Misunderstanding rationality

(Featured image credit: Les Colporteurs CC0)

It is tempting to equate ‘rational’ with ‘right’, and ‘irrational’ with ‘wrong’, but that is mistaken

The lottery is sometimes called a tax on the stupid. The logic behind this opinion is that the expected net gain is negative. In a post entitled ‘The irrationality of Euromillions’, blogger Robert Sharp works out that buying a £2 ticket is only economically rational when the jackpot is £190 million or more. Yet I am pretty sure that the anonymous Belgian who pocketed £134 million on 2 June (the highest Euromillions win so far this year) will not be too concerned about whether the purchase of that ticket was irrational or not.

Another situation that is often wheeled out as an example of irrationality is fear of flying. The headline of an Independent article  from 2009, “Why fear of flying is just plane stupid”, is pretty blunt. And the facts seem to support it: the risk of dying in a car crash is 100 times higher than that of perishing in an air disaster.

Cognitive errors…

It is hard to estimate with any accuracy how many people choose alternative modes of travel out of fear, but the terrorist attacks of 9/11 led to a large-scale natural experiment that does shed some light. In the months following the events, there was a marked drop in the number of passengers across the USA, and an increase in the number of road traffic casualties. This is strong evidence for the thesis that some people swapped the car for the plane. The estimate of the number of ‘excess’ traffic deaths in the last three months of 2001 varies a bit – 344 (Blalock et al), 353 (Gigerenzer), and 1018 (Sivak and Flannagan) – but it over that period it is significantly higher than the 266 passengers and crew who were killed in the planes.

We often take decisions based on what we see as their possible direct consequence. The lottery ticket buyer has the fabulous prize in mind, the person avoiding air travel is guided by the horrible prospect of dying in an air disaster – irrespective of the likelihood of either event.

Rationalists would be quick to point out how this relies on two common apparent cognitive errors. The salience effect or saliency bias refers to situations where our actions are influenced by what is highly visible and prominent. This makes both the lottery player and the aviophobe overestimate the likelihood of the desired (or undesired) event. Outcome bias makes us evaluate the quality of a decision on the basis of its result. If you win the jackpot, you’re unlikely to say you made the wrong choice buying a ticket. You may even attribute your good fortune to the clever way you decided to choose your numbers. And if you never fly as a matter of course, whenever there’s a crash you will be relieved that could never happen to you.

…or not?

But does that mean these choices are truly irrational? Not really.

For just by looking at the choices we don’t know what reasoning is behind them. If a lottery player’s cost is negligible to them, and the thrill of the draw provides ample enjoyment in return, there is really nothing irrational about their behaviour. Likewise, if avoiding the sheer terror of sitting in a steel tube 30,000 feet up in the air is well worth the inconvenience (and added risk) of driving long distance, and of not being able to visit another continent, that is a perfectly rational course of action for someone with fear of flying.


“You’ll never get me into a plane” (picture: David Eun CC0)

The normative evaluation of how we make decisions seems to have been strengthened by Daniel Kahneman’s popular book Thinking Fast and Slow , in which he describes a model of two modes of thinking. The Wikipedia entry tags System 1 as “emotional, stereotypic, and subconscious” (traits we associate with impulsive, poorly thought through decision-making). System 2 is “logical, calculating, and conscious” (characteristics of wise, reasoned and well-considered decision-making). These terms barely conceal the implication that the latter is superior to the former.

Yet we have survived and prospered for tens of thousands of years with both systems working in tandem, making mostly the right trade-off between the speed of one and the thoroughness of the other. That demolishes any suggestion of inherent superiority.

Rationality is a matter of preference

Acting rationally also does not imply any moral rectitude. The work of economist Gary Becker, crystallized in his essay Crime and Punishment: An Economic Approach, shows that it is perfectly possible to be rational and criminal at the same time. David Friedman, a law professor at the University of Santa Clara, summarizes it clearly in a chapter in The New Economics of Human Behaviour: “Criminals are rational. A mugger is a mugger for the same reason I am a professor – because that profession makes him better off, by his own standards, than any other alternative available to him.”

And the rationality in the behaviour of a mugger relieving a passer-by of the contents of his wallet is not different from the rationality of, for example, the Member of Parliament voting for an income tax increase on the well-heeled. Both extract resources from someone else to further their self-interest.

So if we are tempted to look down on someone because we believe they are acting irrationally, we really ought to pause for a moment, and consider not what they’re doing, but why they’re doing it. When we see someone buying a lottery ticket, and on that basis alone decide they’re not being rational, we’re arguably being influenced by outcome bias ourselves. Unless we really know people’s preferences, we cannot judge their rationality.

Whether what they’re doing is right… that is a matter of our preferences.


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

What’s the price?

(featured image: pexels)

Internet companies are trying to charge different prices for the same thing depending on who you are. Is that a problem?

Prices are a funny concept. We unthinkingly treat them as if they are a physical attribute of an item – like the weight or the height. To some extent that is not even so crazy. Even if we know that they can be a bit fuzzy, we’ll have an approximate idea of what something costs (certainly for stuff we buy regularly), and so what we’re prepared to pay. Or do we?

In his 1983 paper Transaction Utility Theory, behavioural economist Richard Thaler describes a thought experiment that goes something like this. You’re on the beach on a hot day with only iced water to refresh you, and for the last hour you’ve been imagining a nice cold bottle of your favourite beer. Your friend next to you gets up to make a phone call (this is 1983!) and offers to bring you a beer from a place close to the phone box. But of course, beer at the beach might be expensive, so he asks you the maximum price you’re prepared to pay, and promises to only buy you a beer if it costs no more than your limit (which economists call the reservation price). What would you be willing to pay?

Beer money

When it was stipulated that the only place selling beer was a fancy hotel, people were on average willing to pay $2.65; when it was a small, run-down grocery store, their average limit was just $1.50 (1983 prices of course). Thaler points out that this is inconsistent with standard economic theory. There is no material difference between both situations that would explain the difference, let alone one of this size.

So, we don’t really know the value of a beer – nor of pretty much everything else we regularly, or seldom, buy. Prices are emergent. In an individual transaction both parties may bargain until they settle on a price that both are happy with (and there is a whole range that would qualify). In a larger market, economists talk of the market clearing price, the price at which supply and demand are exactly equal.

This bugs suppliers, though. Obviously, at this equilibrium price, those for whom it is too high will not buy. But there are lots of people who would be prepared to pay more than the market price, but who don’t need to. The supplier is leaving money in the customer’s pocket.

So inventive sellers come up with ways to separate price sensitive customers from those with a higher reservation price. Three decades ago, for example, it was rumoured that companies supplying both computer equipment and medical systems used to charge a hefty premium for PCs for hospitals, simply because they were painted white instead of the drab Pantone 413 beige that was the norm back then. That may be a tall story, but one industry that has long been successfully using price discrimination instruments is airline travel.

The art of discriminating

An aeroplane is essentially a bunch of seats flying from A to B. And yet the people whose bums are on those seats may have paid a price that differs by a factor 20 or more. Low cost airlines correctly assume that people who need to fly at short notice will be prepared to pay more. So prices start low if you book months in advance, going up as the date of travel gets nearer. Premium airlines offer different classes (while often the seats are exactly the same), but more importantly, they try to distinguish cheapskate tourists from business travellers with deeper pockets by restricting cut-price tickets for travellers staying a weekend.


Cheap and expensive seats – can you tell which is which? Picture: StelaDi CC0]

One company (no longer in existence) was even more creative. GetGoing.com asked you to pick two different destinations. You didn’t know which of the two you’d be flying to until the moment you paid. This kind of offer would clearly only sensible for a price-sensitive holidaymakers, with no risk that even the most astute among people prepared to pay full whack would snaffle a huge discount. This is also why you see student or senior citizens’ discounts, reduced-price train tickets for families and so on.


Now with big data, of course, companies can simply rely on user profiles to figure out whether we’re a tightwad or a big spender. Amazon was exposed as early as 2000 using cookies to set the price according to whether you were a regular or a new customer. They apologized and vowed never to test prices based on customer demographics. But even if Amazon is lying low in this respect, others have happily been playing around with price discrimination, according to a study by Christo Wilson and others at Northeastern University.

Most recently, Uber (who else?) has been in the news again, as they too are looking to charge you what they think you’re willing to pay. And perhaps Amazon too is back dabbling with price discrimination, as economist Cyril Morong wonders, looking at their latest move in their battle with Walmart where they offer a discount to their Prime service if you are receiving government assistance.

Do those who cry foul at this kind of manipulative trickery have a point? Few people seem to have a problem with more conventional forms of price discrimination. On this basis, the criticism of more modern approaches seems a bit inconsistent.

Consumer power

That may be the case, and perhaps it is indeed economically sensible, rational and perfectly legitimate for companies to seek ever more sophisticated ways to channel as much of the economic surplus of a transaction into their coffers. That is not without danger though. Misguided as the suspicion of consumers may be, a company that is seen as manipulative and devious risks losing its reputation, and its customers to the competition.

But what if the competition too resorts to ever cleverer price discrimination methods? Will we eventually all pay our reservation price for everything we buy?

It is unlikely we will get to that point any time soon. A Bloomberg article about Uber’s latest ploy hints at ways in which riders can fool the algorithms that try to guess our willingness to pay: feed it confusing signals. Open the app at random and check prices, even if you have no intention to ride – just to make yourself look price-sensitive.

We consumers are on the whole a reasonable and tolerant lot. We don’t mind a bit of advertising, and we’re OK with websites using cookies – even if it is to help companies make more money. But when we think this kind of stuff is going too far, we strike back. In with the ad blockers and with our browser’s private mode to prevent sites tracking our moves.

And so it is more generally with the experience we get from internet companies. It’s great that we can take a ride with Uber, but any company that underestimates the resourcefulness of people who feel they’re actually being taken for a ride does so at its own risk and peril.

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

A tale of two tolls (behavioural economics and ethics edition)

(featured image: S Lowe)

Why do we happily pay some tolls, and not others? And how ethical is it to encourage people to take the toll option?

On the way back from a few days at the seaside, Google informed us that traffic ahead was getting worse, so we swapped the A road for the B-roads. The trade-off seemed straightforward enough: we would be saving an estimated 20 minutes by getting off the main road. But in general the trade-off is more complex.

As Wikiman Rory Sutherland eloquently argues, taking the rural road, even if the main road is clear and faster, may make sense. When you find yourself unexpectedly in heavy traffic on the motorway, unless you have an exit just ahead, you’re really stuck. If you need to catch a flight, opting for a route that is nominally slower, but which has less risk of catastrophic delays is a clever move.

And of course, the duration of your journey may not be your main concern. B-roads can be a much more pleasant and relaxing drive. You’re generally in more interesting surroundings, and you may even make unexpected, quirky discoveries. So it was during our unplanned detour.


Picture: David Smith

It is not unusual to come across unusual little octagonal buildings along England’s roads. These are former tollhouses, located at roads and bridges up and down the country where the land owner was allowed, by law, to collect a fee from users. Almost all of them are now quirky dwellings, but England wouldn’t be England if it didn’t still have some active tollhouses. One of them (which has a more conventional shape) is that at Swinford Bridge, a crossing over the river Thames near Oxford.

Pedestrians and cyclists can cross free of charge, but for cars there is a charge of 5p. (That may sound a laughably small amount, but Oxfordshire County Council estimates about 10,000 vehicles use the bridge every day. That generates a revenue of more than £500 per day – ample to leave a nice, tax-free profit after paying two people to collect the money.)


A complete bargain (photo: Simon Li)

The toll bridge is popular because the nearest free bridges upstream or downstream are about 7 miles away. 5p (even if you need to queues for a minute at the toll gate) is a small price to pay to avoid the diversion. But are we always this rational with such choices?

Toll aversion

Aside from historical relics like this, there are other tolls in the UK, including the relatively recent toll road North of Birmingham, built to relieve one of the most congested motorway stretches in the country. The original M6 was designed for 72,000 vehicles per day, but by the late 1990s had to cope with peaks of up to 180,000.

The 27-mile long M6 Toll now carries roughly a quarter of the total traffic now (in 2016 it averaged 48,000 vehicles per day; for the M6 the figure was 122,000).

It’s hard to say whether that is an appropriate ratio. But the toll road also has a capacity of 72,000 vehicles per day, so it is well underutilized. And despite having since gained a fourth lane, the M6 is still congested during rush hour, while the M6T is always clear. Why are not more people diverting to the toll road? Is it because the economics don’t stack up?

For sure, the M6T toll, at £5.50 for a car, is over 100 times that of little Swindon Bridge. How much time would one need to save for this to make sense? The 27 miles typically takes 24 mins on the M6T. On a busy M6 that can easily take half an hour more, and if there’s an accident at Spaghetti Junction it could be even more.

Time stuck in stop-start traffic is pretty much totally wasted. Is that half hour worth £5.50?

That depends on how you frame it. The headline price is very salient – the signs along the M6 advertise it in big print. That may put people off. Nothing or £5.50? Easy choice – especially if what you get in return is not clear. At Swinford bridge you know you’re certain to avoid a 15 minute, 7 mile detour. Here it’s much more speculative.

If you consider the actual cost of driving 27 miles, that looks a bit different though. In the UK the tax people consider a reimbursement of 45p/mile fair compensation for all the driving costs that does not constitute a taxable benefit. On that basis the comparison now looks like this: pay £12.15, or pay £17.65 and (maybe) save half an hour. That seems a better deal than £5.50 versus nothing.

m6-tollBut few people make that kind of calculation. We are more likely that to construct a narrative on a foundation of optimism bias: it will be fine today. Or it won’t be that bad, really. Even if the messages on the overhead gantries try to influence us, they’re hardly persuasive.


Just knowing that there is congestion is not enough to make most of us reconsider – optimism bias is a very powerful force. But what if the duration of the journey either way was shown? Imagine the sign said something like these two signs?


The first one gives drivers the facts – much better than ‘congestion’ and ‘clear’. The second one goes even further, and shows how much time you would lose, swapping the loss frame from the payment to the lost time. Behavioural economics would suggest that this might get more people to take the T-road.

What about the ethics?

Some people might question the ethics of these hypothetical signs. Clearly they would benefit the operator of the (currently actually loss-making) M6 Toll road. But does that make it unethical? A driver opting to go down the M6T also benefits.

We generally don’t see anything unscrupulous in salesperson highlighting the benefits of what she is selling to the prospective buyer. And that is precisely what the fictitious overhead signs would be doing. Nobody is surreptitiously lured into something they don’t want to do – they simply point out the difference.

And yet, somehow that phraseology seems manipulative.

But before we condemn the trickery of behavioural economics, perhaps there is something else to consider. Both Swinford bridge and the M6T save their users time, and both provide additional societal benefits: lower pollution through shorter journeys or less congestion.

But despite its low toll, Swinford bridge is very profitable (it sold for over £1 million in 2009 and brings in a gross revenue of nearly £200,000 each year). And that is mostly economic rent – simply exploiting the ownership of the piece of land on which the bridge sits.

So maybe the ethics of the little bridge at Swinford and its tax free profit are actually more questionable than the hypothetical use of behavioural economics to invite people to use the M6T.

In (behavioural) economics, things are not always what they seem.

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

Coase on the coast

(featured image: Adam Piggott CC0)

Should we negotiate more with our fellow humans for a better life?

A couple of weeks ago, I spent a few days on the English south coast, in unexpectedly clement weather. While we had brought weatherproofs and an umbrella, we ended up having to go and buy sunscreen to avoid incinerating our skin as we sat on the seafront, eating our crab sandwiches.

Several other people were sitting alongside us on a row of benches, each enjoying in their own way the warm sunshine and the gentle bubbling of the surf. To our far left was a middle-aged lady, engrossed in a rather fat book, to our right a retired couple drinking tea from a thermos. Then two women appeared – mother and daughter, it seemed, to occupy the bench between us and the woman with the book. The former was sharing an ice cream cone with her dog, the latter was talking rather loudly on her mobile phone.

After maybe ten minutes, all of a sudden, the book lady got up, trotted over to the two women next to us and complained in an agitated tone about how the phone conversation made it impossible to concentrate, gesticulating with her free hand. Then she shoved her book in her bag and stormed off.

Dirty socks

What we had been witnessing was the manifestation of what economists call a negative externality. That is a cost incurred by someone who was not involved in the action or transaction that imposed it. (There are also positive externalities, when a third party enjoys a benefit resulting from an action or transaction without taking part in it.) Often it is society as a whole that gets lumbered with negative externalities. A factory which, in the process of producing widgets to sell, pollutes its surroundings, imposes a cost on the people living in its vicinity: maybe people contract illnesses as a result, or if not that, there is a constant smell, or the deposits from the chimneys make the land of their vegetable patches infertile.

But externalities happen all around us, not just in the real economy, but in households (leaving the lights on, or the dirty socks on the floor), workplaces (interminable procedures for travel approval) and indeed the Dorset seafront.

In 1960, Ronald Coase wrote a classic (and very readable) paper on the problem of externalities, The Problem of Social Cost. (It would contribute to his being awarded the Economics Nobel prize in 1991.)  Being an economist, Coase was concerned with the efficient use of scarce resources, and he approached the problem of externalities as one of bargaining between the participants. Provided transaction costs were sufficiently low, negotiation between the different parties would – as elsewhere in markets – lead to an efficient solution.


A sharp mind in a strong body – Coase aged 98 (Source: University of Chicago Law School, via Wikimedia Commons)

Coase died in 2013, aged 102, but imagine he had sat next to us watching the tide come in. He might have explained to the ladies that they could have negotiated either a price for the right to yak on a mobile phone, or a price for the right to peace and quiet, and compensate the other party accordingly.

That may sound bizarre, but it does happen in practice. An boss ordering pizza for her team working late is actually compensating them for a negative externality – getting home late. The negotiation may be implicit here, but it can be pretty unequivocally present as well. To a kid who would rather go to a friend’s birthday party, being taken by their parents on a boring visit to great-aunt Sal is very much a negative externality. And the discussion leading to parental concessions to ensure junior behaves on the trip is very much a negotiation. Some might be inclined to talk about bribing and blackmail, but of course the embarrassment of misbehaving offspring is a massive negative externality for parents, well worth some insurance premium.

Costs and payments

Still, we don’t often go about negotiating with strangers about the externalities they (or we) cause. One explanation Coase offers is that the transaction cost is often a barrier. Operations like discovering who to deal with, informing them one wishes to deal and on what terms, conducting negotiations leading up to a bargain, drawing up the contract etc, Coase says, “are often extremely costly, sufficiently costly at any rate to prevent many transactions that would be carried out in a world in which the pricing system worked without cost.”

But it’s not as if there is no ‘cost’ involved when we don’t negotiate. The lady with the book suffered an emotional cost. She went over to the lady babbling on her phone to tell her exactly what she thought of the breach of peace – while she could have opened a negotiation.

Maybe it’s the idea of turning what we see as social interaction into a market transaction that repulses us? In a 2014 Slate article, Christopher Buccafusco and Christopher Jon Sprigman, two US law professors, described their investigation into the problem airline passengers fighting over the use of reclining seats. When the person in front puts their seat back, that takes away valuable knee space for the person behind them. The authors looked at whether passengers could be encouraged to negotiate.

Their surprising finding was that people were more inclined to trade the space if what was being exchanged was not money, but small gifts of equivalent value. This violates standard economic theory, which holds that getting a gift should never be better than getting the gift’s value. But it is consistent with human nature: when we’re invited to a dinner party, we take wine or flowers, rather than transferring the equivalent amount to the host’s bank account.

Nonetheless, getting people to negotiate about seafront peace and quiet seems even harder than getting them to do so on aeroplanes.

The Coase in the machine

But on the way home, driving along a 2-lane dual carriage way, continually being held up trucks driving at 59mph overtaking  one driving at 58mph, made me think. Could we delegate the negotiations we humans seem to find costly or embarrassing to technology?


“Paid enough? Go ahead, overtake!” – photo: Phil Hollenback CC0

Just imagine a lorry that can communicate with the cars behind it, allowing the driver to offer a certain amount of money to hold them up while it overtakes a slow vehicle in front. Upon successful negotiation, the money is transferred and the driver will be able to pull out and start the manoeuvre. If the drivers behind are unhappy with the offer, the truck will stay in the inside lane and allow them to pass – and the negotiation can start with the next bunch of cars. (Alternatively, the car drivers could pay the lorry driver to stay in his lane of course.)

Obviously, in a few years’ time we’ll all be riding in autonomous vehicles, and we won’t even be involved in such negotiations. But how will our cars decide for us? Much thought has been going into figuring out how self-driving cars should react in case of an unfolding accident. Should they safeguard the occupants, or other road users – elderly people, children, bank robbers. A team at MIT developed a cool simulation to see how people think machines should make decisions involving such moral dilemmas.

But perhaps the trolley problem is not the only decision-making challenge for autonomous vehicles. Maybe we need to provide them with the means to negotiate with other vehicles depending on whether we’re in a hurry or on a leisurely trip.

And so the insights of the great Ronald Coase’s might continue to serve us well into the 21st century, even as we delegate more and more decisions to machines.

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

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 Behavioural economics, Cognitive biases and fallacies, politics | Tagged , , | 1 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