Tim Harford The Undercover Economist
  • THE UNDERCOVER ECONOMIST STRIKES BACK
    “Every Tim Harford book is cause for celebration. He makes the ‘dismal science’ seem like an awful lot of fun.”
    – Malcolm Gladwell, author of “The Tipping Point”
  • Adapt – Why Success Always Starts with Failure


    “A highly readable argument... Very impressive”
    – Nassim N Taleb, author of “The Black Swan”
  • Dear Undercover Economist
    “Harford has a knack for explaining economic principles and problems in plain language and, even better, for making them fun”
    – The New York Times
  • The Logic of Life
    “Highly readable, funny and daringly contentious... a whopping good time.”
    – The San Francisco Chronicle
  • The Undercover Economist
    “Required reading”
    – Stephen J. Dubner, co-author of “Freakonomics”
Undercover Economist

Metropolitan myths that led to Brexit

The Eurosceptic myths that fuelled hostility to the EU are obvious enough. We were told that the NHS was being destroyed by immigration, when more than a third of UK-based doctors qualified overseas. We were told that the EU is a fat bureaucracy, when it employs about as many people as Lancashire County Council. And we were told that the UK was being buried in red tape, when the OECD reckons it is one of the least regulated economies in the developed world.

It is easy (and useless) to sneer. Yet the metropolitan elite that voted so enthusiastically to remain cherishes its own myths, and those myths did plenty to undermine the cause of remaining in the EU.

Here are four tenets of the trendy centre-left of British politics: first, soaring inequality means that ordinary people haven’t shared in the benefits of economic growth; second, rich people and big companies don’t pay taxes; third, gross domestic product (GDP) is a statistic that misses what really counts; and finally, economists are reliably wrong. Flip through The Guardian, browse the popular economics books in your local bookshop, and tell me that these ideas aren’t taken for granted among the chattering classes.

Before the referendum, Anand Menon, director of the “UK in a Changing Europe” project, was speaking at a town hall event in Newcastle. He explained that most economists thought Brexit would depress the UK’s GDP. “That’s your bloody GDP,” yelled a heckler, “not ours”.

Look again at the four articles of centre-left faith. If they are true, then surely the heckler was right. But while there is a little truth in each of these four beliefs, there is less than you might think.

It is true that income inequality in the UK rose very sharply during the 1980s. But by most measures it has been pretty flat since 1990. The top 1 per cent continued to pull away in the 1990s — although not this century — but, counterbalancing that, the gap between people at the 10th percentile and the 90th percentile actually fell between 1990 and 2013-14. Broadly, income inequality is a problem that emerged in the 1980s and has not worsened since. (The Institute for Fiscal Studies report Living Standards, Poverty and Inequality in the UK: 2015 was my source; the 2016 version has been published since this column went to press.)

The pressing issue for the UK has not been rising inequality but weak growth that has affected most people not only during the recession of 2008, but in the five years before and after it. The problem is not that income growth benefits only the rich. The problem is that there’s been little income growth to benefit anyone at all.

The second article of faith is that rich people don’t pay taxes. If true, it would hardly matter to ordinary voters if Brexit hurt the rich or drove them away.

But the richest 1 per cent of taxpayers pay nearly 28 per cent of income tax. And, with about 9 per cent of post-tax income, they presumably also pay about 9 per cent of VAT, which is close to being a flat tax. Of course, some other taxes are grossly regressive — most notoriously the council tax, which the European Commission did urge the UK government to reform — but the rich certainly pay enough tax that the public purse would sag if they disappeared. London, too, generated more than 25 per cent of UK taxes, and that proportion has been rising, according to Centre for Cities, a think-tank. After Brexit, who knows?

What about the idea that GDP itself is flawed? Well, yes. It is flawed. It measures things that do not matter and misses things that do. But a sharp hit to GDP will also be a sharp hit to our everyday wellbeing: people will lose their jobs; schools, hospitals and public services will be squeezed as tax revenue dries up.

Consider an alternative measure: the Social Progress Index, an attempt to measure what matters in global human development with more than 50 different indicators — including access to nutrition, healthcare and schools. The SPI explicitly excludes financial indicators. Yet there is a very high correlation between the SPI and GDP. (For my fellow nerds: Michael Green, director of the SPI, tells me that the correlation is 0.88 when GDP is measured on a log scale. That’s high.) As a measure of human welfare, GDP completely fails in theory. In practice, however, it is not such a bad yardstick.

Finally, there is the low reputation of economists, the result of a global financial crisis that only a few in the profession warned us against. But the institutes that analysed the risks and rewards of Brexit can hardly be blamed for that. The Institute for Fiscal Studies is full of experts on tax and household income; the Centre for Economic Performance studies globalisation, technology and education. Blaming these people for not foreseeing the collapse of Lehman Brothers is like blaming a brain surgeon for the spread of obesity.

Many of the people who rightly scorned the myths put around by Eurosceptics should examine their own fond beliefs. The lesson of the referendum campaign was that emotion trumps rational analysis — and that is not just true of the Leavers.

Written for and first published at FT.com

Free email updates

(You can unsubscribe at any time)

Undercover Economist

Brexit and the power of wishful thinking

‘We are quite capable of clinging on to our beliefs by picking whatever facts support them’

So what happens now? As the aftershocks of the Brexit vote rumble on, people keep asking me to predict the future. I am not sure that is a useful exercise. But there’s a lot to be learnt from examining the recent past. The unexpected result has important lessons for anyone who wishes to see what might be ahead.

The first is that wishful thinking is surprisingly powerful. A few years ago, the economist Guy Mayraz conducted a simple experiment at Oxford university’s Centre for Experimental Social Science. Mayraz ran sessions in which the participants were shown 90 days of “wheat prices” (actually based on historical price data) and asked to predict the price of wheat on the 100th day. In addition to being paid for accurate forecasts, half the experimental subjects were told they were “bakers”, who would profit if the price of wheat fell, and half were “farmers”, who would make money if the price of wheat rose.

Logically, a farmer should make the same forecast as a baker, since the forecast does not change the outcome, and both of them are paid for accuracy. If people want to hedge even these small bets, farmers ought to forecast a lower price so that if the unwelcome outcome happens, at least they’re rewarded for their forecasting.

But that’s not what Mayraz found. Instead, nearly two-thirds of farmers predicted higher-than-average prices, and nearly two-thirds of bakers predicted lower-than-average prices. People tended to predict that their dreams would come true.

The same seems to have happened to Remain supporters — a group that included most of the British and international political and business establishment. Months before the vote, betting markets suggested that the chance of a Leave vote was about one-third. Given the likely consequences of such a vote — including the collapse of the pound, the resignation of the prime minister and a prolonged period of rudderless uncertainty — a one-third chance was worth taking seriously. Yet most of the elite seemed unwilling to countenance it. That is wishful thinking at its finest.

The betting markets, influenced as always by the weight of money, may have been displaying some wishful thinking of their own. Even during a mid-June run where nine out of 10 polls showed the Leave side ahead, the markets never gave Leave much more than a 40 per cent chance. With hindsight that seems odd.

The second lesson is that confirmation bias is everywhere: we are quite capable of clinging on to our beliefs by picking whatever facts support them. Remain voters now see the catastrophe they expected; Leave voters see a gratifying shake-up that will turn out fine in the end. Contrary evidence is easily dismissed. Eurosceptics claim that a drop in high-street footfall, a drying-up of job adverts and the emergency lockdown of property funds are not because real economic damage is in prospect, but because of gloomy Europhiles talking Britain down. The convenient thing about this argument is that it can never be falsified: recessions can always be blamed on the lack of faith of unbelievers.

I’m as guilty as anyone of confirmation bias. Like most economists, I expected that the consequences of a Leave vote would be ugly. The immediate collapse of the pound and the FTSE 100 made me feel I was right, although really they demonstrated nothing more than the fact that traders shared my view. Yet when the FTSE 100 recovered, that did not reassure me at all. I recalled that the index contains largely global companies and says little about the UK’s economic prospects. I accepted bad news when it chimed with my beliefs, and dismissed good news when it did not.

Perhaps the most important lesson is that we spend too much energy trying to foretell the future, and too little trying to be resilient whatever happens. The referendum result was unpredictable but the likely short-term consequences of a Leave vote were perfectly clear. Former deputy prime minister Nick Clegg, aka “Mystic Clegg”, outlined them in an article on the eve of the referendum that now seems clairvoyant. But most of what Clegg wrote could have been foretold by any well-informed observer who bothered to think through the consequences.

Perhaps Clegg learnt the trick of thinking things through from Vince Cable, his former colleague. Cable and I briefly worked together in the scenario planning department at Shell, a fascinating place to think about the future. Rather than making forecasts, good scenario planners sketch out different possibilities and bring together people with different perspectives to work through the details. The end result will be several plausible, internally consistent and emotionally compelling stories about the future. The scenarios will highlight hidden connections and make distant consequences seem real. But, importantly, the scenarios will also contradict each other.

It’s time for more serious scenario thinking about the UK’s future in Europe. Because scenarios are persuasive stories, they can help us face up to uncomfortable prospects and think clearly about possibilities we would rather ignore. And because scenarios contradict each other, they force us to acknowledge that, in the end, we cannot actually see into the future. As a result, we move from a sterile question to a fertile one — from “What will happen?” to “What will we do if it does?”

Written for and first published at ft.com.

Free email updates

(You can unsubscribe at any time)

Undercover Economist

Did economists fail us over Brexit?

‘Most economists think the British public shot itself in the foot, and did so against expert advice’

You may have heard the jokes about economists equivocating and squabbling. Ronald Reagan had the best one, about an edition of Trivial Pursuit designed for economists: “There are 100 questions, 3,000 answers.”

But mainstream economists did not disagree on the EU referendum question. A fortnight before the British public voted by a slim-yet-definite margin to leave the EU, the Centre for Macroeconomics, a research group, polled a group of expert economists, asking them what they thought the overall economic consequences of such a vote would be. The response was unanimous, and the only point of contention was whether Brexit would be bad, or very bad.

Leave voters did not seem to take this unprecedented warning very seriously; nor were they moved when nearly 200 economists signed a letter to the Times; nor were they dissuaded when three of the country’s leading economic institutes issued a sober analysis in a joint statement. All these were independent views, separated from the sometimes-political world of the International Monetary Fund or the Treasury. Yet they were ignored.

In short, most economists think the British public shot itself in the foot, and did so against expert advice. So should the profession have done things differently? Or would further expert views have been drowned out in the media maelstrom?

Most newspapers had a strong preconceived view, for or against. The BBC attempted to be impartial, but many critics complain that the corporation gave too much airtime to the fringe views of a few pro-exit economists, and failed to challenge obvious falsehoods. (I should declare my own BBC role: I presented a short radio series fact-checking the referendum campaigns.)

But Paul Johnson, head of the well-respected Institute for Fiscal Studies, thinks economists have a case to answer. “It is always a mistake simply to look at the media as a scapegoat. The real failings were with my profession.”

Perhaps. But the campaign left little room for subtle analysis of the difference between a single market and a free-trade agreement, or the “lump of labour” fallacy that each immigrant worker deprives a local of a job. Even the baldest falsehoods could not be expunged.

The Leave campaign’s most prominent claim — “We send the EU £350m a week, let’s spend it on the NHS instead” — was untrue and exposed as false repeatedly by the head of the UK Statistics Authority. Yet, if a recent Ipsos Mori poll is to be believed, 47 per cent of people came to believe it.

John van Reenen, the outgoing director of the LSE’s Centre for Economic Performance, doesn’t think the profession should be too down on itself. “I’m proud of what we did,” he says, and had economists engaged more “in my frank view, it would not have made a jot of difference.”

Van Reenen has a point; there is a limit to how much experts can achieve by simply presenting the facts.

Dan Kahan is a Yale law professor who studies the way we debate controversial political issues such as climate change and gun control. He points out that our reasoning about such issues is often bound up with passionately held emotions and values. For example, people who oppose government regulations are often instinctively sceptical about climate change, fearing that it is a Trojan horse for state control; alternatively, people who distrust large corporations instinctively embrace the scientific evidence on climate change, yet tend to dismiss scientists who say that genetically-modified food is safe. Giving people evidence that threatens deep beliefs is often counterproductive, because we start with our emotions and trim the facts to fit them.

A possible solution, argues Kahan, is to recruit experts who confound people’s stereotypes — a sharp-suited City economist explaining that EU laws protect workers’ rights, perhaps. Another possibility is to acknowledge people’s values before presenting the evidence — for example, the desire to preserve a community’s character in the face of immigration. And Kahan and his colleagues have also found that when evidence is presented by a diverse group of experts, people receive it with a more open mind. Perhaps the problem is that economists don’t seem very diverse.

Ultimately, there is no substitute for sustained public engagement — a lesson scientists have learnt the hard way. After years of bruising anti-scientific backlashes, they realised they were being ignored on issues such as genetically-modified crops or vaccines because they shied away from media appearances. Charlatans and activists stepped into the vacuum.

So, in 2002, the Science Media Centre was established, with a mission to ensure that when science or pseudoscience hit the headlines, respected scientists would be available to speak to journalists. “It’s a long slog,” says David Spiegelhalter, a statistician and risk-communication expert at Cambridge university. But he says the SMC has led to better science reporting.

Should economists do likewise? Perhaps. But the conduct of the campaign will not make that easier. Economists who did step forward to explain the issues were dismissed as corrupt or incompetent in the crudest terms. Nobody could be blamed for wishing to duck such mud-slinging. But if the events of the past few weeks have shown us anything, it is that public life is too precious to be entrusted to politicians alone.

Written for and first published at ft.com.

Marginalia

What I’ve been reading in June

In between all that Brexit, I’ve read some cracking books of late.

TED Talks by Chris Anderson. I wrote a piece about the book here; as a person who devours books about public speaking, this is by far the best I’ve seen. Bravo. (US) (UK)

The Shock of the Old by David Edgerton. (US) (UK) Fascinating take on the history of technology, pointing out that the innovations that make the headlines aren’t necessarily the ones that really make a difference. All hail to concrete and the bicycle!

On a similar topic but a more upbeat vein, How We Got to Now by Steven Johnson is jolly good. Much I did not know here. (US) (UK)

Other People’s Money by John Kay is a magnificent book – witty, penetrating and wise. All about the gap between what the financial sector does and what it could and should do – but by someone who really understand the sector in some depth. (US) (UK)

Algorithms to Live By by Brian Christian and Tom Griffiths. (US) (UK) This one is very good too: using computer science to help understand how to get a date, how to sort a bookshelf, how to find an apartment, how to tidy your desk, etc. etc. You learn a lot about computer science and a fair bit of self-help too. It’s not quite as good as Brian Christian’s The Most Human Human (US) (UK) but since that’s perhaps the best book I’ve read all decade, no big deal.

Or if you fancy reading one of my books, I’m rather proud of The Undercover Economist Strikes Back.

Free email updates

(You can unsubscribe at any time)

 

11th of July, 2016MarginaliaComments off
Undercover Economist

We’re all winners or losers now

‘When we stop admiring how much the pie has grown, we start fighting each other for a larger slice’

When it became clear that the British really had voted to leave the EU, markets around the world dropped to the floor and began twitching. One couldn’t help but be reminded of the collapse of Lehman Brothers almost eight years ago, but perhaps the Brexit vote isn’t so much an echo of Lehman Brothers as a sequel in a long story of economic disappointment.

The 2008 financial crisis shaped the rhetoric of the Brexit debate, of course. Had this vote taken place 10 years ago, the fact that most economists thought that leaving would be an act of awful self-harm would have carried weight. The prospect of investment banks moving their activities to Dublin and Frankfurt would have seemed like a disadvantage, not a two-for-one offer.

There is a good argument that, despite the crisis, both economists and banks have something useful to contribute, but that case was never likely to be heard. It was too much to expect a subtle and well-reasoned public debate on the risks and contributions of London as an international banking centre. The most prominent claim of the Leave side — that the UK spends £350m a week on EU membership and could spend that on the National Health Service instead — sailed untouched through the campaigning, despite being simply false.

But some of the seeds of this vote have been growing for much longer in the fertile soil of economic grievance. The UK, like every other major developed economy inside or outside the EU, was growing more quickly per capita before 1973 than afterwards. Slow growth has been the norm across the G7 for four decades. It has been exacerbated in anglophone countries by a sharp increase in top-income inequality in the 1980s and 1990s, which has meant that the benefits of even this modest growth have not been widely felt.

And so we see a desire to upend the status quo: Brexit, of course, but also the rise of radical politicians from Marine Le Pen in France, Donald Trump in the US and Geert Wilders in the Netherlands on the right, to Jeremy Corbyn in the UK, Trump’s opponent Bernie Sanders and Alexis Tsipras in Greece on the left.

But perhaps there are deeper forces too. In 2005, the Harvard University economist Benjamin Friedman published The Moral Consequences of Economic Growth, arguing that times of broad-based economic progress also tended to lead to moral progress, to “greater opportunity, tolerance of diversity, social mobility, commitment to fairness, and dedication to democracy”. Reverse the economic gains and the moral gains may also evaporate.

Friedman argues that we naturally compare ourselves to others. We can compare ourselves to TV celebrities (a recipe for misery, since they are all richer, more famous and better looking than we are) or, as per the platitude, to “those less fortunate than ourselves”.

But, usually, we make one of two simpler comparisons. We can feel happy because we’re doing better than we were last year, or we can feel happy because we’re doing better than our neighbours. In good times, everyone can feel happy for the first reason, but not everyone can feel happy for the second.

When prosperity is broad-based and growing, it’s cheering to look back at how far we have come. We can relax, knowing that we’re earning twice the salary, that we own a larger home, that there are savings and a pension. But in bad times there is little solace to be had by thinking about the past. All we recall is that our younger selves had hair, muscle tone and, above all, a bright future — each of which time has taken away from us.

And so in tough times we resort to the other comparison available to us: are we doing better than our neighbours? When we stop admiring how much the pie has grown, we start fighting each other for a larger slice. Many people who voted to leave did not see EU membership as a joint project for mutual benefit but as a zero-sum game that Britain was losing and Brussels was winning.

Economists are naturally inclined to see the world as a place where everyone can prosper. The Trumpish rhetoric of winners and losers is alien — and alarming — to us. But that is the world in which we now live. The economics of Brexit are daunting but, with goodwill on all sides, they are manageable. It is the zero-sum politics that worry me.

It would be wrong to suggest that economic suffering inevitably produces a backlash. In the UK, the people who have struggled most since the crisis have been the young — and, in a sadly inspiring act, they were the ones who voted overwhelmingly to stay in the EU.

Still, the economic backdrop clearly matters, both in its own right and because of its political effects. Those of us who are committed to openness and prosperity for everyone, regardless of their nationality, now have a long campaign on our hands. We should start by accepting that, if we cannot bring back broad-based and growing prosperity to the advanced economies, Brexit will not be the last political shock we must face.

Written for and first published at ft.com.

Free email updates

(You can unsubscribe at any time)

Undercover Economist

How do you make the Olympics pay? Fudge the figures

‘Hosting the games is not unlike building a church for one single, glorious wedding celebration’

Rio de Janeiro is about to host the Olympic Games. Good luck to it. Brazil is burdened by a political struggle that has seen President Dilma Rousseff impeached, a sharp economic downturn and a public health crisis in the form of the Zika virus. On top of that, Rio has to stage the Olympics.

Don’t get me wrong: I loved the London 2012 Olympics. It was a superb spectacle in its own right and there’s an impressive legacy — some great sporting facilities, a lovely park and new housing in a city that desperately needs it. I just doubt that it was worth what it cost. Very few Olympic Games are.

This shouldn’t really be a surprise: hosting the games is not unlike building a church for one single, glorious wedding celebration. The expensive facilities will only be fully used for a short time. They will then either be underutilised or, at best, cleverly reworked at some expense. It’s possible to adjust and dye a wedding dress so that it can be worn again but this is a pricey way to get a posh frock.

A new survey by economists Robert Baade and Victor Matheson provides a good starting point to understand the problem. Every host of the summer Olympics from Seoul in 1988 to Rio in 2016 has spent many billions of dollars on the games. The cheapest by some margin was Atlanta in 1996, which cost the equivalent of $3.6bn in today’s money; the most expensive was Beijing in 2008, a national vanity project that cost an astonishing $45bn. A reasonable assumption is that today it costs at least $10bn to host a summer Olympics — the last to be cheaper was Sydney in 2000. (These numbers are collated by Baade and Matheson.)

Given likely costs of more than $10bn, an Olympic Games is all but guaranteed to lose money. A host city might expect roughly $4bn in revenue: $1bn from ticket sales, $1bn from sponsors, $1bn from local broadcast rights and $1bn as a share of the International Olympic Committee’s global broadcast deals.

How on earth to make the numbers stack up, then?

The answer is simple: fudge them. The London Olympics and Paralympics provide a shining example. Afterwards, the government reported that the final cost of £8.77bn (around $13bn) was more than £500m “under budget”, which is true after a fashion, since the budget had been revised to £9bn several years before. But the original estimate for the games, back in 2005, was £2.4bn. If you can say that an event that cost almost four times the original estimate was “under budget”, you can say anything you like. At least the London organising committee didn’t emulate its counterpart at Nagano, the host of the 1998 Winter Olympics, and burn its financial records after the event.

Organising committees are always sure to hire consultants to produce enormous estimates of the spillover benefits of hosting the games. These consultant reports generally ignore substitution effects — for example, that a dollar spent on an Olympic ticket might well be a dollar not spent on some other local attraction. They also gloss over the risk of “crowding out”, where, fearing crowds and high prices, some tourists avoid Olympic cities. The UK had fewer visitors during July and August in 2012 than in the same months in 2011; Beijing’s hotels suffered a fall in occupancy during the 2008 Olympics.

Faced with PR fluff, the rule of thumb among serious economists is to divide all such benefits by 10 to get a more realistic figure. Most rigorous studies have found very modest spillover benefits at best.

In the case of the Sydney Olympics, researchers concluded that the wider Australian economy had actually been damaged by hosting the games.

But what if it’s not enough just to claim that hosting the games makes sense? What if the host city actually hopes to turn a profit, rather than merely forecast one — or, at least, to produce broader benefits that justify the expense? That’s a tall order, but here are three suggestions.

First, make sure the games take place during a recession, so that the spending can boost aggregate demand. Rio did get this one right but, of course, nobody can forecast recessions eight months ahead of time, let alone eight years.

Second, be a hidden gem, so the games serve to spotlight your qualities and boost tourism for many years afterwards. Rio, London, Beijing, Athens and Sydney hardly qualify here but Barcelona did: in 1990 the city was half as popular as Madrid with tourists but, by 2010, it had outstripped its rival. The Olympics can perhaps take some credit. Utah, similarly, enjoyed greater success as a destination for skiers after the Salt Lake City winter games. (Candidates for the 2024 games include Paris and Rome, not exactly well-kept secrets — and Budapest, which is a more plausible candidate to earn a lasting boost from tourism.)

Third, and most importantly, launch a cut-price bid in the wake of a disastrous games. Los Angeles in 1984 achieved the near-impossible and turned a profit because, after the ruinously expensive Montreal event of 1976, LA was the sole bidder for the games. They were hosted in the Los Angeles Coliseum, an ageing stadium that had been second-hand even when it first hosted the Olympics in 1932.

The best way to host a profitable Olympics is to do it twice, both times on the cheap.

Written for and first published at ft.com.

Free email updates

(You can unsubscribe at any time)

Undercover Economist

How to fuel a rewarding culture

‘Money matters, but sometimes we find financial incentives to be insulting or grubby’

Here’s an age-old management conundrum: who should be rewarded for high performance, and how? As Diane Coyle, the economist and former adviser to the UK Treasury, recently observed in this newspaper, the answer to the question is usually self-serving. Simple and easily monitored jobs, such as flipping burgers, are natural candidates for performance incentives. Yet somehow it’s the inhabitants of the C-suite who tend to pick up bonuses, despite the fact that their complex, hard-to-measure jobs are poorly suited to the crude nature of performance-related pay.

But let’s assume that managers really want to answer the question. The answer is deliciously complex. Money matters, but sometimes we find financial incentives to be insulting or grubby. And we can respond keenly to non-financial rewards such as praise, status or the satisfaction in a job well done.

So managers might try running experiments to see what works in a particular situation. There is a long tradition of this, going back to Harvard professor Elton Mayo’s productivity trials at Western Electric’s Hawthorne works in the 1920s and early 1930s.

The Hawthorne experiments themselves, alas, were flawed and have been mythologised. But more modern experiments are revealing some intriguing results. I reported a few years ago on the curious alliance between “Farmer Smith”, owner of a large British fruit farm, and three economists, Oriana Bandiera, Iwan Barankay and Imran Rasul. Bandiera and her colleagues designed and tested different incentive schemes on Farmer Smith’s farms. (The deal: he got higher productivity; they got the data.)

The fruit farm experiments show that financial incentives do matter, at least for casual immigrant labour on fruit farms. First, a piece-rate scheme boosted productivity by 50 per cent; then, performance pay for the front-line managers ensured that work was no longer assigned as a favour to friends, and productivity increased another 20 per cent; then, a tournament encouraged workers to sort themselves into productive teams, and productivity increased by a further 20 per cent.

In another study by Bandiera (with Nava Ashraf and Kelsey Jack), hair stylists in Zambia’s capital Lusaka were recruited to sell condoms and give advice on HIV prevention. It turned out that celebrating the top performers at a public ceremony proved a far better approach than providing financial incentives to sell more condoms.

But sometimes neither a public ceremony nor a financial incentive is appropriate. Consider the case of long-haul airline captains. Unlike part-time condom agents or fruit pickers, these senior pilots have high-status, six-figure salaries and powerful unions to defend their pay and conditions. Nevertheless, a recent experiment conducted by Greer Gosnell, John List and Robert Metcalfe examines what can be done to influence the behaviour of these star players.

Gosnell, List and Metcalfe teamed up with a commercial airline that wanted to encourage captains to save fuel. Broadly, there are three ways to do this: before take-off, by carefully calculating fuel requirements; after landing, by switching off some engines while taxiing; and during the flight, by carefully adjusting the flap settings and negotiating the most efficient altitude, speed and course with air traffic control. The airline’s own data suggested that captains could potentially save 3 to 6 per cent on fuel — a substantial financial and environmental gain. But how to incentivise them?

Gosnell, List and Metcalfe designed an experiment that did not rely on paying bonuses. Instead, the captains were told that their company was running an experiment with the aim of saving fuel, and that the researchers would maintain anonymity for all the captains. There would be no financial incentives and no league tables.

Instead, the captains were split randomly into four groups. The “information” group received monthly feedback reports detailing how often they had saved fuel before, during and after each flight. The “target” group received the same reports but were also set targets to improve their performance. (The reward for hitting the target was a hearty “well done!”) The “incentives” group were told that for each target they hit, £10 would be donated to the charity of their choice — a total donation of £240 was possible if all three targets were hit across the eight months of the study. A control group was simply told that a study into fuel efficiency was taking place.

The most obvious outcome was that there was a large and lasting “observer effect”. Merely telling captains that the experiment was happening prodded them into being more careful and saving a lot of fuel. It is always possible that the sudden switch to fuel-saving behaviour had a cause that was nothing to do with the experiment but there are no apparent alternative explanations.

The second outcome was that all three treatments saved fuel compared with the control group but setting targets (with or without the charitable donation) had a particularly notable effect. And the third outcome was that captains who hit their targets were substantially more satisfied with their jobs.

“I just couldn’t believe the impact we had on job satisfaction,” says Metcalfe. Far from annoying the captains, the fact that the company was taking an interest in fuel saving, and acknowledging success, seemed to delight them.

No performance scheme will fit every occasion but the fuel-saving study does suggest an approach worth trying more broadly. If you want people to do a good job, tell them what success looks like to you — and that you’ve noticed when they’ve achieved it.

Written for and first published at ft.com.

Free email updates

(You can unsubscribe at any time)

Undercover Economist

The dubious power of power poses

‘Many notable results in psychology are now being questioned because later research has reached different conclusions’

Imagine that a group of researchers set out to explore the idea that adopting a “power pose” could make a real difference to how we thought and acted. High-power poses include standing with hands on hips, feet planted confidently apart, or lounging back in a chair with feet on table and hands behind head; low-power poses include slumped shoulders and folded arms.

The researchers asked 200 people to adopt such poses, then tested the levels of two hormones in their saliva: testosterone, associated with high status, and cortisol, associated with stress.

The astonishing findings? Well, actually, there were no astonishing findings: the power poses seemed to make no difference worth mentioning. High-power poses were correlated with slightly lower testosterone and slightly higher cortisol — the opposite of what might be expected, but tiny and statistically indistinguishable from chance.

Now imagine that a second group of researchers re-examined the same hypothesis. There were some small variations, and the study was smaller (42 participants). The new study did produce remarkable findings: high-power poses boosted testosterone and lowered cortisol. Low-power poses had the opposite effect. The scale of the effect was described as “a whopping significant difference” by one of the researchers — more formally, the sizes were both practically large and statistically significant. (Statistical significance is a test of whether the result might easily have been a fluke; it’s possible to have small but statistically significant results, or large but statistically insignificant results.)

Faced with both these research findings, published in reputable journals, what should we think? The natural conclusion is that the second study was a fluke, and that standing in a bold pose for a couple of minutes makes no difference to hormone levels. Being open-minded people, we might also be intrigued by the faint possibility that the second study had uncovered a genuine and important result.

This is a hypothetical scenario, I should emphasise. It hasn’t happened. The studies did take place but not in this order. The smaller study was conducted by Amy Cuddy of Harvard and Dana Carney and Andy Yap of Columbia. It inspired a book, and a TED talk that has been watched 34 million times. The larger study was conducted by a team led by Eva Ranehill. But the smaller Cuddy-Carney Yap study didn’t come second; it was conducted first. The Ranehill team’s study came later.

This story will sound familiar to some. Many notable results in psychology are now being questioned because later research has reached different conclusions. Last year, the “Reproducibility Project”, a large collaborative effort reproducing 100 studies in psychology, published the unnerving finding that only 36 per cent of the replication attempts had produced statistically significant results.

But it is not easy to know quite what to make of that percentage. Failing to find a statistically significant effect in a replication does not simply discredit the original work. For example, some replications find similar effects to the original studies without achieving statistical significance. That means the replication provides (faint) support for the original study rather than evidence against it.

Wharton psychologist Uri Simonsohn suggests a replication attempt should use a substantially larger sample than the original, so it is likely to estimate effects more precisely. If the replication fails to find an effect, that’s not proof there’s no effect; it does suggest, however, that the original study was a fluke.

Columbia University statistician Andrew Gelman suggests a simple rule of thumb that I followed in the opening paragraphs of this column: mentally reverse the order of the studies. Imagine the “replication” came first, and the “original” study came later. Being published first should not be a privileged position from which our conclusions can only be budged with extraordinary evidence. Gelman’s rule of thumb helps us avoid doggedly sticking to the status quo.

But perhaps the most important lesson is to remember that while “statistical significance” sounds scientific, it’s hardly a cast-iron endorsement of a result. The theory behind statistical significance assumes that a single pre-chosen hypothesis will be tested. In practice, researchers rarely pre-specify their hypothesis. They can test dozens, or hundreds — and sooner or later a pattern will emerge, if only by chance.

Imagine testing the idea that vitamin supplements boost childhood achievement. OK. But only for girls? Only for boys? Only for children suffering a poor diet? Only for under-10s?

An unscrupulous researcher can grind through the myriad combinations until a statistically significant pattern appears. But, says Gelman, there is no reason to think such unethical behaviour is common. More likely, researchers gather the data, look informally at the patterns they see, and only then choose a few hypotheses to test. They will tell themselves — correctly — that they’re being led by the data. That’s fine. But nobody should take seriously a test of statistical significance that emerges from such a research process: it will bring up fluke after fluke.

There are various technical solutions to this problem. But a little common sense also goes a long way. When a study of 42 subjects inspires 34 million people, it’s not unreasonable to go back and check the results.

Written for and first published at ft.com.

Free email updates

(You can unsubscribe at any time)

Undercover Economist

Worth the wait?

‘If you miss your plane or your train, it hardly matters that the queue was a nice place to chill’

I love queues. Not that I love queueing — I may be English but I’m not that English. But from a safe distance, queues are fascinating. They’re less fun if they cause you to miss your flight. In mid-May, two-hour queues for security at Chicago’s Midway airport had just that effect. Jeh Johnson, the US Secretary for Homeland Security, offered travellers some meditative advice: “Contemplate increased wait times as you travel.” I’d hope we can do a little better than mindful meditation.

There are three very different perspectives on queues: psychological, engineering and economic.

The psychological perspective tells us that much of what makes queues unpleasant is nothing to do with the waiting time. If a queue carries risk (you may or may not make your flight), then it is far more stressful. So are queues that are confrontational, unfair or require constant monitoring for queue-jumpers or the sudden opening up of new lines.

A single serpentine queue, secure against cheats, can be a perfectly civilised place to stand and check email or read a paperback. With a bit of cleverness, the queue may be a pleasure — as at well-designed theme parks — or an unobtrusive virtual version, as when you collect a ticket from the supermarket deli counter and do some shopping while waiting for your number.

There are, however, limits to the psychological approach. When the Eyjafjallajökull eruption shut down air travel across Europe in 2010, I found myself queueing for train tickets in Stockholm Central Station, along with almost everyone else in Sweden. Thankfully, the queue had a counter system: simply take a ticket, and wait for your turn. I sat in a café, sipping espresso and typing on a laptop as I waited. But, after a pleasant three-quarters of an hour, I did some mental arithmetic, and realised that the queue was approximately 14 hours long. In the end, if you miss your plane or your train, it hardly matters that the queue itself was a nice place to chill.

When psychology fails, engineering must take the strain. A well-engineered queue copes gracefully with periods of high demand, and balances the cost of waiting against the expense of overproviding idle service staff.

Queue engineers understand that queues can have strange properties. Imagine the queue at a busy post office. During the mid-morning lull, roughly one person a minute arrives and one person a minute can be served. The queue will fluctuate — and, alas, there will never be a negative number of people in the queue — but we can expect it to stay fairly short. Then, during lunch hour, extra people arrive and the queue starts to lengthen — two people, then four, five, 10. As the rush subsides, the capacity of the ticket office again begins to match the inflow of customers: one person arrives each minute, and one person is served each minute.

Annoyingly, even though the inflow and outflow of people from the queue is the same as it was in the morning, the afternoon queue is about 10 people long. It will stay 10 people long until the capacity of the ticket office is greater than the inflow of customers. Once a serious queue has formed, it needs attention or it can linger indefinitely.

That brings us to the economic perspective on queues. Queues are a terrible, inefficient waste of time. If the resource in question is genuinely limited, then the existence of a queue shows that it is being underpriced. If everyone had to pay to join a queue, the queue itself would be shorter, because some people would decide not to bother. Those who did queue would earn back their entry fee in time saved, while the person selling tickets for the queue would make some cash.

In other cases, however, capacity should expand to keep the queue short. Imagine a line so long that most passengers would pay $50 to skip it — probably a good description of the two-hour queues at Midway. Hiring extra Homeland Security staff would save $50 worth of frustration for every extra person they scan from the line.

. . .

How many people could an extra security team see? One per minute, perhaps? Fifty dollars a minute would surely pay for some extra personnel. The problem is that the security team is unlikely actually to receive the $50. In an alternate universe, passengers would have a whip round, hire more agents, and the line would move just fine.

But in the world in which we live, queues remain. Part of the cost is imposed on foreigners, whose annoyance barely registers on the system. (This is particularly true of immigration checks.) For example, on a recent trip from South America to London, I chose to change at Madrid rather than at Miami because I’ve had terrible experiences at Miami. That’s bad for the US economy but security screeners, customs officers and immigration officials respond to political signals, not market ones. The US political system is hardly likely to dance to my tune.

Looking on the bright side, I hear that Reagan National Airport, often used by members of Congress as they fly in and out of Washington DC, works like a charm.

Written for and first published at ft.com.

Free email updates

(You can unsubscribe at any time)

Undercover Economist

How the sense of an ending shapes memory

‘Composers, novelists and film directors try to end on a high. Restaurants keen to manipulate their online reviews have found a similar trick’

Many years ago, I listened to a string quartet perform a challenging piece of contemporary music. The piece, we were told, represented a journey of suffering and redemption. It would descend into discordant screeching for nearly 20 minutes before finally resolving harmoniously. The small concert hall was packed — there were even people seated on stage behind the performers — so there was little choice but to stick it out.

Everything unfolded as promised. The performance sounded like a succession of cats being tossed into a food processor. Eventually, though, the dissonance became resonance, the chaos became calm. It was beautiful.

But then came a sound that had not been in the score; the electronic peal of a mobile phone rang out across the tranquil auditorium. To make matters worse, the beeping arpeggios were emerging from the pocket of an audience member who was sitting on the stage. He was so close to the performers that he could easily have been downed by a solid backhand swing with the viola. It must have been tempting.

The music had been ruined. But it’s curious that 20 minutes of listening can be redeemed or destroyed by what happened in a few moments at the conclusion.

Daniel Kahneman, psychologist and Nobel laureate, tells a similar story about a man enraptured by a symphony recording that is ruined by a hideous screech — a scratch on the vinyl — in the final moments.

“But the experience was not actually ruined,” writes Kahneman, “only the memory of it.” After all, both concerts were almost complete when interrupted. The lived experience had been unblemished until the final moments. The remembered experience was awful.

When we recall things — a concert, a holiday, a bout of flu — we do not play out the recollection minute by minute like a movie in our minds. Instead, we tell ourselves a little story about what happened. And these stories have their own logic in which the order of events makes a difference.

Consider Jenson Button’s 2009 season in Formula One. The British racing driver easily outpaced his rivals in the first seven races of the season, building a vast lead. Then, as the relative performance of the cars changed, Button failed to win any of the remaining 10 races. His rival Sebastian Vettel couldn’t quite catch him, though, and Button became champion with a limp fifth place finish in the penultimate race. One pundit defended Button against his many doubters with the feeble line: “There have been many less gifted world champions than Jenson Button.”

But imagine if the order of results had been reversed. After being beaten in almost every one of the first 10 races by Vettel, Button would have mounted a magnificent comeback, sealing his world championship with a victory in the final race. The same results in a different order would have told a very different story. And the story matters.

Kahneman and his colleagues have run a number of experiments testing these ideas. In one, people were asked to hold one hand in painfully cold water for 60 seconds. Some subjects then had to keep their hand in the water for another 30 seconds while a hidden valve released fractionally warmer water. So, which experience was worse: 60 seconds of pain, or 60 seconds of pain followed by 30 seconds of somewhat lesser pain? The experimental subjects preferred the longer experience with the happier ending.

In another study with Don Redelmeier, Kahneman surveyed colonoscopy patients every 60 seconds while they underwent a distinctly uncomfortable procedure, producing a minute-by-minute record of just how painful the colonoscopy was. Then, Redelmeier and Kahneman asked the patients “the total amount of pain” they had experienced. The responses were strongly correlated with the average of two factors: the pain experienced at the worst moment, and the pain experienced at the end.

This is summarised as the “peak-end” rule. Our memories of experiences are governed by — of course — the most memorable things about them. Had the doctor left the probe inside the patient, without prodding around, for an extra 10 minutes, the final moments wouldn’t have felt too bad and the entire memory of the procedure would have been less grim.

No wonder Jenson Button’s 2009 season seemed mediocre: his peak performances were great but his final performances less so. And no wonder that disruptive mobile phone was so aggravating: since the best moment of the music came at the end, one ringtone managed to spoil both the peak and the end.

Of course, it is no coincidence that the best bit of the music was at the finale: composers, like novelists and film directors, try to end on a high.

Restaurants keen to manipulate their online reviews have discovered a similar trick: twice recently I’ve dined at restaurants in unfamiliar towns that were highly rated on TripAdvisor. Both times, the food was good but unremarkable. Both times, the proprietor pressed gifts upon us as we left — a free glass of grappa, a nice corkscrew. It seems that when people thought back and wrote their reviews, they remembered this pleasant send-off. That makes sense: if you want people to remember you fondly, it’s best to engineer things so that the last thing they remember of you is something other than signing a bill.

Written for and first published at ft.com.

Previous

Elsewhere

  • 1 Twitter
  • 2 Flickr
  • 3 RSS
  • 4 YouTube
  • 5 Podcasts
  • 6 Facebook

Books

  • Messy
  • The Undercover Economist Strikes Back
  • Adapt
  • Dear Undercover Economist
  • The Logic of Life
  • The Undercover Economist

About Tim

Tim Harford is an author, columnist for the Financial Times and presenter of Radio 4's "More or Less".
Learn More ›

Subscribe to TimHarford.com

Enter your email address to receive notifications of new articles by email.

Tim’s Tweets

Do NOT follow this link or you will be banned from the site!