The job advert issued by Dominic Cummings for people to work in government has attracted a wide range of comments. One particular focus has been on the sorts of skills he is looking for.
Computer science, forecasting, artificial intelligence, causality theory — all these topics excite his interest. Cummings advocates a small selection of scientific papers with which applicants should be familiar. He believes that humanities graduates are unlikely to be aware of them.
The papers are indeed quite challenging mathematically. Even the smartest arts graduate might struggle to cope with their content simply because of the language — maths — in which they are written.
The implication is that those with expertise in the humanities need not apply. Indeed, economics is conspicuous by its absence from the now notorious reading list.
I can empathise with his focus on the hard sciences. But economics does have one very powerful, general insight into behaviour that Cummings should heed. In fact, everyone — whether working in the civil service, think tanks or university social science departments — should be familiar with it.
It is, quite simply, that agents respond to incentives. When the set of incentives faced by an individual, a company, or a government changes, behaviour changes too. Different decisions are made as a result.
Two snippets of recent news, chosen almost at random, can illustrate the power of the concept.
Beggars have started to travel from Glasgow to Carlisle to ply their trade. In Scotland, the penalty for aggressive begging is up to a year in jail and a £5,000 fine. In England, it is only a £1,000 fine. These facts are sufficient to explain why Carlisle has become more attractive.
On a note that will be more relevant to most people, more than a million people a month now fail to turn up for GP appointments. From June to November last year, a record 7.8m patients did not attend.
This bad behaviour imposes extra costs on the NHS and makes it more difficult for people who really need to see a GP to get appointments.
A simple solution is to charge for visits to the GP surgery. It would not eliminate the problem, but it would make a big difference. Once having paid, people would be much more likely to turn up.
Any proposal to introduce charges in the NHS causes the left to froth at the mouth. The standard argument is that charges would deter poor people from accessing healthcare.
It does not seem to do so in countries such as Ireland and Sweden. Both charge people to see their doctors. The impact is mitigated in the former by an annual cap on charges, and in the latter low-income people can visit for free. Other EU countries have similar schemes — in France, the principle of health services is pay upfront, get reimbursed later.
It is not necessary to believe that people act in a completely rational way all the time. They obviously don’t. But incentives work. Empirical examples of the principle can be found every day, in every situation.
A simple, sensibly designed set of incentives is worth a tonne of regulation. A clear understanding of this principle should be the key thing Cummings considers when hiring a new set of government “weirdos” to shake up the civil service.
As published in City AM Wednesday 15th January 2020
Calling all employers: what was in your Christmas stocking? Did you find the latest gadget designed to enhance productivity?
The innovative device, featured in the media during the festive season, is a toilet with a downward sloping seat. The company which makes it, StandardToilet, has conducted extensive tests. A slope of 13 degrees is exactly the right tilt to make workers feel miserable without causing any lasting pain.
This way, it is alleged, staff will spend less time on the lavatory and more on their work. Output per worker will rise.
The very slow growth in productivity, the shorthand word for “output per worker”, was a defining feature of the past decade. Typically, productivity in the UK grows by around two per cent a year. During the latest decade, annual growth was barely above zero, at 0.3 per cent a year.
Higher productivity growth means that real wages can increase faster. Bigger pay packets mean more tax receipts for the government, so spending on public services can also rise.
Clearly, productivity growth needs to be boosted. But contrary to the claims, devices such as the sloping toilet may be one of the problems rather than the solution.
Recent years have seen a surge in technical innovations designed to control in ever more detail the tasks which workers perform. This is particularly the case at the lower end of the pay scale — think of warehouse jobs, delivery services and the like.
This ultra-micromanagement of time is intended to increase productivity. Its side effects include high employee turnover, resentment, and sheer bloody mindedness.
Why bother to make even the slightest bit more effort than your contract specifies under such conditions? Especially when, with the UK at full employment, you can walk into an equally crap job just down the road the very next day.
Nobel laureate George Akerlof addressed these issues in a famous paper 40 years ago entitled “Labor Contracts as Partial Gift Exchange”.
It was stimulated by a study which found that a group of women in a low-level, routine job exceeded the minimum work standards of the firm by an average of 15 per cent. This could not be explained by the standard economic theory of rational behaviour.
A key point for Akerlof was that the women did not work in isolation from each other. They interacted. He argued that, through these interactions, the workers acquired sentiment both for each other and for the firm. As a result, a situation developed which depended on the “norms” of gift exchange.
On the workers’ side, the “gift” given was work in excess of the minimum work standard. On the company’s side, the “gift” was wages in excess of what these women could receive if they left their current jobs.
The new tools of time and task micromanagement do the opposite. They are counter-productive, ensuring that virtually no employee will do more than the absolute minimum required by the contract. Why not try fur-lined, heated lavatory seats instead?
As published in City AM Wednesday 8th January 2020
Behavioural economics — which extends the ability of economics to explain the world — has become very fashionable.
Richard Thaler, Nobel Prize winner for his work in this area, observed that most of the time, the rational choice model of standard economics works well. People gather information on the various alternatives open to them, and choose the one which fits their preferences most closely.
Behavioural economics comes into play when people are observed to deviate from the predictions of this model. To explain this, Thaler points out that economists borrow bits from psychology and add them on to their basic rational choice theory. He has shown how various psychological concepts, such the so-called “endowment effect” and the “sunk cost fallacy”, can be used to explain why people make irrational choices.
Yet one important piece of psychology which economists have not used so far is the concept of cognitive dissonance.
The US psychologist Leon Festinger encountered a classic example when he infiltrated a group which believed that a catastrophic flood would end the world on 21 December 1954. The appointed day came and went, and the world was still there. Rather than processing this information rationally and abandoning their discredited beliefs, group members adhered to them even more strongly. The fervour of their proselytising increased.
Another important instance is seen in the Ukrainian famine of the early 1930s. On Stalin’s instructions, the Soviet military and armed members of the Communist Party seized the grain in the Ukraine. Millions died as a result.
The ardent young Communists turned on the Ukrainians and accused them of terrorism. The peasants were, despite all the evidence to the contrary, deliberately starving themselves to death in order to discredit socialism. Any Party member who disagreed was shot.
A much less harrowing example is given by Jeremy Corbyn and his cult during the General Election here.
The Friday before polling day, I was in my home town of Rochdale, having a drink with a couple of long-standing friends, both very experienced local Labour members. They were certain that Labour would lose. Brexit was certainly being mentioned on the doorsteps as a reason why voters weren’t supporting Labour, but the main reason was Corbyn.
My friends made specific predictions on the basis of rational analysis of the evidence. The neighbouring seat of Heywood and Middleton (majority 8,000 in 2017) would be lost. It was. Elsewhere in Greater Manchester, the Tories would win Leigh, Labour since 1922. They did.
Information of this kind, gathered on the ground by experienced agents, was fed back to the Labour leadership throughout the campaign from all over the country. It was pointedly ignored, possibly on the grounds that the informants were right-wingers trying to discredit the Dear Leader.
The electorate took some time to discover the reality of Corbyn. But once they had done so, far from displaying cognitive dissonance, they made a rational choice.
Now Labour, if it is ever to recover from defeat, should learn the same lesson.
As published in City AM Wednesday 18th December 2019
It would take a heart of stone not to be amused by Emmanuel Macron’s current predicament.
The French President is trying to position himself as the leader of Europe. But at the same time, the streets of the major cities in France are, quite literally, ablaze. France’s public services are crippled by the biggest strike in decades.
The reason is the massive unpopularity of Macron’s proposed reforms to public sector pensions.
The retirement age in France is still only 62, compared to 66 in the UK. In general, the proposal is not to increase the age, but to pay slightly reduced benefits before the age of 64. However, the most contentious part is to modify or even scrap completely the scams under which many public sector workers get to retire much earlier on full pension.
France faces a serious pension funding problem. Spending on pensions costs no less than 14 per cent of the country’s GDP. Only Greece and Italy are higher in the entire developed world.
That is probably why opinion polls put support for these reforms among the population as a whole at around 70 per cent, with even greater support among the young, even if many from the minority directly impacted have taken angrily to the streets.
Still, pension reform is known to be potential political dynamite — and not just in France. Raising the pension age for women has become an issue in the current General Election here.
The Women Against State Pension Inequality (WASPI) campaign argues that when the retirement age was raised for UK women in a series of reforms, the 3.8m affected women, born in the 1950s, did not have enough time to adjust.
Despite that fact that this is not mentioned in Labour’s manifesto, John McDonnell has pledged to compensate these women. The cost is a mere £58bn — around three per cent of GDP — almost all of which would need to be borrowed.
As it happens, considered purely in isolation, a reasonable case can be made for increasing the general level of the basic state pension in the UK. Pension costs here are below the OECD average as a percentage of GDP, at only half the level of France. But this would not be a free lunch. Other aspects of public spending would have to be correspondingly reduced.
The myth persists that people are investing in a funded scheme with their taxes. They pay the money in when they are working, the investments grow, and there is a pot earmarked for them at their retirement. In reality, the cost of paying an individual’s pension falls entirely on those who are working during his or her retirement.
For anyone in work, the government’s promise of a pension in the future is rather like a slightly dodgy IOU. The amount you will end up getting depends upon how fast the economy grows over the coming decades, how long people live, and ultimately on the generosity of those in employment when you retire.
Political debates on pensions are usually rather depressing for economists because of either the inability or the reluctance to understand this point.
Much as it sticks in the throat to say so, President Macron is to be admired for the stance he is taking.
As published in City AM Wednesday 11th December 2019
Image: President Macron protests by Jeanne Menjoulet via Wikimedia licensed for use CC BY-2.0
An important piece of social news emerged last week. According to the Office for National Statistics, the divorce rate in 2018 fell to its lowest level for nearly 50 years.
The overall trend is clear and well-established. The divorce rate rose steadily from the late 1950s, with sharp rises immediately following the Divorce Act of 1969, to the early 1990s.
Since then, with minor blips, the rate has fallen. It now stands at just over half the level of its 1993 peak.
Economic theory has a lot to say about marriage and divorce.
This may be surprising to many. But economists believe that the theory’s basic model, that of someone making a rational choice from the alternatives on offer in any given situation, is universal in its application. The institution of marriage is a key social phenomenon, and so the rational choice model ought to be able to give an explanation as to why it exists.
The Chicago economist Gary Becker received the Nobel prize in 1992 for his pioneering work in this area. Essentially, the participants in a marriage reap what economists see as the gains from trade. One partner goes to work and earns money, and the other raises children and does housework. By each concentrating on the activity which he or she does better than the other, both parties benefit.
Implicitly, Becker took as the social background to his theory the institutional structure of marriage and the family as it existed in the Midwest of the USA in the 1950s. Gender roles have certainly evolved since then, but his basic insights remain valid.
A much more general theory of marriage is linked with the work of economists such as Bob Rowthorn, former head of the Cambridge economics department.
In this approach, marriage is seen as an institution for providing couples with the confidence to make long-term investments in their relationship. The basic theme is that marriage should be seen as an institution for creating trust between individuals in the sphere of family life.
Given this emphasis on both trust and the long term, it is curious that many metropolitan liberals, not least Supreme Court justice Lady Hale when she headed the family courts, appear to see marriage as no better than any other form of family structure.
The empirical evidence overwhelmingly supports the special value of marriage for the individuals concerned, for their children, and for society. Indeed, there are few hypotheses in the social sciences which receive such clear confirmation from serious research.
For example, most children grow up to be useful and well-adjusted members of society regardless of family structure. But the incidence of crime and mental illness among children whose parents have divorced, while low as a proportion of all such children, is much higher than it is among those whose parents remain married.
The falls in the divorce rate can be seen as rational learning by the generation who were children themselves when divorce was at its peak. They see the costs imposed on them. And society as a whole will reap the benefits in years to come.
As published in City AM Wednesday 4th December 2019
One of the first tasks facing whoever becomes chancellor after the General Election will be choosing the next governor of the Bank of England.
Getting to make this choice would be a key step in the plans of Labour’s shadow chancellor John McDonnell to shake up the Bank of England, but his radicalism is not simply a matter of practical policies. McDonnell appears to want a new intellectual approach at the central bank and across government, one not based on existing economic theory.
Many people on the left, such as McDonnell, have criticised austerity ever since the financial crisis of the late 2000s. They often believe this to be a criticism of conventional economic theory.
But these are two different things. They are not the same. It is certainly possible to construct a coherent critique of austerity purely on the basis of standard theory.
Economists disagree on the matter because of different interpretations of the empirical evidence, rather than on the basic theory.
However, Labour’s election manifesto appears to want to ditch the most powerful insight of the whole of economic theory. Namely, that people react to changes in the incentives which they face. If incentives change, behaviour changes.
For example, even the most dedicated critic of economics will slow down when approaching a speed camera if they are exceeding the speed limit. On the open road, there is a chance of being stopped by the police, but it is very small. Yet the probability of incurring a penalty rises sharply in the presence of a speed camera. And so behaviour changes — the driver slows down.
We can apply this insight to taxation. Already, the top one per cent of earners in the UK pay well over 30 per cent of all income tax, according to calculations by the Institute of Fiscal Studies.
Before tax, to be in the top one per cent you have to earn at least £166,000 a year. After tax, the figure falls to £111,000. So these individuals — just over 300,000 of them — are already handing over one third of their income to HMRC.
We know from Labour’s manifesto that McDonnell wants to raise many billions more in tax from the top five per cent of earners.
Much of this would potentially fall on the top one per cent, where the money is really concentrated. But this would require these taxpayers to stand still and wait to be plucked like so many golden geese. Economic theory, supported by a vast amount of empirical evidence, suggests that this just would not happen.
he same ignorance of fundamental, well-supported economic theory is seen in Labour’s policy on corporation tax. The proposal is to raise the rate from 19 to 26 per cent.
The basic problem here is that if the tax rate changes, the behaviour of a company will change.
The company may hold down wages. It may not take on extra staff, or even get rid of existing employees. It may cut dividends, so that pension funds and the income of pensioners is reduced. Or it may slash investment, so that the workers in the firms which supply to the company will suffer.
Instead of jettisoning economic theory, Labour should learn from its most powerful insights.
As published in City AM Wednesday 27th November 2019
Image: John McDonnell by Sophie Brown via Wikimedia licensed for use CC BY-SA 4.0
Ardent Remainers had a rare bit of good news at the end of last week.
The latest statistics for the German economy showed that, contrary to expectations, it had not fallen into recession in the July-September period.
Economists have come to define a recession as a period when a country’s GDP falls for two quarters in succession. There is no firm scientific basis for this — it has simply emerged over time as a consensus among the members of the profession who follow these things.
A more decisive indicator is if the economy shrinks compared to the same quarter in the previous year. A cynic might say that the less stringent definition, a fall for just two successive quarters, gives economists more to write about, as such events are more frequent than year-on-year falls.
Still, the two-quarters fall is the definition commonly used. German GDP fell in the second quarter, April to June, but it rose — by all of 0.1 per cent — in the third quarter. A recession has been avoided, and Remainers were relieved by the demonstration (if weak) that the EU’s powerhouse economy was not in decline.
Not so fast. We should take economic statistics such as the growth of GDP with a firm pinch of salt, and remember that, outside of the financial markets, almost all economic data is estimated.
We cannot put the economy onto a set of scales and weigh it. Its size has to be estimated, using a wide range of basic information and assumptions. And that means that the numbers cannot be taken at face value.
For example, commentators noted that the fall in German GDP in the second quarter had been revised, from minus 0.1 per cent to minus 0.2 per cent.
Changes of this magnitude are small fry. The US has probably the most sophisticated system for estimating the so-called national accounts — the maze of economic statistics which make up the economy. Yet during the depth of the financial crisis, in the fourth quarter of 2008, American GDP was initially estimated to have fallen by 6.4 per cent at an annual rate. As more information came to light, over a period of years, the figure is now believed to be a drop of 8.2 per cent.
And this is by no means the biggest revision which has been made. In the recession of early 1975, for example, GDP was initially thought to have fallen 11.4 per cent in the first quarter. The figure was eventually revised to just 4.7 per cent.
These revisions are inevitable given the assumptions — many of them arbitrary — that have to be made in economic analysis. How do we measure the output of the armed forces, for example, or civil servants? There is no market in which these outputs are sold which might give us some clue. The answers are provided by a set of conventions.
Economic statistics therefore paint a broad picture and do not have the precision with which the media has come to treat them.
As a rule of thumb, quarterly GDP growth of 0.5 per cent or above is a sign of a healthy economy. The German data shows that, in terms of the big picture, the economy is currently struggling, regardless of whether an individual estimate is plus 0.1 or minus 0.1.
As published in City AM Wednesday 20th November 2019
The media has been awash over the past week with stories about the thirtieth anniversary of the fall of the Berlin Wall.
My favourite vignette concerns a couple living in East Berlin who were delighted to have a telephone installed in their apartment only weeks before the Wall came down. They had been on the waiting list for 19 years.
This captures the essence of socialism. The system could generate a tolerable standard of living for citizens, but it was grossly inefficient and run for the benefit of the producers rather than the consumers.
The old nationalised industries in Britain also offered us a glimpse of what life would be like under socialism. Under British Rail, new heating stoves really were installed in station waiting rooms on the very day that the line was closed to traffic for ever.
In the 1970s, people routinely waited at least six months for the nationalised telecoms company to install a domestic phone line.
This “producer-just” attitude persists. Today’s Labour leadership has, for example, defended firefighters in the face of the recent damning criticism of their performance in the Grenfell Tower tragedy. For Jeremy Corbyn and John McDonnell, the interests of the producers — the public sector workers, even the fire chief who could retire at 50 with a pension of £140,000 a year — come first.
Some will feel that this is unfair to socialism. Socialism in practice may have had its faults (such as the liquidation of nearly 100 million people by their own governments), but a better, different kind of socialism is apparently on offer in the future.
Remarkably, the leaders of the Communist Parties in Eastern Europe appear to have believed the same thing. The ideologists in their Politburos described countries such as East Germany as examples of “actually existing” socialism — in contrast to the nirvana which would exist at some unspecified time in the future.
But we can only judge a system by its performance in practice, not by some Platonic ideal of what true believers imagine it might do. Everywhere it has been tried, socialism has been a failure. This simple fact cannot be repeated too often, particularly to younger generations to whom 1989 may seem as remote as the days of the Roman Empire.
Modern history has provided us with a whole series of what are termed natural experiments.
We cannot set up (as in the natural sciences) a laboratory in which one society is started up on socialist lines and the other on capitalist ones, and then observe their performances over time.
But we can observe the United States and the Soviet Union, West and East Germany, South and North Korea, China when it was purely socialist and China when it subsequently embraced a market-oriented economy.
In every single case, capitalism has delivered better outcomes: higher living standards, longer life expectancy, more holidays, more provision of health and education — more of almost everything except slave labour and environmental pollution.
Capitalism can be criticised, but its faults are nothing compared to those of socialism.
As published in City AM Wednesday 13th November 2019
Image: Berlin Wall by Ira Gorelick via Pixabay
The young contestants on Lord Sugar’s reality TV show The Apprentice sparked outrage last week.
They appeared to have virtually no knowledge about the Second World War.
The online clips of the sequence capture to perfection their expressions of bovine outrage at even being expected to know such an esoteric thing as how long the War lasted. For many viewers, it was a powerful argument for an immediate raising of the voting age to at least 35.
Trying to be fair, I reflected on my own knowledge of the Boer War. The gap between that and my birth is not that much more than the one between World War II and Lord Sugar’s wannabes.
I have to confess that it is very sketchy. I know it was around 1900. The Boers nearly pulled off a shock victory, but our boys pipped them in extra time. And there was a siege at somewhere called Mafeking.
That said, the Second World War is of far more historic significance than Britain’s colonial skirmishes. The participants in The Apprentice might, one would think, be expected to know more about it.
Their ignorance reflects a more general trend towards what can be described as a shortening of cultural memory.
This can be seen in many seemingly unrelated ways.
Baby names, for example, are important indicators of culture. The example may seem slightly frivolous but, as the American polymath Stephen Pinker has written, the choice of name “encapsulates the great contradiction in human life: between the desire to fit in and the desire to be unique”.
Whenever there is a royal birth in the UK, the name given to the new baby enjoys a surge in popularity as people copy what is fashionable. But there is also the desire to give your baby a distinctive name.
In the twenty-first century, the turnover in popularity of names has increased dramatically. It was steady for almost the whole of the twentieth century – once a name had made it into the top 10, say, it tended to stay there a long time. But the “churn” is now running at a rate nearly 10 times greater.
The outcomes of behaviour on the internet provide a more obvious example. At any point in time, the most popular memes on Twitter or the most viewed videos on YouTube are often millions of times more popular than the average. But their popularity is ephemeral.
In what many would regard as far more serious areas of economic activity, away from popular culture, this shortening of collective memory may have disturbing consequences for all of us.
In financial markets, for example, there is a traditional tale that a crisis happens precisely when the institutional memory of the last crisis has faded. But if traders and decision-makers develop shortened memories, crises could become more frequent.
Already bodies such as the International Monetary Fund are warning about the build-up of private sector debt, even though it is scarcely a decade since such debt precipitated the global financial crisis.
The apparent ignorance of The Apprentice candidates offers a warning to us all.
As published in City AM Wednesday 6th November 2019
Image: World War Two posters by Andrew Curtis via Geograph licensed for use CC BY-SA 2.0
Brexit is about much more than the economic costs and benefits, but the idea that the former dramatically outweigh the latter has become the received wisdom in much of the media.
Report after report emerges which purports to show that, under any of the various trade arrangements envisaged, the UK will be worse off as a result of Brexit.
These studies are not wrong. They all use perfectly standard economic theory to arrive at their conclusions. But they are misleading.
The real problem is that they miss out key bits of the story. We can think of the classic tale of the person dropping his car keys in the street at night. He only looks for them under the street lamp, where the light is.
In the same way, standard economic analysis of Brexit only illuminates part of the landscape.
The explanation of why trade occurs between countries was given 200 years ago by the great English economist David Ricardo. It is still the basis of the modern economist’s understanding of trade.
Ricardo imagined, to illustrate his theory, a world with just two countries and two products. His examples were England and Portugal, and cloth and wine – but they could have been any countries and any products.
Ricardo asked a simple but profound question. If England could produce both cloth and wine more efficiently than Portugal, why would trade take place at all? How could the more efficient country, England, benefit from trade?
His answer introduced the fundamental concept of comparative advantage. England had an absolute advantage in producing both cloth and wine, but the country should choose to specialise in producing the one in which its advantage compared to Portugal was bigger. Both would benefit if England produced only cloth and Portugal only wine, and they traded.
Economics has moved on in the past two centuries, but the concept of comparative advantage, modified by factors such as the distance between countries, is still seen as a key determinant of trade patterns.
In terms of Brexit, introducing complexities like tariffs into the picture essentially affects the amount which is traded, and not the structure of trade in terms of who sells what to who.
If the basic pattern of trade is fixed by comparative advantage, then if Brexit means higher tariffs for the UK, as a country we will lose out. In a nutshell, this is what lies behind all the negative assessments of the impact of Brexit.
However, the key word in the last paragraph is “if”. Like almost all economic theory, these models are static. They assume that the network of trade is fixed, and analyse the consequences of changing prices through tariffs.
The EU has become mired in regulation and the level of innovation is low. Outside the EU, the UK could alter the patterns of trade by innovating in, say, biotech or AI-related products and services. It is this dynamic response, and not the static one, which will determine whether or not Brexit is a success.
Economic models which claim to analyse the impact of Brexit are true – but only up to a point, Lord Copper, as the saying goes.