de rigueur. Regular passengers would ignore the signs because they already know what they say. Occasional travellers might be intrigued by them, but both would still need to judge the height of the escalator and the congestion level to know whether they should stand on the left or are allowed to walk. Nope, the old heuristic won’t be beaten by something like that.


Not so suitable for escalators (Photo: Simon Murphy)

A dynamic eye-catching Walk/Don’t Walk sign at the start of the escalator would require a bit of technology to control it, but it might be more effective. At least it gives a visual cue when it matters. Or maybe the original technology could be reused? A looped spoken announcement saying “Walk on the left”, or “Don’t walk on the left” according to the situation would probably become annoying very quickly, however. And in both cases it would still be tempting for anyone on the left hand side, seeing half a dozen empty steps ahead, to start walking again to close the gap, of course immediately followed by everyone behind.

Seeing the light

Perhaps road traffic can provide inspiration after all. Environmental cues can influence how we behave, and traffic engineers in Japan (and elsewhere) have used this to great effect. Grooves in the so-called Melody Road produce a well-known tune you drive over them at a particular speed. And while sound might not be a good choice to influence pedestrians on an escalator, light might be a better cue.

A mechanism developed by three sports scientists, Liang Huang, Jie Zhuang and Yanxin Zhang, offers intriguing possibilities. Using a system of running lights they were able to control a subject’s walking speed in order to analyse their gait. So imagine a strip of lights on either side of an escalator. If the lights move at the same speed as the steps, people would be nudged to stand. When the situation allows for it, the lights on the left hand side could be made to move at a faster pace, thus encouraging them to walk (and help distracted passengers keep out of the way of their hurried fellow travellers).

The tube network becomes ever busier. TfL will need to come up with ways to maximize throughput in the stations during peak times, and avoid having to temporarily close them for safety reasons when they become congested.

Literally nudging people on escalators would probably be a violation of safety regulations, but a metaphorical nudge might just make travellers see the light and let go of the old heuristic.


Posted in Behavioural economics, Economics | Tagged , | Leave a comment

The largest anchor ever

(Featured image: clmper)

How do we know what is a reasonable amount of money to pay?

How much would a rational person be prepared to pay for, say, a cauliflower? They’d go down the ranking of all possible purchases, and for each one compare the utility it provides with that offered by the vegetable, until the utilities are equal. Whatever price they’d pay for these equivalent goods, they’d hand over for the cauliflower.

But we ordinary humans don’t possess the power to determine the price independently. Our willingness to pay – whether for a cauliflower, a used car, a sofa, a haircut or a holiday in the Mediterranean – is strongly influenced by what others are happy to pay. And that information comes to us via the market.

In the supermarket, it is advertised on the shelves. Conventional purchases in shops, certainly in the West, involve little negotiation. In practice, it is generally a binary decision: are we willing to pay the asking price or not? Either something is too expensive (but by how much?) or it is cheap enough (but how much more could it cost before we bail out?). So we’re generally not really aware of the maximum price we’d be willing to pay.  Perhaps surprisingly, that means we don’t know how much we truly value the things we buy.

A now classic experiment by Dan Ariely, George Loewenstein and Drazen Prelec, then at Harvard University, illustrates this very clearly. They showed 55 students six products with an average retail price of $70, without mentioning the price. The students were then asked whether or not they would be prepared to buy each good for a dollar amount equal to the last two digits of their social security number. After this response, they stated their maximum willingness-to-pay (WTP) for each item. Despite being wholly irrelevant, the last two digits of the social security number turned out to be very influential on the WTP. The table shows that, on average, people whose number ended in 80-99 were prepared to pay $37.55 for the rare wine, over three times more than those with a number ending in 00-19.

This phenomenon is known as anchoring. The first piece of information (in this case totally unrelated and arbitrary) weighs heavily on the subsequent decision-making. In practice, you can see this effect being used in situations where it is customary to negotiate.

It is telling that sellers on car boot sales or flea markets almost never advertise prices on their stall. This forces the buyer, who is used to seeing prices displayed, to ask for it. And that gives the seller the upper hand, as their ‘asking price’ will then be the anchor in the negotiation. Of course they’ll aim high, so they can concede a little to the buyer, who will immediately get the feeling they’re getting a bargain.

First mover advantage

Intuitively you would assume that being the first to put your cards on the table undermines your negotiating position. But this is not borne out by the research, as Leigh Thompson, a psychologist at Northwestern University’s Kellogg School of Management, explains. On the contrary, several investigations show a powerful first mover effect, especially if both parties are equally well prepared and have similar leverage points.


A steal at 5,000 DM!

The uncertainty around the ‘right’ price is not just on the side of the naïve party in a negotiation either. In a 1997 study by Thomas Mussweiler  and others at Würzburg University, a 10-year old car with numerous defects was shown to 60 mechanics and car dealers. They could inspect it and were then asked to estimate the selling price, but not before the researcher gave his opinion. In half the cases he told the expert “I thought the car should sell for about DM 2,800”; the other half were told the selling price was “about DM 5,000”. The average estimates of the second group turned out to be DM 1,000 higher than those of the first group.

Even jokingly demanding a ridiculously high salary ($1million) was found to influence the eventual offer in a study by Todd Thorsteinson at the University of Idaho. Certainly worth bearing in mind during your next job interview!

A huge anchor

It is hard not to see last week’s media attention for the UK’s “Brexit bill” in this light. This is the amount that the UK will need to hand over to cover outstanding liabilities when it finally leaves. And in mid-November of last year the EU was first in putting forward the amount of €60 billion ($63 billion, £51 billion at the time of writing).

It then went quiet for a while. Yet the British government didn’t take advantage of this lull to try and regain the initiative: last week the figure was back in the headlines with a vengeance.

Just like in a divorce, this will be the subject of negotiations. The €60 billion is well underpinned and defensible. But there are numerous components to the calculation, many of which are ambiguous. The UK had a chance to develop its own version yet it failed to take the initiative, not once, but twice.

Perhaps the poor preparation of Brexit in the first few months is to blame. It is not surprising that a Brexit ministry without proper accommodation, so that it reportedly had to hold meetings in the local Starbucks, doesn’t have the resources to develop its own calculations.

But maybe there is more to this than meets the eye, and the lack of counteroffer is deliberate. Arguably this scenario plays into the hands of the British government. Just like the naïve buyer on a flea market will walk away with the idea they’ve bargained well if they knock a few pounds of the seller’s anchor price, any reduction to the €60 billion can be sold at home as a victory in the negotiation.

Yet just like it is the flea market seller that has the last laugh, so it is likely to be for chief EU negotiator Michel Barnier – thanks to a €60 billion anchor, the largest anchor ever.


Posted in Behavioural economics, Cognitive biases and fallacies, Economics | Tagged , , | Leave a comment

A new challenge for behavioural economics

Can behavioural economics play a role in the process of policy-making, as well as in its implementation?

(Featured image: Unsplash)

Many of the best known examples of everyday Behavioural Economics are about health, wealth and wellbeing. Not surprisingly, these three central aspects of our life feature in the subtitle of Thaler and Sunstein’s popular book on the discipline, Nudge.

As they explain, we have many conflicting aspirations, and we’re not very good at weighing them up. We want to lose weight, but we also prefer snacking on a chocolate bar or crisps, rather than on a banana. We want to keep fit, but we also like to lie in in the morning rather than to go jogging. We look forward to a comfortable retirement and know we should save for it, but we also like spending money here and now.

A rational person would simply attribute weights to these conflicting options, and work out what, on the whole, is the optimum combination. Thaler and Sunstein call such a mythical person an ‘econ’, short for “homo economicus”. But we are not econs – we are humans, subject to all kinds of biases and temptations that trip us up and make us regret our choices, over and over again.

Thankfully, behavioural economics offers an arsenal of tricks to help us to make better choices. Nudges can make the option we think we ought to take easier than the tempting alternatives.

But it is important to realize that this is not about one choice winning and the other losing. It is not because we save more for our retirement that we stop spending money today on things we like. Neither do we give up chocolate altogether because we start eating more fruit. Behavioural Economics is not about replacing surrender to immediate temptation by surrender to austerity. It helps us find a compromise between the tempting and the reasonable that is good enough.

It is almost as if deep inside there are two separate single-minded, simplistic entities with opposing aims. Each has complete disregard for their counterpart: “I like chocolate!” versus “I want to be healthy and fit!”. And behavioural economics speaks to a third, impartial entity. This kind of inner benevolent dictator uses  tricks like putting an apple on the counter the night before to reduce the temptation to grab a bar of chocolate for our lunch bag. This seems to model how we function much better than Thaler and Sunstein’s mythical econ, in which everything is integrated in one, rational entity.

A larger scale

So we appear to be capable of recognizing the conflict between two aspirations within ourselves, and use behavioural economics to help us take a more beneficial course of action. Can this be done on a larger scale?

Setting and implementing public policy shows some interesting parallels. In public life it’s not various aspects of a single individual that needs to be reconciled, but the diverse demands and expectations of a single population. Policy related to the regulation of road traffic is a good example. None other than Cass Sunstein wrote a Bloomberg view column this week with the provocative title “No One Needs to Die on America’s Highways”. And a last week a member of the Green Party in Belgium wrote an opinion piece (in Dutch – Google Translate) pleading for cities that make it possible for young children safely to cycle to school.

Both articles take a one-sided view in which road traffic safety dominates. Such single-minded views are characteristic for campaigners, who, pretty much by definition, don’t have to worry about ‘the other hand’.


This crumple zone comes with a price tag (picture: Pixel-Mixer)

But of course there are other hands. Not only are there the petrolheads who would resist, out of principle, lowering the speed limits or indeed imposing any speed limits at all. There is also the economic reality of the burden of regulation. The safety-related features of our cars, from seat belts and crumple zones to anti-lock brakes and air bags, cost a lot of money. So does building bridges and underpasses to reduce the chance of collisions at junctions, and so on. And of course reducing the speed limit means it takes longer to get from A to B – and that too has an economic cost.

There is nothing wrong with campaigning in a single-minded way. Just like it’s fine for part of us to insist on having more chocolate, so it is fine for people to argue in favour of taking road space away from cars and giving it to cycle lanes. That is, as long as there is a mechanism to balance these one-sided demands with the ones with which they are in conflict.

The checks and balances of policy-making usually work well in modern western society. Policy is generally developed through debate, often involves the input of impartial experts, and eventually ends up being a compromise that is ‘good enough’ for almost all parties.

But what if it isn’t? What if single-minded campaigners actually gain the power to push through their single-minded cause, disregarding opposite views? Or what if a binary vote, like in a referendum, becomes the dictatorship of a slim numerical majority?

Such situations may have seemed a remote, perhaps only theoretical, possibility just a year ago. But now, there is much evidence that the checks and balances that are intended to protect liberal democracy are under pressure. We see an authoritarian president with a rather cavalier attitude towards his country’s constitution and to the separation of powers. In the British parliament, we see the most ineffective opposition for generations, and a depressing abdication of its scrutinizing role, thus bolstering the unchecked power of the government.

Behavioural economics is being used in public life. More and more governments are setting up teams to use behavioural insights to support the implementation of policies and to influencing the behaviour of their citizens.

But perhaps a much bigger public life challenge for behavioural economics is the safeguarding of a liberal democracy from domination by single-minded views, and that ensures a sensible compromise between the conflicting demands of the population. When it concerns nudging us individuals into healthier, greener or safer behaviours, behavioural economics can appeal to our inner benevolent dictator. Unfortunately, in the domain of policy-making and government there is no such benevolent dictator.

I am not normally attracted by campaigning, precisely because of its single-minded nature. It is hard not to invoke the ‘other hand’ as soon as I see even a hint of a campaign. But a campaign to encourage behavioural economics practitioners to find ways to curb authoritarian policy-making certainly has some appeal.

That’s the kind of campaign that I can wholeheartedly get behind.

Posted in Behavioural economics, Economics, politics | Tagged , | Leave a comment

The irrationality of charity

Featured image: Howard Lake/Flickr

When charitable donations are concerned, we seem to be even more irrational than otherwise

What is an acceptable salary for a CEO who works in the charity sector? For the Daily Mail, a British tabloid newspaper, £200,000 would seem to be way too much. Earlier this month, they published a typical rant about, an internet platform to make collecting donations easy for individuals who take part in sponsored events and small charities that don’t have their own card facilities.

In it they berate the organization for ‘raking off £20 million a year’ and for paying its staff an average of annual salary of £60,000. Soon after, many Twitter users voiced their disapproval with JustGiving’s practices. Not all of them were typical Daily Mail readers – this is what ITV journalist Robert Peston tweeted:

Even some charities tweeted suggesting they’d take their business elsewhere.

Charitable donations are a most intriguing subject for anyone fascinated by the rationality or otherwise of our behaviour. For a start, giving itself seems puzzling. The kind of altruism that has us simply donating our money away seems the starkest violation of the self-interested, utility-maximizing, rational homo economicus.

If you drop the narrow perspective and look a bit further than that axiom, however, there are plenty of entirely plausible and indeed rational explanations. But that doesn’t mean we’re not irrational when charitable giving is concerned.

One of the most compelling demonstrations is Peter Singer’s Drowning Child thought experiment:

Imagine you’re walking through the park and see a small child in a pond. The water is shallow enough for an adult to be able to stand up comfortably, but deep enough for a toddler to drown. There’s nobody around so you’re the only person capable of saving the little kid. Yet you’re wearing an expensive pair of shoes and your best suit, worth in total a good few hundred pounds. What would you do? Let the child drown to spare your attire? Or would you get into the water and save the toddler, and ruin your shoes and suit?

As Singer says, most people at least say they wouldn’t hesitate to rescue the child. Yet if we are prepared to sacrifice a couple of hundred pounds to save a drowning child in the park, how come we’re far more reluctant to donate a similar amount to save a starving child thousands of miles away? Would a rational person consider the life of one unknown child so much more worthy than that of another unknown child?

One explanation for this is the identifiable victim effect. We appear to be much more easily engaged when there is a concrete beneficiary of our charity than when they are abstract. Some charities use this explicitly. Plan International, for example, invites people to sponsor individual children, providing detail about their lives and facilitating direct corresponding between the donor and the recipient (even though the money donated doesn’t go directly to the child in question). And many other charities use personalized material to nudge prospective donors to open their purses.

There is even more intriguing charity-related behaviour. Nichola Raihani and Sarah Smith examined publicly visible donations (on a site like!) to look for evidence of competitiveness among donors. They found that male donors donate significantly more when the fundraiser is an attractive female, and when they are responding to a large donation made by another male donor. (The difference is a factor 4.) There was no such competitive behaviour among female donors giving to attractive male fundraisers.

But perhaps more irrational still is our unease with the notion that only a fraction of our donation eventually reaches the deserving recipients. This is what appears to be behind the outrage of the Daily Mail. There are two concepts in conflict here. On the one hand, we want the cause we support to obtain as much money as possible. On the other hand, we don’t want part our donation to be ‘wasted’ on other activities, like administration, or worse still, the CEO’s salary.

This is a form of mental accounting that can trip us up in other situations too. Many people value ‘free shipping’, saving for example $6.99, more than a $10 discount on the price of the item. We consider delivery charges a form of waste – we’re only interested in the item. Likewise, we resent having to pay a ‘booking fee’ for concert tickets, or to pay a surcharge for using a credit card.

But there is probably also an element of fairness at play here. In 2008, On Amir, Dan Ariely and Ziv Carmon conducted an experiment in which they assessed the willingness to pay for recovering the data of a faulty computer disk. They looked both at the value of the data, and the amount of time spent by the technician to recover it. To reclaim 5 years’ worth of data, the participants were willing to pay 67% more for the recovery if it took 12 hours rather than 5 minutes. In the case of 1 month’s worth of data that difference was nearly 150%. It appears we would feel seriously ripped off if we were charged a large amount of money to recover valuable data if it took very little effort. We would feel exploited.


It is probably this kind of emotion that most fuels the indignation about the fees JustGiving charge (and the salary its boss receives). It is unfair for people to profit from our generosity by taking a slice of our selfless sacrifice.

Of course, most of our decisions are driven by emotions. We choose what we feel is best. But if we feel conflicting emotions, weighing them up can be very difficult. We feel good if a cause that is close to our heart gets lots of money, but we feel bad if that is only part of the total that is donated. If the latter dominates our thinking, we risk demanding that, in the words of Anne-Marie Huby, the co-founder of, “a good charity should be a poor one, run on a shoestring.”

Professional fundraisers and platforms like help increase the amounts donated to charity by a much larger margin than their costs. But if our decisions are guided by the negative emotion that comes from the perception that a slice of our donation is wasted and channelled off unfairly, we ultimately damage the cause we find important.

And that is as good a definition of irrationality as any.

Posted in Behavioural economics, Cognitive biases and fallacies, Emotions | Tagged , , | Leave a comment

When a devil’s advocate throws in his wig

(featured image: Sarah-Rose/Flickr cc)

Sometimes you need to choose between the quest for wisdom and maintaining sanity

David Sipress is a celebrated cartoonist whose work regularly features in the New Yorker. One of his drawings, so old the author cannot even remember where it was first published, had gone viral, and in a heartfelt article in the same magazine last weekend, Sipress reflected on the reasons why.

The title of the article (How to stay sane as a cartoonist in Trumpland) and the caption underneath the cartoon got me thinking too. Following news bulletins, newspapers and social media can indeed do more than just provide you with information. A new Government measure can, depending on our political leanings, provoke anger or pleasure, and a report on crime figures can make us frightened. Even the stock market movements can affect our mood and our behaviour – Richard Thaler’s famous advice to investors (in Misbehaving) is to buy a diversified portfolio, and then “scrupulously avoid reading anything in the paper aside from the sports section”.


Stuck in a filter bubble

And if we want to keep well-informed, it is very likely we will be regularly confronted with views that are violating our beliefs and values, and that can feel distinctly uncomfortable – never more than in recent months.  There is a trade-off to be made between knowing what’s going on, and keeping mental stress at bay. At some point we may have to start cutting down on the news for our own sake.

But perhaps it is actually a good thing for our mental health to expose ourselves to viewpoints that are conflicting with our own. In 2011 Eli Pariser coined the term filter bubble, referring to the tailored internet newsfeeds that serve to reinforce what we think, and to shield us from information that would broaden or challenge our world view. Now, a google search for the same term yields nearly 20 million hits, suggesting it is a topic that draws attention.

You’d be wrong to blame your filter bubble only on Facebook, Yahoo News, or the Huffington post, though. As Pariser writes (in a post about a Facebook study on the exposure to ideological diverse news and opinion), who our friends are, and who we choose to follow are at least as important as the filter algorithms.

So if we want to pop that bubble, we need to be deliberate in seeking out contrary views. You could, for example, do what Gillian Tett suggests we do with our Twitter timeline: “replace half of those you follow with others who espouse a radically different view”). But is simply knowing that there are views that differ from ours enough – or should we go further and try to understand the people holding these views?

Siding with the devil

This is where devil’s advocacy comes in. Acknowledging contrary views may hamper the formation of a bubble, but the easiest way to respond is to cling even more strongly to your own position and dismiss the alternatives outright: they are wrong and misinformed, and you are right. What if, instead, you tried to actually argue in favour of the opposite side? You would force yourself to override your own preconceptions and move – even if only temporarily – closer to that opposite viewpoint. Most importantly, you would develop a logically plausible case for something you disagree with.


For many years, I have been driving my children up the wall by taking the opposite viewpoint to theirs. When they argued in favour of the legalization of soft drugs, I made the case against; when they hailed the smoking ban in pubs, I (a non-smoker of over 20 years) took pro-smoking side and so on. And then, one day, one of them came home saying they played devil’s advocate in a discussion on student loans. That was one of my proudest moments as a father…

Playing devil’s advocate is not just an instrument for annoying your offspring, though. It is also a rather potent  way to combat an array of cognitive biases (including old favourites the availability heuristic and the confirmation bias), and to become more comfortable with cognitive dissonance – something president Truman’s two-handed economists would have been proud of.

Is there is risk that your defences might crumble and that you might actually change your mind? Sure –in fact that is the purpose of playing devil’s advocate. Scrutinizing both your own and the opposing viewpoint allows you to adopt the one that is the most plausible. Arguably playing devil’s advocate protects you from a barrage of discordant views with the power of logic and reason. It might even help my good friend Oliver Payne, who recently tweeted


The boundary of advocacy

What you do as a devil’s advocate is very much what defence lawyers do. If their client is innocent, they look for the evidence and build a persuasive defence case on the basis of it. If their client is guilty, but there are mitigating circumstances, their goal is to convince the court (or the jury) to see the situation from their client’s side and take into account these extenuating circumstances.

But what if their client is obviously and undeniably guilty as sin, and where there are no such moderating conditions, no matter how hard you look for them? There is not really much to do for an advocate in such situation.

And it is the same for a devil’s advocate. When, for example, a president invokes alternative facts to his heart’s content, and acts blatantly in his own self-interest rather than in the interest of the nation, then devil’s advocacy has reached its boundary.

When someone’s behaviour is indefensible, persisting would not lead to wisdom, but to insanity. That is when a devil’s advocate has no choice but to throw in his wig.

Posted in Behavioural economics, Cognitive biases and fallacies, politics | Tagged , | Leave a comment

Just in time, year after year

Why I keep procrastinating about my tax return

The British are known for the occasional eccentricity. One of these quirks is their tax year: for reasons that escape me, it runs from 6 April to 5 April the next year. Annual income tax returns, if you submit them online, are due at midnight on 31 January. In principle, all information to complete it should be available some time in May for most people. Yet only just over half of the 11 million tax payers that must submit the return, do so before the end of the calendar year, and of the remaining 5 million nearly 20% failed to meet the final deadline last year.

Marshmallows may help explain this phenomenon. Walter Mischel, a psychologist at Stanford University in the 1960s, conducted a series of experiments on 4- and 5-year old children to gauge their self-control, to find out to what extent it might predict how they fared later in life. In one variant a child was brought into a room, and asked to sit down at a table on which there was a single marshmallow. The experimenter then told the child he was going to leave the room for 15 minutes, and the child could either eat the marshmallow at any time before he returned or, if it was left untouched, there would be a reward of a second marshmallow.

marshmallowtrumpFew of the children managed to leave the sweet alone for the full duration, despite the prospect of double the amount of marshmallow after a mere 15 minutes. It’s easy to snigger at how these small children struggle to delay gratification and give in to temptation, but we adults shouldn’t really laugh too hard.

A popular case study in behavioural economics is that of the failure of many people to save enough for their retirement. They succumb to the temptation of spending small amounts of money now, rather than investing it into a pension fund that will produce a much bigger reward later: a process that is very similar to eating the marshmallow now instead of waiting and receiving two later.

My procrastination is the time equivalent to marshmallows and pension savings. Of course I could spend a couple of hours on a Saturday afternoon, now and then throughout the year, keeping my paperwork up to date so that when the tax year ends, all I need to do is one final check. But I don’t – there are far better things to be done, and there’s plenty of time left, so I put it off.  Bizarrely, even the knowledge that I would receive a sizeable tax refund has not proved enough of an incentive to make me submit the return in May (and get the refund before the summer holidays) rather than in January (and get the money nine months later).

Commitment ahead

Is there a way I could save myself the stress of finding not only enough time in the few remaining weekends before the deadline, but also of finding all the documents I need to work out what I am owed or owing?

Maybe I could try to rely on a pre-commitment device. Behavioural economists Richard Thaler and Shlomo Benartzi developed the Save More Tomorrow approach, in which employees commit in advance to allocating a proportion of future wage increases to their retirement plan. This kills two birds with one stone. First, it decouples the decision from the implementation, which reduces the cognitive load (“Perhaps I should first buy this new phone/a new pair of shoes and start saving next month”) and makes it a fait accompli. It also avoids the loss aversion that might hold people back: their take-home pay will still go up when they get a salary rise, only by less than it otherwise would have been.

Dan Ariely has investigated (together with Klaus Wertenbroch) how such a pre-commitment approach might affect his students’ tendency to procrastinate with the papers they need to submit during a semester. In the video below he explains how he offers them the choice between a single, end-of-term deadline, and setting their own intermediate deadlines, with a certain grade penalty if they are late.

One might argue that the rational thing to do for a student is to go for the standard option: this leaves the maximum amount of freedom to submit when you want, with no risk of a penalty in case some unexpected event prevents you from submitting according to a self-imposed, earlier deadline. And yet, as he describes in Predictably Irrational, students with well-paced intermediate deadlines achieve higher grades.

So maybe I can penalize myself for not keeping on top of my income tax paperwork – no beer money for a month if I failed to organize my receipts and invoices in the previous month? If I’m not trusting myself, I could rely on an external referee, and hand them my beer money for a whole year, only to get it back if they have verified I’ve completed my monthly task. There is even an internet platform, stickK, which makes it easy to set up the contract. One of their suggestions is to forfeit the money to anti-charities: if you fail to meet your goal, your stake is donated to a cause that you despise, just to motivate you that little bit more.


And yet… there is one thing that is overlooked in all of this. The unspoken assumption has been that having to race against the calendar, if not the clock, to fill in all the boxes on the tax return, to have to rummage through stacks of documents, to trail through emails and bank statements, to struggle with half-completed spreadsheets – that all this is a bad thing. And perhaps it is: spreading the volume of work out over a longer period of time would most likely take away most of the last-minute stress.

But what about the incredible feeling of relief when I finally hit the Submit button, with a few days to spare? Would I feel anything like it if I was ready many months before the deadline, and all I needed to do was simply copy some numbers from a well-maintained spreadsheet?

I doubt it. The sense of liberation and satisfaction I experience when I can finally draw a line under the bureaucratic chore is well worth the hassle, the agony and the panic.

It is, all in all, a good trade-off for me, and that is why am sticking to my routine of procrastinating for eight and a half months, and two weeks of mad rush, year after year.

Posted in Behavioural economics, Cognitive biases and fallacies | Tagged , , , | Leave a comment