A voter’s burden

(featured image: carnagenyc)

Voting, like so many things in life, is about the difficult task of weighing up many, often conflicting options. Maybe a (behavioural) economics perspective can help cut through it?

On 18 April, Prime Minister Theresa May called a snap election in the UK for 8 June. On the face of it, she is after an enhanced parliamentary majority to give her “a strong hand” in the Brexit negotiations  and deliver the “best possible deal for Britain”. But even though Brexit dominates much of political life, there are other domains in politics that are pushed by the various parties. The ruling Conservative party did come up with a proposal to cap energy prices (see The anti-nudge), and their manifesto also contained a surprising plan for funding long-term adult social care.

The idea itself is rather simple. At present, if you need social care at home, you need to pay for it yourself until you have less than £23,250 left, excluding the value of your home. The proposal is to raise that threshold to £100,000, but to also include the value of the home (which may then need to be sold after your death to settle the balance). Initially, the proposal stipulated that, other than the £100k floor, there would be no limit on the total amount paid by the patient.

That would have been bad news for anyone owning even a modest home, and so there was a hurried correction to the proposal to cap the total amount payable (to a so far unspecified level). But even this amended version implies a considerable shift of the adult social care cost from the state to the (well-off) individual. Naturally, the opposition parties, well, oppose the idea, and would pretty much maintain the status quo, continuing to fund adult social care largely out of general tax and national insurance revenue.

This binary choice relating to a single policy domain, between more and less individual contribution to the cost of care, conceals a remarkable complexity, though. Choosing – and hence voting – is a daunting task. How to approach this challenge?

A material world

Let’s start looking at the trade-offs involved at the unequivocally rational end of the spectrum, and consider the material costs and benefits. Would the Conservatives’ proposal leave you richer or poorer? That looks straightforward enough. Sure, it depends a lot on whether you own your home, and if so, how much it is worth, but that can all be worked out.

Financial Times journalist Chris Giles calculated what percentage of your wealth would be eaten up if the total care costs before you finally died were £150,000, as shown in the graph below. And there were several more attempts to determine more specifically under what circumstances you’d gain or lose.


What will be left of my fortune? (source: Chris Giles)

All that is on the assumption that you will need the kind of care in question, of course. If you die without needing long-term care, it doesn’t matter how much you’d have had to pay: you can leave your estate to your heirs intact. But the state is carrying an ever increasing cost burden for long-term care, you are likely to pay for it anyway while you are still alive through higher taxes.


With the uncertainty around what you will and won’t need when you’re old and decrepit, it looks as if establishing a clear-cut cost-benefit case is not that easy.

Belief and conviction

Might strong ideological convictions help us be more decisive? If you believe that it is a sacred right to leave your fortune to your children, then you should clearly not support the Conservative manifesto proposal. If you believe that inequality is a growing problem, and that one of the forces that sustain is in inheritance, then the proposed measure is for you. If you hold both beliefs, though, you’re still no further towards a decision.

Maybe you have been a working class Labour voter or a middle class Conservative voter all your life, convinced that your preferred party serves your class the best. But here are the Tories, proposing to take money away from the middle class, and Labour rejecting a measure that would free up billions in the social care budget by making the well-off pay more. That doesn’t help either.

Context matters

When people’s preferences are weak, or ill-defined, they are often influenced by the so-called choice architecture – the context that surrounds the situation in which you have to decide. But the context varies dramatically depending on who you listen to.

Is it a quadrupling of the amount you can keep, from £23,250 to £100,000? Or is it a dementia tax, that will force people to sell their home, reclaiming much of the one bit of wealth that people of otherwise modest means can pass on to their children? Is it a regressive measure or a progressive measure?

Is it fair because it makes people contribute according to their means, especially those lucky enough to have profited from the astronomical increase in property values? (An average house in the UK costing about £12,000 in 1976, was worth more than £200,000 forty years on, an increase by a factor 16. In London house prices grew by a factor 32 over that period.) Or is it unfair because it only affects those unlucky enough to be affected by a terrible long-term condition like dementia?


Principles or pragmatism?

Even the hasty amendment of the manifesto proposal can be framed in two ways. Was it Mrs May’s umpteenth “U-turn”, showing what an unprincipled politician she is? Or was it a sign of pragmatism – changing your minds when the facts change (as the great economist Keynes never said – it was Paul Samuelson)?


An impossible challenge

And this is of course a matter of just one policy. Let’s imagine that you managed to weigh up all these elements and come to a conclusion. But now let’s say that you are feeling very strongly about Brexit. If you’re against the Conservative social care proposal, but you want to give Theresa May the strong hand she’s asking for, what do you do? Or if you think it is a good idea to let well-off people pay for their social care, but you want to give Mrs May a strong parliamentary opposition (or even, hoping for a miracle, stop her gaining an overall majority), what do you do?

Reflecting on all this, I should probably count myself lucky: being a foreigner, I don’t get to vote. But I will spare a thought for the poor British voter, and I sympathize with those who, having thoroughly thought everything through, decide to abstain. Some challenges are just impossible.

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

It’s a bargain… or is it?

We are happily handing over large amounts of private data to internet giants. How come? And does it matter?

Our lives are, in a way, an endless string of trades. Have a look at your latest current account or credit card statement. Barely a day goes past when you don’t wave our contactless card in front of a reader, or enter its digits into a web form. Add to that the numerous direct debits for utilities and subscriptions as well as still plenty of cash transactions. And of course there is the source of all that money for many of us: the income we get in exchange for our labour. Humankind is clearly of the trading kind.

One reason we do this in such a matter-of-fact way, almost carefree, is that we generally feel well protected in doing so. When we hand over money, whether that is for groceries or for a car, there are laws that safeguard our rights and ensure the goods supplied are fit for purpose. Additional measures shield us in situations where there is a risk of abuse or confusion (e.g. distance selling) or with more complex services like insurance. There are cooling-off periods in which we can change our minds free of charge, and ombudsmen to complain to when we have a problematic transaction. And if our employer is cheating us, we can go to an employment tribunal.

Moneyless trading

But we also trade without money. Perhaps you are in a car-sharing arrangement with a colleague for your commute: today you drive, tomorrow it’s their turn etc. Even if no money changes hands, it’s still an economic transaction. We water the neighbours’ plants when they’re on holiday (or they water ours), we lend out our ladder or borrow a friend’s car battery charger.

Behind such trades is the notion of reciprocity, one of the oldest and most prominent motivational factors in social interaction. We do something for someone else, in the expectation that they will return the favour at some later point. We experience this even when we don’t interact directly with people we know. Internet forums like Mumsnet, or those catering for people suffering with medical conditions or psychological distress are full of users more than willing to support others without any financial compensation. Youtube bulges with videos uploaded by selfless individuals showing how to change a washer in a tap, or replace the brake pads of a classic car.

No need for a safety net of guarantees, trading standards or tribunals here. What is at stake in social networks like this are things like friendship, reputation and status, and that seems to be enough to keep the intense economy of social trades going.

But the internet has also brought a new kind of trade. We could call these unconscious trades: we are not immediately aware that we are exchanging something for what we get. I am, of course, referring to the ‘free’ services from the likes of Google, Facebook and Twitter. But of course, as the old saying goes, there is no such thing as a free lunch – certainly not when it concerns giant companies. Or as Andy Lewis tweeted

Generous with data

Much has been written about the fact that we freely supply these companies with all manner of personal information. Our date of birth, the school we went to, our relationship status, but also where we are (how else do you think Google Maps can tell us how quickly the traffic flows in our street?), what websites we’ve visited (mmm… cookies!) or the contents of our emails.

How come we are so generous with our data?


My street is busy. But how does Google know that?

One explanation is that we don’t even realize how valuable our personal information is. In the real world, that would be akin to someone giving away an old, tatty painting, not realizing it is actually an old master, worth millions. Now, our individual data may not be worth that much, but when you look at Google’s or Facebook’s revenue from advertising, you see it’s not peanuts either. The ability to tailor ads to us, the user, that they offer is exceptionally valuable to advertisers. AVG, an antivirus company, used to offer a browser extension that monitored your usage and settings for Facebook and Google, and estimated how valuable you were to them. In a 2014 AdAge article, a journalist reports she was worth $20.75 each year to Facebook, and no less than $223 to Google.

Does that mean we are being abused as consumers, in a way that, in the real world, would have trading standards up in arms?

A fair bargain?

There are two parts to the answer. The first one is concerned with the bargain itself – are we getting ‘value for money’ from the trade (even though we don’t pay anything)? Hal Varian, Google’s Chief Economist worked out, in a presentation on the Economic Impact of Google, that the consumer surplus per user of Google’s search facilities is of the order of $500 per year. He is quite upfront about the fact that it is a back of the envelope calculation, but that is not an insignificant user benefit.

But perhaps you should ask yourself: how much would you need to be paid to give up Google (or Facebook, or Twitter etc.)? That will give you an idea of what these services are worth to you. Then you should ask yourself, would you ‘sell’ your private information for that amount?

The other part is more tricky. Is a trade, in which we don’t really realize what we are ‘paying’, a fair one? Imagine I offer you, say, a ‘free’ tablet computer, but unbeknownst to you, I extricate £100 from your pocket. That may be an excellent price for the device you received, but if you are not aware that you’re actually paying for it, it’s difficult to maintain the transaction itself is fair.

Arguably, the same applies in our relationship with the internet giants. Of course, we have all clicked the button to say we have agreed with their general terms and conditions, and their rights to use the information we supply as described. Some even go so far as to require that you scroll down all the way to indicate that you have read all sixty-odd pages. But even if the exchange is fair from a market viewpoint, that is perhaps not enough to deny they are exploiting our attraction to what is free to lure us into divulging our private information for their own lucrative aims.

Are we getting a good and fair bargain from Google, Facebook and co? To answer that question we need to consciously consider the trading relationships we have with them. But few of us do that, and so we really have no idea.

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

The anti-nudge

(Featured image: Publicdomainpictures)

Politicians’ love of behavioural insights is quickly forgotten if votes can be bought with populist measures

When elections are imminent, talk is cheap, and cheaper still are politicians’ promises. Sure, there is little doubt that British Prime Minister Theresa May’s decision to call a snap election on 8 June was inspired by the Brexit challenge she faces. But there is more to governing than just extricating the country from the EU. So the election manifesto needs to contain some other pledges that might engage the electorate more than what role the ECJ should or shouldn’t play.

One such promise from the Conservative party is to cap the energy prices for people who are on the so-called standard variable tariff (SVT). Estimates vary, but up to 18 million households (out of a total of just over 26 million) are assumed to pay this tariff. The government reckons that with a price cap could save these households up to £100 per year.

Switching is hard

The UK energy market is characterized by fixed price deals, which freeze your gas and electricity prices for a given period of time. Almost all suppliers offer a range of such deals, with varying durations. All of them also have a standard rate, to which you revert at the end of your fixed deal (unless you select a new deal from the same supplier, or switch to a different supplier). The SVT fluctuates with the prices on the world markets, and is the default for whoever is not on a fixed option.

As the market got privatized and deregulated, comparison websites appeared, allowing consumers to check prices and identify the best deal for them. This, together with a landscape that contains six large suppliers and more than 40 smaller ones, was expected to provide sufficient competition to maintain a dynamic market with low prices.

So why do millions of people pay “too much”? Ofgem, the UK energy regulator, reported that 7.7 million switches of electricity or gas took place in 2016, up 28% from 2015, and a six-year high. But that still leaves nearly 20 million users who didn’t switch. Maybe some of them moved to a new suitable fixed-price deal with the same supplier. However, only 15% of users regularly switch, and as if to illustrate the point, Energy Secretary Greg Clark admitted he never switched electricity or gas suppliers himself because it is “a hassle”.

All else is never equal

Unilateral interventions like a price cap generally suffer from a fundamental flaw, for they come with the implicit assumption that nothing else will change. Somehow, so the logic goes, the supposedly ill-gotten profit by the energy giants can be siphoned back to the hapless consumer without any side effect.

But things will change, and there will be side effects. Suppliers are not stupid, and will respond to such interventions in any number of ways. They will seek to save costs (almost always labour cost, thus reducing employment), or they will invest less (which will mean dearer or less reliable supply in the future). Or maybe they will raise the prices of the fixed tariffs to protect their revenue, thus transferring money from the pockets of the astute customers to the passive ones. And if the price cap is announced in advance, they will raise prices in anticipation.

The least likely course of action is that they will give away part of their profits to the tune of £1.8 billion each year (that is £100 times 18 million households).

But the problem remains that, according to Citizens Advice, some vulnerable consumers lose out to the tune of £300 per year on poor value energy tariffs. In a recent blogpost, James Plunkett points at the existence of two markets as a cause: an active market, in which about 1/3 of the households switch regularly, and a passive market, in which people rarely switch. The energy companies, so the narrative goes, take advantage of the inertia of the passive households to keep their prices high.


So much choice… but who is the cheapest provider? (image: wilhei)

There is nothing sinister about this: companies always try to discriminate between different types of customers. This the reason why retailers give discounts to people who can be bothered to cut out coupons, rather than to everyone, or why travel tickets bought in advance are cheaper than those bought at short notice.

Rewarding bad behaviour

But the problem of the maybe 4 or 5 million vulnerable and poor households, many of whom really would have a hard time switching online, can be resolved through targeted benefits or subsidies. A blanket price cap that would benefit at least 13 million households who are by no means poor or vulnerable, on the other hand is a terrible intervention.

Not only is it terrible from an economic perspective, it is awful from a behavioural perspective too. In a market economy, prices are kept low and quality high through competition. So you need a sufficiently large number of suppliers on the one hand, and discerning consumers on the other hand, able and willing to switch providers if they’re not happy with price or service levels.

If you want to strengthen this mechanism you’d need to nudge consumers towards being more astute and active, by making it easier to do so. Of course, some people are strongly motivated and happily cope with the hassle of finding a slightly cheaper deal. Others have been encouraged by nudges like prompts on bills. Additional nudges might get even more of those who are on the fence to take the trouble to switch. Not only does this save those people money, but more importantly, it also contributes to the good working of the market.

But a price cap is an anti-nudge: it makes it easy for people to do the wrong thing. It reinforces the status quo bias, by giving free money to people who do not switch. For all we know, they stick with the status quo not because of a bias, but because they deliberately prefer to stick with the same supplier (for whatever reason). And that behaviour, which weakens competition, will be rewarded by the proposed measure.

Politicians have been boasting for years about their adoption of behavioural insights – the UK’s was one of the trailblazers in this respect. But what we observe here is that the best behavioural approaches are no match for brazen populism and economic illiteracy.

It is doubtful a nudge would be sufficient to prevent politicians from resorting to such vote-buying tactics. For that, we would really need a rather potent shove.

Posted in Behavioural economics, Cognitive biases and fallacies, Economics, politics | Tagged , , | 1 Comment

Choice and regret (or when there is too much jam)

(Featured image: Jarmoluk)

Freedom of choice is awesome… or is it?

A friend of mine attended a conference recently. It was a huge affair with thousands of participants, and – excluding special events – around 350 sessions, spread over three days. There were 22 rooms in which simultaneous, parallel lectures and workshops were taking place.

If that sounds astronomical, that’s because it actually is astronomical. The logistics involved in this kind of conference are huge, and I take my hat off to the organizers for pulling such an endeavour off with great success.

But something about it makes me feel rather uneasy. Sure, several thousands of participants will have diverse interests within the domain of a conference of this magnitude, and so it is unlikely that all 350 sessions would be of equal interest to everyone. But no matter how you cut it, you can only reasonably attend perhaps 25 of the sessions. A similar problem confronts people attending large festivals like Glastonbury, which this year has a line-up of 88 artists from Aanderson .Paak and the Free Nationals to Wiley. (Presumably ZZ Top were unavailable.) This does not include dozens of smaller acts and non-musical performances.

Too much jam

So how do you choose when there is so much to choose from? Can you imagine anyone working out the different criteria that describe the conference sessions or the bands, assigning weights to them, and scoring them to rank them in order of priority? This is not a sensible way of dealing with what is, essentially, choice overload.

Barry Schwartz, an American psychologist, described this phenomenon in his 2004 book, The paradox of choice. An experiment often quoted to illustrate this was conducted by Sheena Iyengar and Mark Lepper. They offered people gourmet jam or chocolates, either from an array of 6 choices, or an array of 24 or 30 choices. People facing the larger range of possibilities, they found, were much less likely to actually buy. And if they did make a purchase, they were more likely to be dissatisfied with their choice.

As is so often the case, things are not quite that simple, though. In 2010, Benjamin Scheibehenne and colleagues carried out meta-analysis of 50 similar experiments investigating choice overload. They discovered that several among them had found no effect, or even on the contrary, that more options had facilitated choice and actually increased satisfaction. This suggests that the phenomenon does not apply universally.

Still, where it is observed, there appear to be two explanations for people walking away from making a choice. One is the cognitive load of actually choosing – the ‘cost’ of making the decision exceeds the ‘benefit’ that having made a particular choice delivers. It’s just too much bother, and life is too short. The other one is the anticipation that one might feel one should have made a different choice.

Regrets, I’ve had a few

This regret is the stronger if the situation enforces exclusive choices. You may end up regretting choosing pineapple with rose petals jam at the display stand in the supermarket, when you’re putting it on your toast the next day and it’s not quite what you thought it was going to be. But you can purchase your alternative preference next time. The cost of doing so is small.

At a conference or a rock festival, however, opting for one session or artist inevitably means rejecting all other options that exist at the same time. And that can be serious regret: if you lose out on the chance of seeing Foo Fighters at Glasto 17, that chance is gone forever. And what’s worse: it was your choice to miss them, because you were totally in a position to do so!

The larger the number of choices you have to let go, the larger this kind of regret, even – perhaps even more so – for people who are rational thinkers. With each extra possibility the amount of regret increases asymptotically to the level of regret corresponding with not attending at all, and missing everything. As the sum of the heartache goes up relentlessly, it will reach the point where it outweighs the joy of experiencing even the best combination of options, and the rational conclusion is then inevitable: don’t attend at all.

The paradoxical nature of freedom of choice seems to be even starker when it involves irrevocable consequences. On the one hand, you would expect us to want to make such momentous choices ourselves. On the other hand we may find the weight of responsibility unbearable. Imagine you were given the possibility of choosing whether you, or your life partner, dies first. Would you take it, or would you rather pass and let life decide for you?


A boy or a girl – your choice? (image: Jean-Pierre Lavoie)

Thankfully, few choices we need to make in real life concern life and death. But some are truly hard, and life seems to get ever more difficult in two ways. The amount of choice in almost any category you care to think is more bewildering than ever: books, food, music, places to live or go on holiday, dental fillings, furniture, blogs to read… you name it. And at the same time we gain ever more power to actually make choices that used to be completely beyond our reach. It is now possible to select the gender of a baby with reasonable accuracy – yet people are not necessarily happy that we should have this choice: at debating.org, 74% of respondents thought parents should not have that right.

But also for less critical choices, it’s worth reflecting on all this freedom of choice. Maybe, whether we’re conference or festival organizers, a supplier of bread or just an ordinary person planning a family holiday, we should consider giving others less choice rather than more. Perhaps we should not always abdicate responsibility and leave them to work it all out.

But, even then, the paradox is never far away.

More freedom to choose, or less regret? That is a difficult choice.

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

Too much commitment is bad news

Featured image via ytimg.com

What can go wrong when tenacity goes too far

People who keep going in the face of adversity speak to our emotions and our imagination like few other things do. Inspirational books are often written by (or at least about) successful sportspeople and intrepid explorers who achieved greatness through heroic perseverance.

Artists like The Beatles spent many years in relative obscurity, living on a pittance before they found fame and fortune. Long before they became billionaires, businesspeople such as Lord Sugar started out in the very modest surroundings of a council estate. And classic and popular literature alike features characters that pursue their goal without letting up, no matter what difficulties they run into. From adventurers like Ulysses and Professor Otto Lidenbrock in Jules Verne’s Journey to the Centre of the Earth to Charles Dickens’ Oliver Twist and Roald Dahl’s Matilda – we cannot help but be in awe of their tenacity and their unwillingness to give up.

Passion and perseverance

But these are all people who are not like us. Whether they’re fictional heroes or giants of the entertainment industry or business, their experiences seem to be well beyond us ordinary mortals. Not so. Last year, Angela Duckworth, a psychologist at the University of Pennsylvania provided an insight into the science behind our fascination with the determination of these great people. In Grit: the power of passion and perseverance, she describes how this trait outweighs talent in realizing personal goals.

Talent is very much something we are born with, but the good news that from Duckworth’s work is that grit is something we can learn and cultivate. Whenever we encounter an obstacle on the way to our goal, we have a choice: to proceed or to give up. In a sense this is a conflict between the present us and the future us.

The thinking of the present us is dominated by the effort we need to put in now, and by the temptations of the alternatives to persevering. Maybe the ambition of completing a marathon is beyond us anyway, or finding the time to put in the regular hours training is simply not possible. Or perhaps we can start preparing next Monday, and enjoy relaxing in front of the TV just a little longer.

The future us takes a different view. Instead of focusing on the effort right now, it looks back at us from the time at which we have achieved our goal – or have given up. It is the emotions we imagine we will experience – the joy of succeeding, or the regret of having bowed out – that are feeding the passion element in Duckworth’s grit.


I’m a bit tied up now, but let the Sirens sing! (photo: Wikimedia)

But what about the perseverance part? We can use a device from the toolkit of behavioural science to help us that. Interestingly, it is named after one of the intrepid heroes mentioned earlier. The so-called Ulysses contract refers to the encounter of Homer’s hero with the Sirens. The song of these mythical creatures enchanted sailors, making them incapable of rational thought and wreck their boats on the perilous rocks around their island. Ulysses made a pact with his men: they were to put wax in their ears so they couldn’t hear the dangerous singing, and to tie him to the mast, to prevent him steering the boat towards the rocks. He also ordered them to attack him should he break free.

As a more contemporary version of this kind of commitment device, students have been known to give their Facebook password to a friend, ask them to change it and not tell them the new one until after their exams.

The flipside of commitment

So commitment strategies can be a great tool to help us combat the irrational lure of our impulsive System 1, seeking immediate gratification and avoiding effort, and rely on our more rational System 2. But there is a flipside to commitment that could push us back into irrationality.

For it is one thing to make it easier for ourselves to choose the ‘right’ option and make choosing the ‘wrong’ options harder. But it is another thing to be so deeply committed that only one option remains possible. The expression ‘burning one’s bridges’ graphically illustrates the irrevocability of extreme commitment devices. Of course, sometimes there inevitably is a point of no return – for instance in aviation. When a plane is accelerating to take off, there is a point where aborting the manoeuvre  is impossible. There is simply not enough runway to stop safely, and the plane has to take off, no matter what.

Avoiding inappropriate points of no return is not just a matter of avoiding extreme commitment strategies. It is also possible to build up such a strong emotional investment in a pursuit that it becomes impossible to back out.

The present British Prime Minister shows worrying signs that she may be at risk of this. As a Remain supporter (albeit a lukewarm one) she might have looked an unlikely person to negotiate and deliver the Brexit that the British voted for. But a fascinating review by David Runciman of a new biography of Theresa May shines a very revealing light on her character. Her “Brexit means Brexit” was not a hollow phrase, meant to appease the hardliners while she worked out a pragmatic way of limiting the damage. It was the mantra of her determination.

“She takes a position and then she sticks to it, seeing it as a matter of principle that she delivers on what she has committed to,” Runciman writes. “Many of the positions she adopts are ones she has inherited, seeing no option but to make good on other people’s promises. This has frequently brought her into conflict with the politicians from whom she inherited these commitments. By making fixed what her colleagues regarded as lines in the sand, she drove some of them mad.”  This is not what characterizes a pragmatic politician, someone who is a master at the politics which, according to Otto von Bismarck, is “the art of the possible”.

Such blinkered focus on an end goal, strengthening the conviction that one is pursuing a righteous cause is antithesis to rationality. The commitment has then degenerated into the sunk cost fallacy.

And that is, unfortunately, unlikely to end well.


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

The dark side of nudging?

(featured image: socialneuron)

Are companies like Uber really into nudging for evil?

In a recent New York Times article, the maverick taxi company Uber came under fire for “pulling psychological levers” to influence when and where their drivers work. A company whose drivers are employees could simply tell them what to do, but a key feature of the Uber model is that drivers are self-employed contractors, who are free to choose their working hours.

So Uber is resorting to other means to persuade their drivers. It uses a female persona (Laura) to send texts with tips of where to go next, because that is more likely to influence the overwhelmingly male contingent of drivers. It uses techniques from the gaming industry, showing income targets just within reach, to entice drivers to keep on working. And it uses defaults on their screen (“keep driving” is the highlighted answer when a driver is trying to log off).

The article caused many commentators to question the ethics of Uber’s practices: this is a company blatantly pursuing its own interests, exploiting the weaknesses of its drivers to their detriment. Surely this was not nudging for good? The title of a Harvard Business Review article by Francesca Gino, a professor at Harvard Business School was unambiguous: “Uber Shows How Not to Apply Behavioral Economics”. Yet other behavioural experts were a lot more sanguine. Co-author of Nudge Richard Thaler didn’t see anything evil in the practices that were exposed by the NYT, and Dean Karlan, a professor of Economics at Yale University, suggests that what is painted “as mind control” would be regarded as “good business practices in another section of the paper”.

When a win-win is not a win-win, and when a win-lose is still a win-win

Professor Gino concludes her article saying that the great potential of applying behavioural economics in organizations is to create win-wins. How could one disagree with this? After all, the win-win is the foundation of human progress and wealth creation. As Adam Smith said in his Wealth of Nations: “It is not from the benevolence of the butcher, the brewer, or the baker that we expect our dinner, but from their regard to their own interest.”  We win from buying our meat, beer and bread just as much as the suppliers win from selling, and it is win-wins like these that are behind the relentless increase in wealth.

But perhaps there is more to the idea of a win-win than meets the eye.

Imagine you want to buy a used bicycle. You’ve found a splendid specimen in the local ads and you go and see the seller. It looks every bit the part, and you’re willing to pay up to £450. The seller’s minimum  price is £250 (but of course you don’t know that). Any sale at a price between £250 and £450 would be a win-win outcome: both you and the seller would end up better off than before. (The transaction produces a win-win because there is a positive difference between the maximum offer and the minimum accepted prices – an economic surplus of £450-£250, or £200.)


Now one of you breaks the ice and makes the first bid – let’s say it’s the seller, who asks for £400. Great news! You’d be getting the bike for £50 less than you maximum price so you’re already ahead. But rather than accepting that price, you do the air-sucking-through-teeth thing, and you point at some imaginary flaws and say you couldn’t really pay more than £300.

That price would be just as much a win-win, but compared to the opening price of £400 it is a zero-sum move. The seller would lose the £100 you would gain. And no matter how the negotiation proceeds further, even though the eventual outcome is a win-win overall, every move is a zero-sum, win-lose step.

The division of the economic surplus between seller and buyer is often the result of what can be described as trickery. Each side wants to maximize their part of it and so neither party would willingly reveal their maximum offer or minimum acceptable price. Instead, they seek to lay down an anchor (the first amount mentioned acts as a reference point), or make claims that are hard to prove (“I can only afford X” or “I have two other buyers who are willing to pay this price”). Or they may simply bluff (“I’m going to have to walk away”), or use reciprocity to great effect, like this guy who saved $400 on a second-hand motorbike with a bottle of coke worth $1.99.

Is this kind of thing unethical? Hardly. Both parties are better off than they were before the transaction. And it is important to see that the negotiation involves more than just the price – it involves intangibles, such as the perception of what constitutes a reasonable price, the risk of losing the sale (or the item), the value of a favour and even sympathy for the other side. All this is weighed up against the successive moves, and so eventually both buyer and seller still end up with a real win-win.

In the relationship between Uber and its drivers, the situation is very much the same. A driver might consider signing off  because the extra revenue would not outweigh the extra time. But the response from Uber is persuasive, and they stay on longer. Uber has won – but has the driver really lost by changing their mind? Or have they simply reassessed the relative value of the extra time and the extra revenue, and concluded that it is actually better to keep driving for another hour?

Whose interest?

It is interesting to look a bit deeper at the mechanics behind successful nudging. A famously ‘failed’ nudging experiment in the Netherlands found that placing wholemeal bread more prominently and accessibly than white bread made no difference at all in sales. A plausible explanation for this ‘failure’ is that the preferences of the customers were so strong that a simple nudge did not persuade them enough to change their choice.

Nudging can really only work on you if your preferences are relatively weak (or indeed undefined, as discussed here). If you know very well what you want (or don’t want), nudging will have little or no effect on you. If mussels are your favourite and you don’t like beef, you will be immune to the nudges on the menu William Poundstone discusses in this article.

If our weak preferences make us susceptible to nudging, then of course it opens up the possibility that someone will nudge us in their interest rather than in ours. But is that necessarily unethical?


A nudge in your favour… or in theirs? (photo: Unsplash)

Consider government nudges to encourage us to pay our taxes. Arguably there is a moral obligation on all of us to do so. But tax evasion and (even more so) avoidance are not rare, which gives us a reasonable idea of where some people’s preference lies. So perhaps this is a case of the government nudging us in their interest (to raise tax revenue) rather than in that of the individual being nudged (who ends up paying more in tax, and keeping less). Is this an unethical use of nudging?

Or imagine someone were to ask you whether you would be willing to volunteer with an organization working with young offenders, and counsel the young people for a couple of hours each week during two years. That’s a big ask, and you would probably decline. But if the requester would then ask you whether you would chaperone a group of juvenile delinquents on a one-day trip to the zoo, what would you then say? Robert Cialdini and colleagues did exactly this in a 1975 experiment to explore the so-called Door in the face technique*. If you are anything like their participants you’d be three times as likely to say yes to the second request, than if all you had been asked to do was the chaperoning.

If a charity were to use this technique to increase the number of volunteers, would that be unethical? What if they use reciprocity (by sending you a free pen) or the identifiable victim effect (depicting and naming an emaciated child) with their request to donate, to make you pay more?

Careful now

We all have strong preferences for or against things. But we are also all often in situations where our preferences are weak or ill-defined. That is when we can be swayed by the environment, and in particular by the way in which it interacts with our cognitive biases.

But neither the swaying in itself, nor the fact that the outcome of the nudge appears to favour the nudger more than the nudgee should be seen as signs that something sinister is going on. Remember – as even the NYT article says – nudges happen “without a whiff of coercion”. People are always free to make any choice.

Does that mean nudges are not manipulative? No, they definitely are. But they are not inherently more so than more direct interventions. Everything that is aimed at changing behaviour, from a simple informative sign to bonus payments and threats of severe punishment, is manipulative. So we should not get hung up about this characteristic either.

For by their nature nudges are gentle. They do not ban any possibilities, nor do they materially affect the incentives of any available choice. That makes nudges pretty poor tools to manipulate people into doing something that goes against their self-interest. If you want to mislead, defraud or fool people, look elsewhere.

So nudges that actually reduce nudgees’ welfare may be rare. But if we insist that only nudges that demonstrably increase the welfare of the nudgee are ethical, we are being way too purist. It is not because psychology is not explicitly used to someone else’s benefit, that it is to their detriment. Much like the negotiations in the win-win domain of economic transactions, they tip the balance from one win-win to another one.

Let us not get all hysterical about “psychological trickery” – the ‘brilliant jerks’ at Uber may deserve criticism for many of its policies and practices, but their use of behavioural science should not be among them.

And let us bear in mind that it is not because we cannot see a clear win-win in a behavioural intervention that it is evil.

* Thanks to Richard Shotton for helping me turn a vague recollection into a precise description.


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

The problem with preferences

(Featured image: Images Money)

Can an objective measure like money help determine what we really, really want?

A cynic, Oscar Wilde wrote in his play Lady Windermere’s Fan, is “a man knows the price of everything, and the value of nothing”. This suggests that price and value are two different concepts, which apparently corresponds well with what we see around us. We generally don’t pay our hosts when we’re invited to a dinner party, but we take a gift. Neither the cost of the ingredients nor the price of the gift are particularly material here. We normally don’t ask friends to contribute to the cost of running our car when give them a lift.

Behavioural economics too seems to fit this perspective. One of its findings is that people appear to behave differently depending on whether they believe themselves to be in the market domain (where prices are essential) or in the social domain (where they don’t figure at all). An often quoted example is that of the Haifa daycare centre featuring in Uri Gneezy and Aldo Rusticini’s paper “A fine is a price”. Most parents habitually picked up their offspring before the stated closing time of the centre. They did so without any external incentive: social norms (and quite likely avoidance of guilt) were enough – but not for all of them. When a fine was introduced for picking up one’s child late, however, the number of parents who did so went up. They saw what was intended as a fine (a social intervention) as if it were a price (a market instrument), and simply ‘bought’ extra childcare time.

So there may be a social domain and market domain, but what the parents’ behaviour highlights is that there is a connection between the two. It seems to be possible to put a price on embarrassment or guilt – and indeed on a lot of stuff that is not quite in the market domain.

Under the heading  ‘The Worth’, Decision Technology regularly reports on their research into what people would be prepared to pay for certain weird and wonderful things. For example, 29% of respondents in one of their surveys would be willing to pay at least £20 to access the information that the Government holds about a person of their choice. More than 50% would pay that amount to have dinner with a guest of their choice.


The price of preference

Part of the implied criticism of the cynic in Oscar Wilde’s play is that some people focus only on money. But money does have an appealing characteristic: it provides an objective yardstick that allows us to indicate relative preferences. If we prefer the real thing instead of a low cost own brand, or a guitar that was once owned by Beatle George Harrison over a brand new model, then that is reflected in the higher price we are prepared to pay. If people are prepared to pay more for a house on a river bank, for a car with enhanced safety features, or for a live performance of an opera on a cinema screen rather than a recording, then that reflects their preference.

And as we’ve seen, that can work just as well outside the obvious market domain. We can ask others (or indeed ourselves!) how much we’d be prepared to know the date of our death (or indeed not to know it). We could ask how much more we’d need to earn if we had to spend another hour commuting each day. I am certain that with just a modicum of imagination you could come up with dozens more such intriguing questions.

What we want, what we really really want

Does using money as a yardstick to measure our preferences really give an accurate picture of what we want? A paper by economists Gerardo Infante, Guilhelm Lecouteux and Robert Sugden casts doubt on that idea. In it, they scrutinize the central assumption of neoclassical microeconomics that people have stable and consistent preferences, independent of context or circumstance.

This assumption implies that we somehow know the value of everything and anything. In order to consistently express a preference of one thing over another, we have to know the value (to us) of each one. And if we have no preference and we’re indifferent, then their value is the same.

That looks not unreasonable at first sight. But can we really maintain a coherent set of preferences on this basis? If you know what an apple is worth to you, and you know what a new car is worth to you, you should be able to express the value of the car in terms of a number of apples. However, you also know the value of a haircut, a pair of socks, or of your broadband contract and so on. So you should be considering all these equivalences whenever you’re presented with a choice. It would likely drive you crazy in no time.

So we have ways of limiting the need for such an astronomical number of possible equivalences, like the use of money, and of mental accounting, by putting items in categories. Even if we cannot consistently compare apples with cars or socks, as long as we’re consistent in comparing apples with cream cakes, we’re fine, right?


My true preference… or is it? (photo: Piotr Adamiak)

Certainly – at least if deep down, we are rational and we know what we really, really want. Our problem would be that we make ‘irrational’ decisions because we are fooled by external (choice architectures) or internal (inappropriate heuristics or beliefs) influences.

But Infante et al argue that, even for a Superreasoner, a hypothetical being with the intelligence of Einstein, the memory of Big Blue and the self-control of Gandhi, it is not possible to establish a context-independent preference between a piece of fruit and a cream cake. There is no reason why Kahneman’s calculating and logical System 2 is the real keeper our true preferences, and that decisions we make in a cold state are ‘better’ than those we make in a hot state under the influence of the emotional and impulsive System 1.

The consequences of unknown preferences

All this explains why and how we can be manipulated. People who want to sell us stuff can influence our preferences. If we don’t always prefer apples over cream cakes (or vice versa), then by arranging them in a particular way (the choice architecture), or by advertising them in a particular way (framing), we can be nudged into choosing one over the other.

This is not new, of course. Marketing people have long known that shiny apples and brightly coloured cans of pop sell better than dull ones. And there is nothing inherently improper about using these insights. We know full well that someone trying to sell us something is doing so in their own interest in the first place.

It does raise a problem regarding paternalistic interventions, though. These rely on the assumption that the preferences of the person being nudged are known. If it turns out that these cannot be known, can governments really be deemed to act in our interest?

Governments may perfectly legitimately seek to improve public health because it keeps healthcare expenditure under control, or to encourage people to use public transport because it reduces traffic congestion and the corresponding cost to the economy.  If it can nudge us collectively to change our behaviour to make this happen: great.

But that is quite different from claiming it is acting in our individual self-interest. If we don’t know our own preferences ourselves, isn’t it a bit arrogant for government to pretend they know?

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