Tim Harford The Undercover Economist

Articles published in April, 2010

Business Life: Nothing lasts forever

First published in Business Life, September 2009

The great economist, Alfred Marshall, came to believe that industrial titans were all but immortal: they “often stagnate, but do not readily die.” That seems sensible. Some companies do seem to last forever. The most valuable companies of 1912, just two years after Marshall wrote those words, included Procter and Gamble, General Electric, and Shell. All three companies not only survived but prospered. Procter and Gamble was the 52nd most valuable company in the world. General Electric was the seventh most valuable, and Shell the fifth. By 1995, according to data collected by the economic historian Leslie Hannah, all three companies had risen up the ranks, and were the tenth, second and most valuable companies in the world.
That might not seem surprising: it was what Marshall expected. But looking closer, was he really correct?
The world’s most valuable company in 1912 was US Steel, a gigantic corporation even by today’s standards, employing 221,000 workers. This was a company with everything going for it: it was the market leader in the world’s leading economy, in an industry that never became obsolete. US Steel even had luck on its side, dodging a bullet when Standard Oil and Du Pont were broken up by the US government in 1911. And yet over time US Steel disappeared completely from list of the most valuable 100 global companies; last year it was not even in the top 500.
Third on the list in 1912 was the British company J&P Coats, which was acquired by Viyella in 1986 and which has dropped out of the top 100. Next was Pullman, bought out by Wheelabrator-Frye in 1980. Other top ten companies included Anaconda, Singer, American Tobacco (now American Brands), and International Harvester. None made the top 100 in 1995.
While Pullman suffered from being in a declining industry, others seemed to limp into obscurity for no obvious reason. If General Electric was such a star, why did its old rival Westinghouse Electric not keep up? If Procter and Gamble flourished in fast-moving consumer goods, what happened to Cudahy Packing?
Ten of Hannah’s top 100 vanished inside a decade. Over half disappeared over the next 83 years, and just 19 retained their position in the top 100. In fast-growing new industries, too, from the dot-coms to the early auto industry and even the 15th century printing industry, most firms disappear.
The paradox of the marketplace is that it delivers prosperity despite endemic failure – in fact, I would say, because of endemic failure. Many of the titans of 1912 failed because they were replaced by something better.
That is one reason why the financial crisis has posed such a challenge to the idea that markets work well. We are now told that banks are “too big to fail”. That may well be true, but if so, it is a big departure from the lessons of the past.

29th of April, 2010Other WritingComments off

Should I forgo my holiday to vote Tory?

Dear Economist,
Would it be very wrong to go on holiday and miss voting in the forthcoming general election? We have an excellent constituency MP (Tory) who has a reasonably safe seat. I don’t agree with all his party’s policies, such as they are, but he has an excellent voting record and works very hard. I’d like to see him retain his seat – but not as much as I’d like to go on holiday.

Last election, the Liberal Democrats came second in my constituency and Labour came third, with a turnout of 70 per cent. If my vote were material in preventing a Labour government, I would be prepared to forego the holiday. Should I?

Unmarried UK Voter

Dear Unmarried Voter,

The country is fortunate indeed to have voters like you. Many people choose not to vote because of far smaller inconveniences – that it is raining, or that every local candidate is an uninspiring stuffed suit. (Next time, though, might I suggest registering to vote by post?)

I think you can cut yourself some slack. In the vanishingly unlikely event that your vote does make the difference between Tory and Labour government, the margin of victory will be so fine that the joy of Tory voters will be precisely balanced by the misery of Labour voters. For this reason I believe you need not vote at such great inconvenience to yourself. Let us consider your own preferences instead. The Conservatives are far more popular than at the last election, while Labour’s popularity has slumped. Unless there is some local scandal you have not mentioned, there is almost no chance that your MP will lose a safe seat while swimming with a favourable tide. The chance your vote will matter locally must be much less than 1 per cent. If you value that more strongly than a vacation, you’re a committed voter indeed – unnervingly committed. Sounds to me like you need that holiday!

Also published at ft.com.

24th of April, 2010Dear EconomistComments off

Hard cash underpins the spirit of independence

Energy. Ambition. Confidence. Patience. Fearlessness. All these traits are associated with that mysterious quality of “entrepreneurialism”. Self-made men, such as Richard Branson and Alan Sugar, seem to exude different qualities from ordinary wage slaves.

But that is not the way things look to economists. It’s not that we are blind to the very idea that personality matters; it’s just that the evidence suggests a different story.

Andrew Oswald of Warwick University and David Blanchflower, a former member of the UK’s monetary policy committee, have assembled an intriguing picture of entrepreneurialism. They conclude that it is less a matter of character and more a matter of opportunity. The first piece in the puzzle is that many people who aren’t entrepreneurs claim that they wanted to be. Across Europe, 10-20 per cent of workers are typically self-employed, but 30-60 per cent say they’d like to be.

Oswald and Blanchflower also looked at evidence on happiness and concluded that the self-employed tend to be happier. In short, people say they want to be self-employed, and are more satisfied when they are self-employed, but frequently do not take the plunge.

Is this just because they lack that essential entrepreneurial spirit? Probably not. Oswald and Blanchflower originally intended to study the psychological make-up of entrepreneurs. That work went nowhere, because it seems that there is nothing distinctive about the psychological make-up of entrepreneurs.

What matters instead is capital. In surveys, would-be entrepreneurs identify lack of finance as the key obstacle. And in a clever piece of analysis, Oswald and Blanchflower compared people who had recently received a substantial bequest with those who had not. They found that such bequests – essentially twists of fate – were good predictors of whether people became entrepreneurs. They were especially influential on the decisions of people to become entrepreneurs early in life, presumably because older people have other ways to acquire funding.

Another substantial influence on entrepreneurship seems to be family background. People with at least one self-employed parent are two or three times more likely to be self-employed themselves.

Recent research by Simeon Djankov, Yingyi Qian, Gérard Roland and Ekaterina Zhuravskaya examines entrepreneurs in Brazil, with similar results. Unlike Oswald and Blanchflower, the Brazilian research looked at entrepreneurs who employed at least six people. Yet, again, they find that family background matters. If you have entrepreneurs in your extended family, you are much more likely to become one yourself.

There is some evidence that personality matters, although the data do not always reinforce the entrepreneurial stereotype. Djankov and his colleagues find that entrepreneurs tend to be more patient and more intelligent. There was no evidence that entrepreneurs were more confident than non-entrepreneurs, and entrepreneurs actually seem to be more averse to taking risks. Separate research by Djankov and different colleagues suggests that low taxes and efficient regulations are hugely important in encouraging the overall prevalence of new businesses.

The lesson I draw is that entrepreneurs learn from example and can be helped by social connections. The business environment – whether access to capital or freedom from too much bureaucracy – matters a great deal, too.

But the entrepreneurial spirit? Don’t look too hard for it. It may not exist.

Also published at ft.com.

Should I bet against my party winning?

Dear Economist,
As canvassing for the general election gathers speed, I’ve been thinking about the gambling possibilities. Without going into specifics, I’m considering placing bets on the rival team’s victory – as insurance in the event of having to live in a world not entirely to my liking.

Would this be psychologically effective, or am I wasting my money? What price should I place on my political ideals?
Justin, south-east London

Dear Justin,

In principle, your plan makes sense, but the practicalities are more troubling. One possible difficulty is “habituation”, our tendency to adjust psychologically to what we have. Some research on happiness economics suggests that people get used to extra money more quickly than they get used to ongoing situations such as a long commute. I could imagine you winning big when your party loses, but then quickly taking the extra cash for granted, while the smug face of the wrong prime minister infuriates you every day. Would your win have compensated you in any significant way?

A more basic problem is that many people object strongly to losses – a phenomenon called, unmysteriously, “loss aversion”. The idea is that a gain of £10 may be mildly pleasant, while a loss of £10 is infuriating.

Your plan guarantees a loss either way.

I worry that if you won your bet you would find it was scant compensation for your party losing the election, while if your party won the election, the celebrations would be ruined by your knowledge that you lost the bet.

If you do decide to bet, you are best placed to judge how much compensation is necessary. You clearly find the outcome more significant than I do. And a word of warning: you imply that if your favoured party wins the election, the world would be entirely to your liking. I can assure you that this possibility is remote.

Also published at ft.com.

17th of April, 2010Dear EconomistComments off

Why recessions aren’t all about job losses

Imagine a recession on Planet Vulcan. Thanks to weak demand, an able and hard-working Vulcan subordinate is simply not doing enough business to justify his salary.

The Vulcan boss calls his subordinate into the corner office for a frank and logical discussion of the options. They agree that it would be illogical to continue under the present arrangements, and that the Vulcan employee will accept a pay cut of 20 per cent for the time being. Both congratulate themselves on avoiding the bizarre human practice of sacking marginally profitable workers rather than adjusting their salaries.

Back on Earth, people do get sacked in recessions, and the received wisdom is that this is because wages don’t adjust. Some economists talk about “voluntary unemployment”. This odd term calls to mind the scenario of the Vulcan deciding he would prefer to spend some time on the beach rather than take the pay cut. Other economists speak of “wage rigidity”. Whatever we call it, inflexible wages are a puzzle, as the Vulcan approach does seem to have logic on its side.

But still, the story many economists tell is that wages don’t adjust much in recessions, and this fact helps to explain why unemployment takes the strain, plunging and soaring with the economic cycle.

Nevertheless, even if stubborn humans refuse to allow their wages to be renegotiated, there could still be room for wage adjustment because people are always gaining and losing jobs. If employers make generous offers during booms and stingy offers during recessions, we should see more wage flexibility and smaller fluctuations in unemployment as the economy booms and busts. But we don’t.

Or do we? Christopher Pissarides of the London School of Economics points to research showing that workers who hop from one job to another during a boom enjoy a hefty jump in take-home pay, while workers who change jobs during a recession do not. That suggests, claims Pissarides, that wages (or at least the wages offered to new hires) are more flexible than many economic theorists assume.

Not so fast, respond those theorists. In a recent article, Mark Gertler and Antonella Trigari explained why the studies to which Pissarides points might not mean what he thinks they mean: “Suppose, for example, that a highly skilled machinist takes a job as a low-paid cab driver in a recession and then is re-employed as a high-paid machinist in a boom.” Quite so: it is possible for an individual to experience cyclical wages by switching between two professions which themselves might have rigid wages.

This is all good clean fun, as I am sure you will all agree. And I saw the latest contribution to the debate presented by Gary Solon at the recent Royal Economic Society conference. Solon (with Pedro Martins and Jonathan Thomas) has been looking at data that allow him to observe the same companies recruiting over and over again – perhaps many times in a year – for the same entry-level positions. Solon’s data, from Portugal, clearly show something to surprise the economic theorists: wages do fall in recessions after all. An increase in the unemployment rate by one percentage point seems to suppress the real wages offered to new employees by 1.8 per cent. In other words, an increase in unemployment from 6 per cent to 9 per cent would depress the wages for new hires by just over 5 per cent.

This does not mean that wages are as flexible as they should be. The Vulcans might point to our wildly fluctuating unemployment rates and suggest that wages are still not absorbing enough of the strain of economic downturns. After all, a wage cut hurts. Losing your job hurts more.

Also published at ft.com.

Help! I’m trapped in a spiral of superstition

Dear Economist,
When I sneeze, people often say to me – I know not why – “bless you”. I do not reciprocate when others sneeze, for I refuse to subscribe to any form of superstition.
It follows that my well-wishers clearly are superstitious. Therefore, my exceptional politesse dictates that I ought to offer them a blessing whenever they expel extraneous sinal mucus.
But perhaps the whole of humankind believes that everyone else is superstitious and so this absurd tradition continues between people who ought to know better, each of them fearing that they will cause offence. It is an awful superstition-fearing spiral. How, dear economist, to break out of it?
Hugh Costello

Dear Hugh,

There seem to be competing explanations for the “bless you” convention – that the sneeze has driven out an evil spirit, that the blessing wards off the bubonic plague, or even that a blessing restarts the heart after the sneeze stops it.

You conclude that anyone who says “bless you” either believes one of these things, or is indulging another person’s presumed superstition. A more likely explanation is that saying “bless you” sends a signal that you have noticed another person’s existence. It is a weak signal, but not saying “bless you” sends a strong signal that you do not care to acknowledge the presence of another human being.

Here is a parallel. When I wish a colleague “good morning” it is not because I believe doing so will cause a good morning to spring into existence. It is because not saying “good morning” is rude – doubly so if she has said “good morning” to me.

You seem a literal-minded fellow, Hugh. When you addressed me as “dear” economist, were you expressing romantic feelings for me? If so, I am afraid I have some bad news.

Also published at ft.com.

10th of April, 2010Dear EconomistComments off

The only thing worse than high taxes is noticing they’re high

Not all taxes are created equal. They vary in obvious ways, such as who has to send money to the government. They vary in subtle ways, too – such as who actually pays the tax.

For instance, the government may ask the buyers of computers to pay a £50 computer purchase tax; or they may ask the sellers of computers to pay a £50 computer sales tax. It does not take much economics to show that it makes no difference to the taxman who writes the cheque: the after-tax price that buyers and sellers pay will be the same either way.

The same economics will also show who ends up absorbing more of the tax: it is whoever is less price-sensitive. If manufacturers and retailers are willing to expand dramatically or contract production in response to the price the market will bear, it will be computer buyers who pay most of the tax. If potential buyers will be driven away by modest price increases, it will be computer sellers who swallow the tax instead. If both buyers and sellers are price-sensitive, the tax will simply shut down the market.

All this is old news to micro-economists, but now a new strand of economic analysis is changing the way we think about taxation.

Raj Chetty, a young economics professor at Harvard, has been trying to understand how we perceive taxes. Taxes that seem equal to micro-economists may not seem equal to shoppers, or voters, and the difference matters.

In one of Chetty’s most memorable experiments, conducted with Kory Kroft and Adam Looney, the researchers persuaded a supermarket to change the way it displayed certain prices in store. Attached to the standard label, which advertised the price before sales tax, was an extra tab that also gave the post-tax price. The extended label should have made no difference: it was not obscuring the original information, but adding something that almost every shopper already knew. (Chetty’s team confirmed this with a survey: shoppers knew of the existence of the sales tax and most of them knew its level.)

By comparing what happened in the supermarket in the weeks before and after the experiment, Chetty was able to show that the labels made a big difference, cutting demand by about 8 per cent. This is an astonishing result: remember that the new labels told customers nothing they didn’t already know. It just brought it to the front of their minds.

This research on “tax salience” has practical implications. Because customers often ignore taxes that are tacked on at the checkout, they will end up paying a larger share of the tax. Perversely, the political appeal of such taxes may be lower, because voters are reminded of the tax every time they go shopping. A tax-inclusive price – such as Britain’s VAT – shapes purchasing decisions more, but is politically less visible. Small wonder rumours abound that VAT is to rise to 20 per cent after the election.

Chetty’s research also tells us about benefits. He complains about the way tax credits are paid out in the US. These credits should encourage more work, because, in effect, they boost the hourly wages of poor households by applying a negative income tax. But they are not salient: people receive a cheque from the government at the end of the year, and most people don’t understand how it boosts their hourly income. British tax credits are paid more frequently, but again, it is not always clear to recipients how their extra hours are linked to extra cash from the government. A more salient benefit would probably encourage longer hours and help lift families out of poverty. Such benefits have to be noticed to be effective – something you wouldn’t think politicians needed telling.

Also published at ft.com.

If I reduce ‘supply’, will my ‘price’ go up?

Dear Economist,
I was in a long-distance relationship with my ex-girlfriend for six months, I in Bangladesh and she in England. At first, everything was perfect, but then her demand curve for me seemed to be slowly shifting to the left. Paranoid, I started giving her a lot more attention, thereby increasing my supply. My price and marginal utility fell. We broke up once, but I convinced her to come back.
We went back to normal and we talked for hours everyday. I decided I would move for her and managed to get offers to study economics in the UK. She started acting very oddly, and a few days ago, she said she doesn’t love me any more and completely broke up with me.
We’re still friends and she’s started making excuses that don’t seem too rational. I think she’s confused and that she’ll come back if I play my cards right. Should I sharply reduce supply and hope that my price goes up? I love her more than anything. My demand curve for her is perfectly inelastic.
Anon, Bangladesh

Dear Anon,

I would put away the supply and demand curves and treat this as a problem of imperfect information. Before you cross continents, you must figure out what your ex-girlfriend thinks.

You blame your needy desperation for her change of behaviour – and it probably hasn’t helped. Even if you rescue this relationship you’ve basically conceded that you’ll be washing the dishes from now on.

But remember that she reduced demand before you increased supply. Why? Two hypotheses: she was bothered by the long-distance nature of the relationship, or she found someone she preferred to you. Test each against the fact that when you announced you’d be arriving on her doorstep, she dumped you. It’s pretty clear where the truth lies: it’s over. Yes, “sharply reduce supply” to this girl, but not because you hope to win her back.

Also published at ft.com.

3rd of April, 2010Dear EconomistComments off


  • 1 Twitter
  • 3 RSS
  • 5 Podcasts
  • 6 Facebook


  • Fifty Inventions That Shaped the Modern Economy
  • Messy
  • The Undercover Economist Strikes Back
  • Adapt
  • Dear Undercover Economist
  • The Logic of Life
  • The Undercover Economist

Free Email Updates

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

Join 188,947 other subscribers.