Tim Harford The Undercover Economist

Articles published in August, 2018

The Peter Principle is a joke taken seriously. Is it true?

There are a lot of idiots around — just look at the front benches of the House of Commons. But why do the idiots reach such elevated positions?

Many onlookers feel sure that the successors to David Davis, the UK’s former Brexit secretary, and Boris Johnson, ex-foreign secretary, cannot possibly be worse than the men they replace. Mr Davis was at least dignified in his resignation. Mr Johnson is widely regarded as the most disgraceful occupant of the post imaginable. And yet both men departed at the hour of their choosing: they were incompetent yet unsackable.

Part of this displays the awful dynamics of British politics today. But it also reflects something more universal. In 1969, Laurence Peter, a professor of education, and Raymond Hull, a playwright, published a management book that became a bestseller and an enduring classic: The Peter Principle. (UK) (US)

The Peter principle states that “every employee tends to rise to his level of incompetence”. If someone is good at her job, she’ll be promoted into a job that demands different skills. If she’s good at the new job too, she’ll be promoted again, requiring yet another set of skills. One day, she will arrive at a job for which she is wholly unsuited, and there she will stick. Since when did a manager ever get sacked for anything?

The Peter Principle is satire: it mocks management and it mocks books about management. It is striking, then, that most people take it quite seriously. The Harvard Business Review has published numerous straight-faced responses.

Two questions, then: is the Peter principle true? If so, what can we do about it?

We should acknowledge that the Peter principle may be an illusion. The top jobs are difficult to do well; they can make monkeys out of capable people. We shouldn’t be surprised if it is easy to call to mind examples of business or political leaders who have struggled. That would not by itself mean there are systematic errors in the way people rise to the top.

People were not promoted for behaviour that might seem correlated with managerial ability — in particular, those who collaborated with others were not rewarded for doing so

But this year, the economists Alan Benson, Danielle Li and Kelly Shue published what may be the first detailed empirical investigation of the Peter principle. Profs Benson, Li and Shue used data from a company that provides performance management software to sales teams; the data cover 214 firms, more than 53,000 workers and more than 1,500 promotions. This data set was ideal for identifying highly effective sales staff, and also provided enough information to evaluate what happened to a team once a hotshot salesperson was promoted into the role of team leader.

The authors of the paper discovered that the best salespeople were more likely to be promoted, and that they were then terrible managers. The better they had been in sales, the worse their teams performed once they arrived in a managerial role.

What’s more, people were not promoted for behaviour that might seem correlated with managerial ability — in particular, those who collaborated with others were not rewarded for doing so. What mattered were sales, pure and simple.

In short, Professor Peter was right. Brilliant people are promoted until they become awful managers. Perhaps employers are using the prospect of promotion as an incentive, and are willing to accept the collateral damage caused by all these terrible managers. If so, they are probably making a mistake. It would be better just to encourage the star salespeople with cash bonuses instead.

There is a more radical approach. One of my favourite IG Nobel Prizes (the IG Nobels reward “achievements that first make people laugh, then make them think”) was awarded to two physicists and a sociologist — Alessandro Pluchino, Andrea Rapisarda and Cesare Garofalo — who wondered how organisations might evade the logic of the Peter principle.

If performance at one level of a hierarchy is uncorrelated with performance at the next level up, the best strategy is simply to promote the very worst people. Nobody knows whether they will make good managers, but at least they will no longer be dreadful staff — or as Dogbert in the cartoon strip Dilbert put it back in 1995: “Leadership is nature’s way of removing morons from the productive flow.”

There are two difficulties with this approach: first, it may be too extreme to assume that no skills at all carry over from one job to the next; second, if the reward for failure is promotion, then the likely response is an organisation full of people bent on sabotage. So Profs Pluchino, Rapisarda and Garofalo suggest a compromise: promote people at random.

This may be the best response to a world where leaders stick around until they are ready to depart. But there is an obvious alternative: when people are not up to the demands of their job, we should not wait for them to resign. They should be sacked — or, perhaps better, demoted back to the roles where they once flourished. A mistake is regrettable, but stubbornly sticking to the mistake is far worse. Just ask the British electorate.


Written for and first published in the Financial Times on 20 July 2018.

My book “Fifty Things That Made the Modern Economy” (UK) / “Fifty Inventions That Shaped The Modern Economy” (US) is out now (or very very soon) in paperback – feel free to pre-order online or through your local bookshop.

Free email updates

(You can unsubscribe at any time)

The cool tools that are reshaping economics

If Hollywood is to be believed, every mad scientist who ever lived has a laboratory full of bubbling flasks, flashing consoles and glowing orbs. Science writer Philip Ball — who has visited countless research labs — tells me that reality is not so very different: the gear may be more subdued, but the gear is always there.

Science depends on tools, often instruments to detect or measure that which was previously undetectable — think of Galileo’s telescope or Newton’s prisms. Nobel Prizes have often been awarded to the physicists who developed such tools: the cloud chamber (1927); the electron microscope (1986); and LIGO, the laser interferometer gravitational-wave observatory (2017).

What, then, of economics? Economics has its own quasi-Nobel Prize, but it is a stretch to find a single example of a prize being awarded for the development of new tools or instruments. Simon Kuznets (laureate in 1971) probably has the best such claim, for developing the ideas behind the gross domestic product measure. Alas, GDP is a broad aggregate with limitations that Kuznets himself understood all too well.

The great Alfred Marshall described economics as being the study of humanity “in the ordinary business of life”. Unfortunately, in Marshall’s day — he died in 1924 — there was no way to observe the ordinary business of life, except perhaps as an anthropologist. Economists spent a lot of time in armchairs, thinking hard about theory rather than measurement.

Some economists now make progress using old tools from other fields. MIT’s Esther Duflo, winner of the prestigious John Bates Clark medal, answers economic questions using randomised controlled trials. RCTs are typically dated back to Austin Bradford Hill’s 1948 trial of streptomycin for tuberculosis. (Hill was trained as an economist, so perhaps we can score that one for the profession.)

But the holy grail is to be able to observe the ordinary business of life in detail, in real time, and at scale — ideally all three at once. That was once an impossible goal, but three new developments put that goal within reach.

The first is the availability of high-resolution satellite images. In the mid-1990s, an economist named Alex Pfaff realised that these images could be used to answer questions about the connection between development projects and deforestation in the Amazon.

Hundreds of others have followed suit. Satellites can easily measure illumination at night, a simple way to track economic activity and patterns of urban development. It is also possible to measure various kinds of air pollution, and to observe the growth of crops. Algorithms are starting to extract subtle information at scale: how many Ethiopian homes have tin roofs? Which roads in Kenya are in good condition? And ever-cheaper small satellites are taking detailed photographs of everywhere, every day.

An even bigger change is that economists are using administrative data. I realise that “economists are using administrative data” is a contender for the most boring sentence uttered in 2018. But over the past two decades or so, this has been a quietly revolutionary move.

Administrative data are the numbers generated by governments or private companies for the purposes of getting things done. Schools keep track of attendance and grades. Tax authorities know your (declared) income — but also where you live, your age, and perhaps who your children are.

As such records have become digitised, they can be used to answer serious questions in research. For example, tax data can tell us the extent to which the children of rich or poor parents grow up to be rich or poor themselves. These detailed data are now at the forefront of empirical economic research.

According to Dave Donaldson, who like Prof Duflo is a John Bates Clark medallist at MIT: “In my field, international trade, I rarely see a paper that doesn’t use customs-level data. Every shipment generates a record which will specify what it is, where it came from, where it’s going, and the tax paid.”

A third measurement tool is the mobile phone. Every time a call is placed, the phone company generates a record of who called whom, when, for how long, and where the phones were, sometimes to within less than a hundred metres. With that kind of “metadata”, economists and other researchers can ask questions such as: how rapidly are people moving around, and to what extent is that correlated with the spread of an epidemic? Is a city’s transport infrastructure working well? How quickly are refugees integrating into a new society?

This is both an opportunity and a challenge for economists. Data scientist and economist Josh Blumenstock told me that “anyone who graduated with an economics PhD more than five years ago has no idea how to handle this data, and is frantically scrambling.”

Surely the scramble will produce results. At last, it is possible not just to theorise about Marshall’s “ordinary business of life”, but to observe it. Our tools are letting us see something new — and what we can see determines what we can think.

Written for and first published in the Financial Times on 13 July 2018.

My book “Fifty Things That Made the Modern Economy” (UK) / “Fifty Inventions That Shaped The Modern Economy” (US) is out now (or in the US, very very soon) in paperback – buy online or through your local bookshop.

Free email updates

(You can unsubscribe at any time)

What does a robot accountant look like?

What shall we do when the robots take our jobs? Last week’s column discussed mass technological unemployment, and readers were quick to write in with variants on the same common sense suggestion: tax the robots.

“Give a robot a deemed income equivalent to the wage of the human it is replacing, with a default of the average income,” wrote one. A similar suggestion: “If one robot displaced five people earning a modest £20,000 each, then that robot could be said to have an economic value of £100,000 per annum.” Both correspondents proposed an income tax or similar levy on this robot-generated output.

I’m grateful for all constructive comments, but particularly grateful for these because I think they are wrong in a fascinating and instructive way. The fault is mine: I set a trap every time I talk about “robots” and “jobs”. That is not how automation happens.

Consider an idea dreamt up in 1978, released in October 1979, and so revolutionary that the journalist Steven Levy could write just five years later: “There are corporate executives, wholesalers, retailers, and small business owners who talk about their business lives in two time periods: before and after the electronic spreadsheet.”

Spreadsheet software redefined what it meant to be an accountant. Spreadsheets were once a literal thing: two-page spreads in a paper ledger. Fill them in, and make sure all the rows and columns add up. The output of several spreadsheets would then be the input for some larger, master spreadsheet. Making an alteration might require hours of work with a pencil, eraser, and desk calculator.

Once a computer programmer named Dan Bricklin came up with the idea of putting the piece of paper inside a computer, it is easy to see why digital spreadsheets caught on almost overnight.

But did the spreadsheet steal jobs? Yes and no. It certainly put a sudden end to a particular kind of task — the task of calculating, filling in, checking and correcting numbers on paper spreadsheets. National Public Radio’s Planet Money programme concluded that in the 35 years after Mr Bricklin’s VisiCalc was launched, the US lost 400,000 jobs for book-keepers and accounting clerks.

Meanwhile, 600,000 jobs appeared for other kinds of accountant. Accountancy had become cheaper and more powerful, so people demanded more of it.

What does a robot accountant look like? Not C-3PO with a pencil sharpener, that’s for sure. One might say that Microsoft Excel is a robot accounting clerk. A more plausible answer is that there is no such thing as a robot accountant. One day we may have androids sophisticated enough to do everything human accountants do now, but by then the very concept of an “accountant” will have changed beyond recognition.

So it is misleading of me to write of “robots” taking “jobs”. What actually happens is that specific tasks are automated, rather than the broad bundle of tasks that together constitute a human “job”. Automating tasks means reshaping jobs. The process can create jobs or destroy them, and will usually do both.

In their recent book, The Future of The Professions (UK) (US), Daniel and Richard Susskind offer numerous examples of the spreadsheet dynamic in action: algorithms that scan mammograms and spot trouble that humans miss; online tutorials that monitor students and alert teachers to where the child is struggling; “document assembly systems” that quiz clients and then draft legal contracts.

Another example. If I didn’t have the use of email, internet search and a mobile phone, I would need to employ someone as a personal assistant. But I have had these tools for a long time, so I have never had a secretary. Should I have to pay the never-employed secretary’s tax bill, because I own a smartphone?

White-collar anxiety about automation is new, but the problem is old. Mike Mulligan and His Steam Shovel (UK) (US) is a Depression-era children’s book about technological obsolescence. (It is wonderful.) The steam shovel’s name is Mary Anne; she “could dig as much in a day as a hundred men could dig in a week”. So is Mary Anne a “digging robot” who destroyed 500 jobs? Yes; no; maybe. After a while the question seems ridiculous. Nor would many sensible people argue that Mike Mulligan, Mary Anne’s owner, be liable for 500 tax bills.

As any tax wonk can tell you, whatever we choose to tax — land, capital, profits, value-added, imports, wealth, greenhouse gas emissions — inevitably turns out to be a more ambiguous concept than it might appear, especially since ambiguity is often tax efficient.

But the category of “robot” is particularly difficult to define, and therefore to tax. We cannot tax the androids who march into our workplaces, stand by while we clear our desks, then sit down to replace us: they do not exist and it is hard to see why they ever would.

In a world of mass technological unemployment we are certainly going to need to tax something other than labour income alone. There are several plausible candidates. “Robots” is not one of them.


Written for and first published in the Financial Times on 6 July 2018.

My book “Fifty Things That Made the Modern Economy” (UK) / “Fifty Inventions That Shaped The Modern Economy” (US) is out now (or very very soon) in paperback – feel free to order online or through your local bookshop.

Free email updates

(You can unsubscribe at any time)


  • 1 Twitter
  • 2 Flickr
  • 3 RSS
  • 4 YouTube
  • 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

Search by Keyword

Tim’s Tweets

Free Email Updates

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

Join 164,391 other subscribers

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