Tim Harford The Undercover Economist

Articles published in September, 2016

Why central bankers shouldn’t have skin in the game

Pensions campaigner Ros Altmann recently launched an eye-catching attack on the Bank of England for paying generous pensions to its own staff while undermining everyone else’s retirement plan.

 It’s an easy claim to make, and it may well stick. Are central bankers not the Marie Antoinettes of the modern world, recklessly imposing penury on the people while ensuring they are protected from their own ruinous actions?

I don’t see it myself. To start with the most obvious point, it’s far from clear that the Bank really is destroying pensions. It is true that low interest rates make future obligations loom larger in today’s company accounts. This creates a problem for any pension scheme. But, on the other side of the equation, low interest rates have boosted the value of shares, bonds and property and thus the value of most pension schemes. Would pension schemes really be better off had the Bank of England stiffened interest rates in 2008 and tried to engineer a rerun of the Great Depression?

This takes us to a very strange place indeed. Currently, members of the Monetary Policy Committee (MPC) set interest rates without having a strong personal incentive to follow any particular policy. I would have thought that was a desirable state of affairs, but perhaps not. An alternative — presumably the alternative that critics would prefer — is that whenever the MPC cuts interest rates, its members know that they will feel the financial consequences personally.

But we can’t stop by linking senior Bank of England pensions to interest rates. If we are to pay the MPC by results, we must do so in a way that reflects the broader consequences of their actions. Consider unemployment. The MPC is not officially responsible for supporting the job market, but the US Federal Reserve is. And everyone knows the MPC considers the state of the wider economy when setting rates. As the Bank’s chief economist Andy Haldane recently commented: “I sympathise with savers but jobs must come first.”

That is what he says now, but what would Haldane do under the new incentive scheme? He and his colleagues may ignore the labour market unless they have some skin in the game. Perhaps we should draw inspiration from the Roman practice of decimation, where some soldiers in a mutinous cohort would be executed according to the drawing of lots. Execution is harsh, but one could easily make the 100 most senior Bank staff participate in an unemployment lottery. If the unemployment rate is 5 per cent, then five of them — chosen at random — must be punished. If the unemployment rate is 10 per cent, then 10 senior Bank staff will taste the consequences. An appropriate punishment? They shall be condemned to carry out their duties, unpaid, from a queue in the job centre.

Once one starts to spell out exactly how Bank of England staff should be rewarded for each policy triumph or penalised for each misstep, it becomes clear that the whole idea is nonsense. Central bankers will not do a better job if given direct financial incentives to pursue certain policies, and it is quite likely that they will do a worse job. Yes, incentives matter — but they often matter for all the wrong reasons.

The Wells Fargo scandal is a recent example: the bank put pressure on its employees to cross-sell financial products to customers. In response to the pressure, some staff simply opened accounts or set up credit cards for customers without their knowledge. The whole thing is depressingly unsurprising.

The basic principle for any incentive scheme is this: can you measure everything that matters? If you can’t, then high-powered financial incentives will simply produce short-sightedness, narrow-mindedness or outright fraud. If a job is complex, multifaceted and involves subtle trade-offs, the best approach is to hire good people, pay them the going rate and tell them to do the job to the best of their ability.

It would be nice to think that independent central banks could get on with a difficult job without being dragged into politics — but of course that is impossible. Ros Altmann isn’t the first person to try to take a debate about central bank policy into the personal realm. Donald Trump recently announced that Federal Reserve chair Janet Yellen should be ashamed of herself; in the previous US presidential campaign, Governor Rick Perry accused her predecessor Ben Bernanke of treason. Senior Brexit campaigners made similar attacks on Mark Carney.

Central bankers no doubt find such personal attacks vexatious — but they should take comfort in them. When Paul Volcker ran the US Federal Reserve, his policies so enraged building contractors that they mailed pieces of two-by-four to his office; farmers blockaded the Federal Reserve with their tractors. Yet Volcker is now the most respected Fed chairman in history. Effective central bankers inevitably annoy a lot of people; that is why the job is too important to be entrusted to politicians.

Written for and first published in the Financial Times.

It helps any new book to pick up some advance orders, so if you like my writing please consider pre-ordering my new book, “Messy“. (US) (UK) More to follow soon…

Free email updates

(You can unsubscribe at any time)


How insurers keep the money-pump flowing

One of the strangest financial scandals of recent years has been the selling of payment protection insurance. PPI is a curious form of insurance that meets loan repayment obligations for people who become ill or jobless.

 Now, there is an honourable tradition of insurance being bundled with a loan. The practice is discussed in the Code of Hammurabi, which is nearly 4,000 years old. This ancient Babylonian law code includes 282 clauses on “bottomry”, which functioned as payment protection insurance for maritime merchants. Hammurabi specifies that a merchant who borrows money to fund a ship’s voyage is not obliged to repay the loan if the ship sinks.

That was a sensible enough arrangement but there are two objections to PPI in its modern form. One is that such policies were often sold to people under false pretences, including those who would never have been eligible for repayment. For this reason, British banks have had to repay tens of billions of pounds in compensation.

To see why, let’s step back and ask ourselves what insurance is for. Classical economics has an answer: people are risk-averse, which means that they will pay good money to reduce the variability of outcomes they face. If home insurance guards against the loss of a million pounds when my house burns down, I’m happy to buy the insurance even though the insurance company expects to make a profit from it.

But this risk aversion emerges from the fact that money is worth more to poor people than to rich people. Gaining a million pounds would make me rich but losing a million pounds would make me poor. I should not gamble a million pounds on the toss of a coin, because the million pounds I might lose is more precious to me than the million pounds I might gain.

As so often with classical economics, this is an excellent description of how we should behave. It is not such an excellent description of how we actually do behave. Risk aversion can only explain why we insure large risks. It cannot explain why we insure small ones. This is because risk aversion turns on the idea that an extra pound is worth more if you are poor than if you are rich. But having to replace a phone is not going to make the difference between poverty and wealth.

In one of my favourite economics articles, written in 2001, the behavioural economists Richard Thaler and Matthew Rabin point out that anyone who rejects a 50/50 gamble to win £10.10 or lose £10 — apparently a reasonable enough taste for caution — cannot possibly be doing so because of risk aversion. (The degree of risk aversion necessary would mean that the same individual wouldn’t risk £1,000 on the toss of a coin for all the money in the world.) Risk aversion simply cannot explain why anyone would turn down that fractionally favourable gamble. And it cannot explain why anyone would insure a mobile phone.

A better explanation is that we tend to view risks in isolation. Rather than telling ourselves “a lost mobile phone would lower my lifetime wealth by 0.005 per cent”, we tell ourselves “it would be so annoying to have to pay for a new mobile phone”. Isolating and obsessing about risks in this way is arbitrary and illogical. But that does not mean we don’t do it.

At this stage, I would like to introduce you to the idea of a money pump. A money pump is a person whose irrationalities can be systematically exploited for financial gain. The simplest money pump is a person who prefers an apple to a doughnut, prefers a doughnut to a chocolate bar, and prefers a chocolate bar to an apple. Just offer them an apple in exchange for their doughnut plus a penny. They will accept. Then offer them a chocolate bar for their apple plus a penny. Then offer them a doughnut for their chocolate bar plus a penny. They end up with their original doughnut and are three pence poorer. Repeat for ever.

Money-pump arguments are sometimes deployed to object that people cannot be irrational, otherwise they would be bankrupted by money pumping. But economists are increasingly coming to realise that, instead, we should be looking for money pumping in action.

Given our anxiety about small risks, what would the money pumping look like? It would be an insurance policy focused on the narrowest possible slice of risk. It would be sold alongside another product or service, often at the last moment. It would be marketed by creating anxiety and then offering the product to make the anxiety go away. In short, it would look like the collision damage waiver, the extended warranty, and PPI. These bespoke slices of insurance are among the largest money-pumping projects in the modern economy. No wonder the banks abandoned their principles to join in.

Written for and first published in the Financial Times.

It helps any new book to pick up some advance orders, so if you like my writing please consider pre-ordering my new book, “Messy“. (US) (UK) More to follow soon…

Free email updates

(You can unsubscribe at any time)

Autumnal reading

Some exciting books coming along soon. I urge you to buy David Bodanis’s new book Einstein’s Greatest Mistake. (US) (UK) I read it in an earlier draft and it’s a wonderful, fresh and readable take on one of the most fascinating lives in science.

Later in the autumn, Steven Johnson launches Wonderland (US) (UK) – a book about how play shaped the modern world. Bravo for the concept, Johnson’s previous books are smart and thought-provoking and the associated podcast series has been great so far.

Alex Bellos offers us Can You Solve My Problems? – self-recommending for Christmas. Alex’s puzzles in the Guardian have been delightful and he’s a terrific writer. (US) (UK)

Hannah Fry and Thomas Evans, also self-recommending for Christmas, give us the mathematics of the season with The Indisputable Existence of Santa Claus (US) (UK)

If you can’t wait for any of them, try Adam Grant’s Originals (US) (UK) – the TED talk gives a good sense of the book, which is excellent.

Finally, I should perhaps mention that my own book, Messy (US) (UK), is coming soon. More details here.

Advanced orders are helpful for all authors to alert booksellers to demand, so show these folks some love!

Free email updates

(You can unsubscribe at any time)

15th of September, 2016MarginaliaComments off

The hazards of a world where mediocrity rules

“It’s absolutely amazing, but under the right circumstances, a producer could make more money with a flop than he could with a hit.” Thus spoke accountant Leo Bloom, played in The Producers (1968) by the much-mourned Gene Wilder. In Bloom’s thought experiment, a dishonest producer would raise a vast sum by selling the profits of a Broadway show many times over. Provided the Broadway show was a flop, nobody would come looking for their share of the profits and the fraudsters could retire to Rio. If the show was a hit, of course, “well, then you’d go to jail”. That was where Bloom and his partner Max Bialystock ended up: their musical, Springtime for Hitler, was far too good.

In the real world, people tend not to become richer when they do a worse job. There are exceptions, of course. In 2013, a jury found that Fabrice Tourre, formerly a trader at Goldman Sachs, had misled investors about the nature of “Abacus”, a complex financial security — and done so because that was his job. Abacus was, like Springtime for Hitler, a bet on collapse mis-sold to investors who did not seem to fully understand it.

Both cases are extreme examples of “moral hazard” — the odd phrase that economists have taken up to describe perverse incentives that encourage people to be careless, reckless or even outright saboteurs. Moral hazard traditionally applies in insurance cases, and indeed recent reports from Vietnam describe a woman who cut off her hand and foot in an attempt to collect a six-figure payout from her insurance company. There was a rash of such self-harming frauds in the Florida panhandle 50 years ago.

Economics has no difficulty analysing such cases — several Nobel Memorial Prizes have been given to economists who studied moral hazard. Still, they run counter to the mood music of mainstream economics, which tends to strike Panglossian chords. The starting point of modern economics is the perfectly competitive equilibrium, in which resources are allocated efficiently and the market will deliver more of what people really want. Against such a stirring symphony, Leo Bloom and Fabrice Tourre hit isolated, dissonant notes.

Yet there are corners of the economy where poor work is the norm, not the exception. A few years ago, two Italian academics, sociologist Diego Gambetta and philosopher Gloria Origgi, published an article reflecting on what they called “the LL game”. It has since found a catchier term: Kakonomics — the economics of rottenness.

In a kakonomy, mediocrity rules. People not only supply shoddy work and expect shoddy work in return, they actually prefer to receive shoddy work. I’m put in mind of the shared student house in which nobody can quite be bothered to wash the dishes, empty the bins or even buy new toilet paper. The presence of a housemate who bustles around wiping up the filth might seem to be welcome but, in fact, it’s an aggravation because it puts pressure on everyone else to join in.

Gambetta and Origgi observed the LL game being played at an advanced level in Italian universities. Not only would both parties to an agreement deliver low-quality (hence the “LL”), but they would insist to each other that they were doing an excellent job, and pronounce themselves delighted with what they had received in return. For example, a visiting lecturer might agree to deliver a series of eight original seminars and be paid an honorarium of €1,200 in advance. In fact, the payment is six months late and it’s only €750 (some excuse about taxes); meanwhile, the lecturer is mostly on holiday with his family and only gives five lectures, all of which are old hat. Both sides expected as much yet both sides loudly announce they’re delighted with the superb professionalism on show. Meanwhile, they are indeed pleased enough: the faculty has not been embarrassed by some visiting star and retains a larger entertainment budget; the lecturer enjoyed a free holiday without having to do any serious work.

There is something rather charming about a kakonomy at first glance. It can be quite pleasant to relax and be a little bit crappy for a while, and we all know that there is nothing quite so exhausting as a colleague — or, worse, a spouse — who is relentlessly perfect.

But a true kakonomy is collusive, a tacit agreement to be mediocre at someone else’s expense. In the case of many Italian universities, it appears that collusive mediocrity costs Italian students and the Italian taxpayer. (Lacking personal experience, I take Gambetta and Origgi at their word about the quality of most Italian universities.) Once a kakocracy has been established, it is likely to endure: recruiters will be careful not to hire anyone who might not only rock the boat but also repair the leaks and fix the outboard motor.

The spectre of kakonomics is a reminder of the importance of things that cannot be measured: the culture of an investment bank, or a university, may matter just as much as the explicit rules. Even when Bialystock and Bloom went to jail, they moved on to the next scam without missing a beat.

Written for and first published in the Financial Times.

It helps any new book to pick up some advance orders, so if you like my writing please consider pre-ordering my new book, “Messy“. (US) (UK) More to follow soon…

Free email updates

(You can unsubscribe at any time)


Can trivia help us to be less ignorant of our own ignorance?

In the early hours of April 20 1995, police knocked on the door of McArthur Wheeler and arrested him for holding-up two Pittsburgh banks the previous day. Wheeler could hardly have been surprised that the police were on to him: wearing no mask or disguise, he had ambled into the banks during business hours and brandished a gun in full view of security cameras. Nevertheless, he was astonished, protesting “but I wore the juice!” Wheeler had formed the erroneous belief that lemon juice rendered people invisible on video.

Wheeler is now a legend in psychology, since it was his regrettable escapade that inspired two psychologists, David Dunning and Justin Kruger, to figure out whether we have a good sense of our own strengths and weaknesses. Dunning and Kruger set tests of grammar, logic and even having a sense of humour to a group of undergraduates. Then they asked them how they stacked up to others in the group. Was their grasp of logic and grammar better or worse than average? Were they better able than other students to distinguish funny from unfunny jokes?

Most students thought that they were above-average logicians, grammarians and wits but the Dunning-Kruger effect is not mere overconfidence. The competent people in the study had a reasonable grasp of where they stood in the pecking order. The incompetent ones were blissfully unaware of their incompetence. The good students knew that they were good; the bad students had no clue that they were bad.

Perhaps because Dunning and Kruger opened their 1999 research paper with the story of McArthur Wheeler, the Dunning-Kruger effect has now become a popular insult in some corners of the internet. We chuckle at people who are far too stupid to know that they are stupid. Unfortunately, such mockery misses the subtlety and universality of the effect. All of us are incompetent in some areas. When we stray into them, the Dunning-Kruger effect may be lurking.

The fundamental problem is a person trying to diagnose his own incompetence is — almost by necessity — likely to be missing the skills needed to make that diagnosis. Not knowing much grammar means you’re poorly placed to diagnose your ignorance of grammar.

There is, of course, a cure for the curse of Dunning-Kruger: asking for advice or criticism. On the question of whether lemon juice is an invisibility potion as well as an invisible ink, McArthur Wheeler could have benefited from a second opinion. Doing so, alas, would have required him to doubt his own reasoning on the matter; it would also have required him to identify a bright-enough advisor. And all of us — especially high-status people — face the problem that when we are sorely mistaken, our friends and colleagues are often too polite to tell us. Still: two heads are better than one.

In a new book, Head in the Cloud (US) (UK), William Poundstone argues for a fresh defence against Dunning-Kruger catastrophes: trivia. Poundstone believes that a broad base of knowledge helps to clue us in to the times when we are stumbling towards a humbling; if we know a little about a lot, we have more opportunities to catch ourselves in the middle of a Dunning-Kruger moment.

Poundstone’s own research suggests that there’s a correlation between income and general knowledge, over and above what might be expected from education levels. One of many plausible explanations: people with a good grasp of general trivia are people who are paying attention to the world.

This is highly speculative stuff, but thought-provoking. Poundstone is pushing against the tide: educational fashion, as well as common sense, suggests that in the age of the smartphone it is better to focus on critical thinking than on rote memorisation. But he may have a point; any particular fact can be looked up, but without a knowledge base who knows where to start?

Recently, psychologist Sarah Tauber and four fellow researchers posed a long series of trivia questions to hundreds of young people (their average age was 20). An example: “What is the name of the large hairy spider that lives near bananas?” Despite a generous marking system (for example, “teranchula” was considered correct), performance was unimpressive. Fewer than half the subjects knew which country’s capital was Baghdad, or what the spear-like object was that was thrown around in athletics contests. Only 43 per cent knew that the hairy spider was a tarantula, however spelt.

As for more challenging pieces of trivia, the performance was astonishingly bad. Name Flash Gordon’s girlfriend, or the author of The Brothers Karamazov, or the first man to run a four-minute mile, or the mountain range separating Europe from Asia? Out of hundreds of participants, nobody knew. Nobody. There were 50 such questions, questions to which not a single person could venture a suitable guess. And these 20-year-olds were undergraduates, so presumably reasonably smart and ostensibly well educated.

It’s not that young people today are stupid. They’re the most educated generation in history, and their intelligence is higher, at least as measured by IQ tests. It’s just that there’s a lot they don’t know, and (as per Dunning-Kruger) a lot they don’t know they don’t know. I’m not sure that is a problem but it might be. As Poundstone points out, one thing you cannot Google is what you should be googling.

Written for and first published in the Financial Times.

It helps any new book to pick up some advance orders, so if you like my writing please consider pre-ordering my new book, “Messy” .(US) (UK) More to follow soon…

Free email updates

(You can unsubscribe at any time)



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


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

Tim’s Tweets

Search by Keyword

Free Email Updates

Enter your email address to receive notifications of new articles by email (you can unsubscribe at any time).

Join 3,473 other subscribers

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