Economyths - ten ways economics gets it wrong Part 7

At Milton Friedman's 90th birthday bash, his former student Donald Rumsfeld said that "Milton is the embodiment of the truth that 'ideas have consequences.'"54 The subprime crisis, in turn, is graphic proof that economic myths have consequences. These myths include: * The myth that the economy is governed by mathematical laws, so risk can be controlled using equations * The myth that individuals or households act independently, and are immune to herd behaviour * The myth that markets are stable, so the future will resemble the past * The myth that investors or households or firms like Lehman Brothers act rationally, and don't make poor economic decisions that go against their best interests * The myth that economics is an objective, impartial mathematical science, rather than a cultural phenomenon that itself influences the economy Finally, there is the myth - discussed further in the next chapter - that free markets are also fair. In many ways the subprime story was less about risk than about power. Friedman's utopia of a society of maximum individual freedom sounds attractive, until you see how it plays out in the real world. Giving companies like Goldman Sachs the maximum freedom to package subprime mortgages at teaser rates for financially illiterate people, construct multi-trillion-dollar balloons of credit out of thin air, and then extract money off the government when the scheme goes sour, probably isn't what he had in mind.

The crisis, whose consequences are still playing out, is unlikely to be the last demonstration of the power that these ideas still hold. Mainstream economists in academia and government remain blinded by their Pythagorean vision of a perfect economy, and therefore fail to learn from their mistakes. By continuing to propagate these myths, our universities and business schools sow the seeds for future financial catastrophes. Just as faulty risk models make the economy more risky, so a worldview that sees and treats the economy as inherently stable and self-regulating will - by loosening regulations - eventually turn it into the opposite. The point is not just that the economy is reflexive, and therefore hard to predict; but that our ideas and myths have shaped the economy in a particular way, and led to the designing in of instability. This is perhaps the clearest example of how economic theories influence the world, and therefore lose any pretence at objectivity. (As discussed later, this process also works in the other direction: the real economy affects theory by selecting ideas that suit its power structure.) Indeed, it is ironic that, while economics has aligned itself as stereotypically masculine, the actual economy has become increasingly connected, changeable, and unpredictable - all traits that are culturally considered stereotypically female.55 Economists are the jilted lovers of the science world - the more rigidly they approach their subject, the more it mocks them with spurious and headstrong behaviour.

Mathematical models are useful tools for simulating and understanding the economy, but they will never be able to accurately predict its course, or fully capture risk. To quote author and derivatives trader Pablo Triana: "Reality is much more ferociously untamable. The randomness is not just wild; it's savagely uncageable, abominably undomesticated. No equations can subjugate it, control it, or decipher it. Where anything can happen, there are no mathematically imposed bounds."56 Or as science historian Evelyn Fox Keller puts it: "Nature ... is not completely bound by Logos: it remains caught in an essential duality. If in some respects it is subject to the light of reason and order, it is also enmired in the dark forces of unreason and disorder. The forces of unreason, in Greek mythology and drama most often embodied in the earth G.o.ddesses or the Furies, are never fully vanquished, even when they are subdued."57 In its way, the subprime crisis was another manifestation of this eternal archetypal battle between chaos and order; and one in which chaos won.

In the 19th century, a band of outsiders sought inspiration from science and engineering to create a new theory of economics. It is time to do so again.

The Great Yinification.

Scientists have long dreamed of a Grand Unified Theory that will explain all the known physical forces and the evolution of the universe. Some have even hoped for a theory that would unite physics, chemistry, biology, psychology, sociology - all the physical and social sciences - in a single model. The ultimate aim of the Enlightenment, according to philosophers Max Horkheimer and Theodor W. Adorno, was to make "dissimilar things comparable by reducing them to abstract quant.i.ties. For the Enlightenment, anything which cannot be resolved into numbers, and ultimately one, is illusion; modern positivism consigns it to poetry. Unity remains the watchword from Parmenides to Russell. All G.o.ds and qualities must be destroyed."58 As Vilfredo Pareto wrote: "It is only the imperfections of the human mind which multiply the divisions of the sciences, separating astronomy from physics or chemistry, the natural sciences from the social sciences. In essence, science is one. It is none other than the truth."59 What seems to have happened instead, though, is a little different. Call it the Great Yinification. Since the 1960s, as we have seen, many of the most exciting developments in applied mathematics have been in areas such as nonlinear dynamics, network theory, and complexity. They are about systems that are connected, in flux, and resistant to reductionist logic. Rather than providing a single unified theory, these methods see models as imperfect patches; instead of elegant proofs, or reductive formulae, they offer only fuzzy glimpses of the complex reality. Science, I believe, is becoming more open, and less dogmatic. A degree of humility is even sometimes present. When sceptics say that science can offer little to economics-a view that has become popular since the crisis, as a counter-reaction to the antiquated pseudo-science of mainstream economics - they seem as unaware as neocla.s.sical economists that science is moving on.60 As economists revise their field to reflect these developments, for example through the use of agent-based models discussed in previous chapters, or by a s.h.i.+ft from overly abstract and theoretical treatments towards empirical knowledge, the field may lose its aura of h.o.m.ogeneity and stereotypical maleness and attract a broader range of talent. "Because models by their nature represent only a partial viewpoint, partiality or bias cannot be eliminated from theories," wrote economist Paula England. "A greater openness ... is likely to lead to a multiplicity of perspectives that more adequately captures the complexity and diversity of economic activities."61 While feminist thought has reshaped areas of study such as literary criticism and law, I still find it surprising how economics seems to have largely bypa.s.sed criticism - what could be less politically correct than rational economic man? Feminist economics is a growing field and has been influential in academic areas, discussed in later chapters, such as the study of unpaid work, alternatives to GDP, the role of women in economic development, and ecological economics. However, it has so far had little impact on the standard textbooks. As the professor of gender and economic development Lourdes Beneria notes: "Economics is a very hegemonic discipline, even though there are so many heterodox economists that protest this arrogance and this unwillingness to discuss criticisms. Compared to other social sciences that have integrated gender much more easily, conventional economics has been one of the most impenetrable disciplines. It has been difficult, if not impossible, for orthodox economics to incorporate feminist issues."62 Part of the problem, according to Beneria, is that "to deal with gender relations you have to incorporate power into the a.n.a.lysis. Neocla.s.sical economics does not deal with power relations; it tends to focus on purely economic issues."

If finance is the ultimate, almost caricaturised example of the yang economy, then the yin equivalent is the "three Cs" - cooking, cleaning, and caring - which are still dominated by women, especially in developing countries. These activities aren't just underpaid, they are often not paid at all. According to Statistics Canada, such unpaid work is estimated to contribute around $11 trillion to global GDP, yet it doesn't even register as part of the economy.63 While the yin economy lacks financial power - it's less of a bonus culture than a tip culture - and isn't often in the news, that doesn't mean it is without strength and influence of a different kind. In place of capital, it generates social capital, which sociologists define as the collective value embedded in social networks.64 One nice property is that it is less neurotic and fit-p.r.o.ne, and more resistant to crashes, than its yang counterpart. People don't give up on caring for each other as easily as they give up on the stockmarket; when there's a real earthquake, nurses don't run out of the hospitals crying "It's a ma.s.sive earthquake," a la Lehman Brothers.

The most exciting and productive financial innovations to have emerged in recent years are not collateralised debt obligations or credit default swaps, but the range of schemes that provide credit and finance to the world's poor. The pioneer in this area was Muhammad Yunus, whose Bangladesh-based Grameen bank now directs billions of dollars in small loans to entrepreneurs. Most loans are targeted at women because they are more likely to spend it on their families (and pay it back). The bank sets up "solidarity groups" whose members act as co-guarantors. In Kenya, M-Pesa provides a basic banking service, and access to micro-finance loans, by using mobile phone credits as a currency. Peer-to-peer lending, also known as social lending, does the opposite of CDOs - instead of bundling loans together into an anonymous package, it uses the internet to match individual lenders with individual borrowers. Examples are Zopa in the UK and Kiva Microfunds in the US, which places its loans through micro-finance inst.i.tutions. All these schemes leverage social networks and new technology to provide genuinely useful services to people in need.

To build a more balanced and just economy, we first need to rebalance our priorities, and our mental theories and mythologies. As the physician/gambler Cardano wrote: "A man is nothing but his mind; if that be out of order, all's amiss, and if that be well, the rest is at ease."65 In the next chapter, we consider how this might apply to the greatest balancing act of all - that of social inequality.

CHAPTER 7.

THE UNFAIR ECONOMY.

It's not that many young people do not have aspirations. It is that they are blocked ... Such elitism is unjust socially. And it can no longer work economically.

Alan Milburn MP (2009).

History is a graveyard of aristocracies.

Vilfredo Pareto (1916).

Economists are taught that a well-run market economy is fundamentally fair, so our chance of success depends only on merit. The whole point of a compet.i.tive market, after all, is that everyone has an equal shot. This belief in an underlying equality influences everything from taxation policy to the pay packages of CEOs. Yet in recent decades, the income distribution has become increasingly skewed, with most of the benefits of increased productivity accruing to the top few per cent of the population. When US taxpayers found out about the executive pay packages at government-supported insurer AIG, it nearly led to a populist uprising. The income disparities on a global scale are even more staggering. The reason, as this chapter shows, is that markets are not fair and balanced, and the rich really do get richer. To counter this trend we need a new approach to financial compensation.

The physicist Richard Feynman once said that the greatest of scientific facts was that all things are made of atoms. If there is a close second, many scientists would choose the idea that the universe is based on symmetry. Just as matter reduces to atoms, so physical laws can be reduced to a.s.sertions of symmetry. And the discovery of deep symmetries has motivated scientists since the time of the Pythagoreans.

The ancient Greeks believed that the celestial bodies moved around the earth in perfect circles, because those were the most symmetrical shapes - as Ptolemy put it, they alone were "strangers to disparities and disorders." Newton's laws of motion showed that every force creates an equal and opposite force; Maxwell's equations revealed a symmetry between electricity and magnetism.

The deepest theorems of physics are conservation theorems stating that some quant.i.ty remains stable and unchanged. These too are based on symmetry. Conservation of energy - which says that energy can be neither created nor destroyed in a process, but can only change its form - can be shown to reduce to a statement that the laws of physics are symmetrical in time: if an experiment is done at noon, and then repeated under identical conditions after lunch, the result will be the same. Conservation of momentum is equivalent to saying that the laws of physics do not depend on position: if the experimental apparatus is s.h.i.+fted a few feet to the side, the result will again be unchanged.

Symmetry and reductionism are two sides of the same coin, for only by exploiting deep symmetries can scientists reduce complex phenomena to simple equations. The search for new forms of symmetry still plays a guiding role in physics, as can be seen, for example, by the development of the theory known as super-symmetry, which hypothesises that every particle type has a kind of unseen mirror image. Many physicists even believe that, when the universe was born, at the moment of the Big Bang, all forces and all forms of matter were one, in a state of perfect symmetry. As the universe expanded and cooled, these symmetries gradually broke down in a process called symmetry breaking - the force of gravity separated from the force of electromagnetism, electrons separated from protons, and so on. The messy, asymmetric world we live in gradually took shape.

If physicists seem fascinated by symmetry, then those physics groupies known as mainstream economists can't be far behind. In fact, they take the study of symmetry to a more advanced level. Their theories don't a.s.sume that the economy was once in a state of symmetry - they a.s.sume that it still is. This belief is the source of perhaps the greatest economic myth of all - the idea that the economy is inherently fair and balanced.

Mirror image.

When economics began to be mathematicised in the late 19th century, economists were forced by the limitations of their mathematical and computational tools to simulate "perfect markets" that included a high degree of symmetry. For example, the a.s.sumption of rational behaviour is a kind of symmetry, because it says that everyone, given the same preferences, would act in exactly the same way. When perfectly rational people (if they exist) look at one another, it's like looking in the mirror.

The markets were also a.s.sumed to be in equilibrium, which is symmetry in time: the past looks exactly like the future. And these perfect markets were fair and transparent, so individuals and firms were positioned symmetrically in terms of advantages: they all competed on an equal basis and had access to all necessary information, and no individual firm or person was powerful enough to affect prices on their own. The "law of supply and demand," for example, a.s.sumed a large number of essentially identical firms all competing as equals in the same market. Statistical methods, based on statistical mechanics, could therefore be applied.

While these symmetry a.s.sumptions may originally have been essential for computational reasons, they have proved surprisingly resilient over the years. William Stanley Jevons defined the market as: "persons dealing in two or more commodities, whose stocks of those commodities and intentions of exchanging are known to all ... Every individual must be considered as exchanging from a pure regard to his own requirements or private interests, and there must be perfectly free compet.i.tion."1 In the 1960s, Eugene Fama defined his efficient market as "a market where there are large numbers of rational profit maximizers actively competing, with each trying to predict future market values of individual securities, and where important current information is almost freely available to all partic.i.p.ants."2 About the only change, then, was the insertion of the word "almost."

Economic models in general have continued to shy away from distinguis.h.i.+ng economic agents based on power, influence, access to information, connections, gender, race, cla.s.s, or any other characteristic. Milton Friedman even argued that properly functioning free markets would automatically render such differences irrelevant: "There is an economic incentive in a free market to separate economic efficiency from other characteristics of an individual. A businessman or an entrepreneur who expresses preferences in his business activities that are not related to productive efficiency is at a disadvantage compared to other individuals who do not. Such an individual is in effect imposing higher costs upon himself than are other individuals who do not have such preferences. Hence, in a free market they will tend to drive him out."3 According to theory, s.e.xism, racism or any other form of discrimination is inefficient, so in a pure (i.e. symmetrical) market it wouldn't exist. Economic transactions are more or less the same, regardless of who is involved or when they take place.

Of course, no economist would claim that the real economy is perfectly fair or stable, or that each partic.i.p.ant has access to exactly the same information. In 2001, the Sveriges Riksbank Prize in Economic Sciences was awarded to George Akerlof, Michael Spence, and Joseph Stiglitz "for their a.n.a.lyses of markets with asymmetric information" - in which, for example, sellers of used cars know more about the state of the items being sold than the buyers. One reason why mainstream economics has endured so long is that economists are willing to address what they consider to be isolated flaws in their models, and acknowledge phenomena such as "bounded rationality" or "asymmetric information," while leaving their core theories, teachings, and myths essentially unchanged. The truth is that a.s.sumptions of symmetry are implicit in theories such as the efficient market hypothesis, and the fascination with simplified, abstract, reductionist models explains what economists M. Neil Browne and J. Kevin Quinn describe as the "almost complete absence of power from the toolkit employed by mainstream economists." In a survey they performed of sixteen currently used introductory textbooks - major doorstoppers all - they found zero pages that dealt with topics directly related to power (and a total of only 25 pages dealing with issues related to women).4 The trend towards increased deregulation is also based on the picture of free and fair compet.i.tion between equals.

As seen with the subprime crisis, though, these a.s.sumptions soon begin to look ridiculous when you compare them with the real world. Markets aren't just slightly asymmetric, they're totally out of whack. Is it really OK to a.s.sume that Goldman Sachs and subprime mortgage holders are competing on a level playing field and have access to the same information? Is Wal-Mart versus the local cornerstore really a fair fight? And does it really make no difference where you are born, who your parents are, what schools you went to, who your friends are, or what your history is?

Circulation of the elites.

The French statesman Georges Clemenceau is attributed with the saying that "Any man who is not a socialist at age 20 has no heart. Any man who is still a socialist at age 40 has no head."

Following a similar kind of trajectory, perhaps, neocla.s.sical economics started off in an idealistic vein. The aim of people like Jevons, Walras, and Pareto was to put economics on a rational basis, and thus improve the living standards of the general population. Jevons was brought up in a Unitarian tradition concerned with social conditions, and spent much of his free time walking the streets of the cities he lived in - Sydney, Manchester, London - observing the conditions of the poor and contemplating the connections between poverty and economics. Walras inherited his socialist ideals from his father, and spent a number of years working in the cooperative movement before taking up his professors.h.i.+p at Lausanne.

As a young man, Vilfredo Pareto was a dedicated democrat, and took pleasure in attacking the Italian government for corruption and corporatism. After the May 1898 riots in Milan, which were organised by the Italian Socialist Party and resulted in the deaths of hundreds of people, Pareto offered his home in Switzerland to socialist exiles and leftist radicals. Even by 1891, though, when Pareto was 43, it appeared that his head was pulling in another direction. He wrote to Walras: "I give up the combat in defense of [liberal] economic theories in Italy. My friends and I get nowhere and lose our time; this time is much more fruitfully devoted to scientific study."5 He began to believe that his youthful pa.s.sion for leftist ideals had been based on emotion rather than logic, and that all human societies were inherently corrupt and irrational.

Pareto's cynicism about human motivations was no doubt fuelled in 1901 when he returned home from a trip to find that his wife had run off with the cook and 30 cases of possessions. Under Italian law, Pareto couldn't get a divorce. He had inherited a large sum of money from an uncle in 1898, enough to make him financially independent. In 1907 he resigned his university position and retired to his villa near Lake Geneva, where he lived with a woman 30 years his junior called Jeanne Regis, a large stock of the finest wines and liqueurs, and eighteen Angora cats (the house was called Villa Angora).

Pareto continued to blast off incendiary books, articles, and letters, but his aim switched from trying to change society, to a.n.a.lysing it from his detached vantage point - rather as an entomologist might a.n.a.lyse the social goings-on of an ant hill, but with more spite and irony. In his million-word tome Treatise on General Sociology, he argued that human behaviour is driven by irrational desires, which are then justified by particular ideologies. To understand society, one therefore had to focus on the underlying irrational desires, which he cla.s.sified into six types. The most important were innovation (Cla.s.s I) and conservation (Cla.s.s II). Everyone was motivated by a mix of these cla.s.ses, but one could nevertheless speak of "Cla.s.s I" types, who are clever and calculating, and "Cla.s.s II" types, who are slower, more bureaucratic, and dependent on force.

Pareto had earlier discovered the power-law distribution of wealth (the 80-20 rule) in Italy and other countries, and wrote that it "can be compared in some respects to Kepler's law in astronomy; we still lack a theory that may make this law of distribution rational in the way in which the theory of universal gravitation has made Kepler's law rational."6 Today, we would describe it as an emergent property of the economy. In his retirement, Pareto came to see this highly-skewed power law as a kind of snapshot that revealed the underlying dynamics of any society.

At the top is a small elite consisting of a mix of Cla.s.s I and Cla.s.s II people who are engaged in a Machiavellian struggle for power. There is always a degree of social mobility, so the composition of the elite changes as people enter or leave. The balance between the two cla.s.ses therefore varies with time, in a process Pareto called the circulation of the elite. If too many innovative and intelligent Cla.s.s I people (Machiavelli's foxes) get in power, then the conservative Cla.s.s IIs will plot a takeover. If the elite is dominated by Cla.s.s IIs (Machiavelli's lions), then it will become overly bureaucratic and reactive and the Cla.s.s Is will make their move. This process can be smooth and gradual; but, if the circulation becomes blocked, so that "simultaneously the upper strata are full of decadent elements and the lower strata are full of elite elements," then the social state "becomes highly unstable and a violent revolution is imminent."7 Pareto demonstrated his argument with numerous case-studies. Perhaps the best ill.u.s.tration, though, was the coming to power in Italy of Mussolini's fascist government. Mussolini liked the idea of powerful lions taking over from foxes grown corrupt and ineffectual, and appointed Pareto Senator of the Kingdom of Italy. In 1923, he finally managed to obtain a divorce and marry Jeanne Regis, before dying the same year.

How to get rich.

While Pareto's sociological arguments have dated a bit in the last hundred years, his observation that wealth is distributed according to a power law has remained accurate - except that the elite has grown relatively smaller and more powerful. Figure 14 is a summary of how the world's wealth was distributed among the total 3.7 billion adults in the year 2000, according to a United Nations report. Adults required a relatively modest net worth of $2,138 to count themselves in the wealthiest 50 per cent. To be in the top 10 per cent (370 million adults) they needed $61,000. This group owned over 80 per cent of the total wealth. Anyone with $510,000 was in the top 1 per cent (that's 37 million adults). Together, this small sliver of the world population controlled 40 per cent of the world's financial a.s.sets. Contrast that with the bottom half, who collectively controlled about 1 per cent of the wealth. Someone born into the world at random would stand a 50 per cent chance of ending up in that group of 1.85 billion adults.

Figure 14. Bar graph showing the worldwide wealth distribution in 2000.8 The top decile (10 per cent) controls over 80 per cent of the total wealth. Deciles 6 through 10, which represent the bottom 50 per cent of the population, control about 1 per cent in total.

Rather impressively, the power-law distribution of wealth extends all the way up to the world's richest billionaires. In 2009 the world's richest person was Bill Gates, with a net worth of $40 billion. To put that in perspective, suppose that you made a plot of the wealth of everyone on the planet, in order from richest to poorest. If you continued the plot up to the 99th percentile, then the vertical scale of the graph would have to be around half a million dollars (this will have changed slightly since 2000). But if you wanted to contain Bill Gates, or his friend Warren Buffett, then the vertical scale would need to expand by a factor of about 80,000 (see Figure 16 below). Anyone who saw Al Gore's film An Inconvenient Truth, in which he demonstrated the upward trend of global warming, can picture him riding his cherry picker out beyond the furthest reaches of the atmosphere.

Wealth is also of course not distributed evenly in geographical terms. In 2000 the USA and Canada together had 34 per cent of the wealth, Europe had 30 per cent, rich Asian-Pacific countries had 24 per cent, and the rest of the world including Latin America and Africa held 12 per cent. This mix is changing as countries like China, India, and Brazil continue to experience explosive growth and claim a larger share of the world's economic pie.

From these data alone, one can therefore conclude that the world economy is highly asymmetric. A small number of people enjoy a huge proportion of the world's wealth, while billions live in poverty. The same kind of pattern is seen repeating itself fractally over different scales. Every city has its own local elite, as does every country or region. The sprawling metropolis of greater So Paulo, Brazil, for example, now has some 500 helicopters, more than any other city in the world. The rich find them a good way to avoid traffic jams that can extend for over a hundred miles.9 Also they're hard to steal.

Apart from his discovery of the power-law wealth distribution, another aspect of Pareto's work to have pa.s.sed the test of time was his insistence that humans act primarily on the basis of psychological motivations, and justify those actions on the basis of ideology. The ruling elite always has a very good argument as to why it should be in charge and have most of the wealth and be flying the helicopter. Today, that argument goes by names such as the invisible hand, the efficient market, or mainstream economics.

Broken symmetry.

Adam Smith's concept of the invisible hand is usually taken to refer to the price mechanism. However, his first use of the expression, in his 1759 work The Theory of Moral Sentiments, is on the subject of wealth distribution: "The rich ... divide with the poor the produce of all their improvements. They are led by an invisible hand to make nearly the same distribution of the necessaries of life which would have been made, had the earth been divided into equal proportions among all its inhabitants, and thus without intending it, without knowing it, advance the interest of the society, and afford means to the multiplication of the species."10 The invisible hand refers here not to the magic of the market, but to an early version of trickle-down economics.

Since the economy has patently failed to align itself with this happy picture, at the level of individual countries or the entire globe, one might ask what forces have created such a skewed distribution. According to Smith's later work The Wealth of Nations (1776), free markets tend to drive prices towards "The natural price, or the price of free compet.i.tion." That applies to the price of labour, so it follows that an individual's earnings should reflect the person's inherent value to society. Efficient market theory similarly argues that markets allocate resources efficiently, and that includes wages. If Quetelet's picture of the "average man" is correct, and our abilities are randomly distributed according to a normal distribution, then one might expect wealth to be symmetrically distributed in the same way - most people would be in the middle, and there would be only a few who are very poor or very rich. The reality in most countries is obviously very different, so either our financial elites are incredibly talented, or something else is going on.

As discussed in Chapter 4, one prevailing economic myth is that the economy is inherently stable and at equilibrium - i.e., it is symmetrical in time - and so history doesn't matter. However, there is the old saying that "the rich get richer," and it certainly seems that to make a lot of money, it helps to have some in the first place.

Imagine as a thought experiment that a city-sized group of people are given a windfall of $100 each, under the condition that they must keep it invested in a rather volatile and unproductive stockmarket. Each person makes their own investments, with an average real return of 0 per cent and a standard deviation of 5 per cent.

After one year, most people's nest eggs will be in the range of $90 to $110, and will be distributed according to the bell curve with a peak at 100 and a standard deviation of 5. As time goes on, though, the distribution becomes increasingly skewed. If we follow the worth of the investments as they are pa.s.sed down through generations for 150 years (about the age of economics), then the resulting wealth distribution looks like Figure 15, which is quite similar to the actual wealth distribution in Figure 14.

Figure 15. Bar graph showing the result from the computer simulation of the evolving wealth distribution described in the text. The skewed distribution is an example of symmetry breaking.

Obviously this is not a serious model of how wealth changes with time. It only tracks the value of imaginary investment portfolios, and ignores other kinds of economic transactions (more realistic agent-based models can be constructed, if desired). However, it does demonstrate the simple fact that, left to their own devices, investments will tend to concentrate themselves in fewer and fewer hands. To use the physics term, it is an example of symmetry breaking. At the start of the simulation, everything is perfectly symmetrical. Each person has exactly the same initial amount of money. They also have identical chances of success with their investments - no one is a.s.sumed to be more talented at picking stocks. But over a period of time, some start to pull ahead of the pack. The reason is that there is a positive feedback effect at work. A person whose sum has grown already from the initial $100 to $1,000 can hope to make another $100 in the coming year. They might instead lose that much, but at least they have the opportunity. Someone whose savings fund has shrunk to $10 can only hope to make another dollar.

As the simulation is run for more years, the wealth becomes increasingly concentrated, until eventually only a few people are left gambling with the entire wealth of the community. If a person were born at random into such a population, their chance of being in the elite would be negligibly small. So even though the laws that govern the economy are symmetrical and non-discriminatory, the system tends to evolve towards an increasingly skewed state. Time matters.

Indeed, while we think of the current high degree of inequality as being a permanent feature of the human condition, it is actually relatively recent. For over 90 per cent of our existence, up until the development of agriculture, humans lived in highly egalitarian societies. For reasons discussed further below, the last few decades have seen a particularly large increase in inequality. Pareto's power law was known as the 80-20 rule because he estimated that 20 per cent of the population owned 80 per cent of the wealth. In 2000, the share of wealth owned by the top 20 per cent of the world population had grown to 93.9 per cent. The world is probably more unequal now than at any previous time in history. It is therefore impossible to a.s.sume, as mainstream economic theories do, that we are all on a uniform playing field, or that history can be ignored. It isn't just symmetries that have broken; it is an entire economic worldview.

CEO-nomics.

While investment growth is a major factor in determining wealth distribution, at least for those with investments, an equally important determinant is employment income. Here again we see marked differences within societies, between countries, and over time.

A typical measure of income inequality is the ratio of CEO compensation to the average worker's pay. This is currently highest in the United States, with the UK coming second, continental Europe trailing, and j.a.pan far behind. In the 1930s, the ratio in the US was around 80:1. It declined to around 30:1 in the 1960s, partly as a result of increased regulations such as anti-trust laws and the growth of unions. The trend was reversed in the 1980s, with a relaxed regulatory environment under Reagan and the loss of union power with increased global compet.i.tion. By 2001 the ratio had ballooned to about 350:1, and in 2007 it had reached 500:1.11 The 2007 compensation of Wal-Mart CEO Lee Scott, Jr. amounted to $31.2 million, well over 1,000 times what the average Wal-Mart employee could expect to pull in. In 2008, not a great year for the world economy, Blackstone Group CEO Stephen Schwarzman's total compensation was $702 million.12 In 2009, one hedge fund manager pulled in $2.5 billion.13 While CEO pay in the US is headed for the stratosphere, median salaries have stagnated since the early 1980s. Men in their thirties earn 12 per cent less, in inflation-adjusted terms, than their fathers did at the same age.14 Middle-income households work more hours; often rely on two incomes instead of one; hold far more debt; and have access in their jobs to powerful productivity-enhancing technology. But the benefits of this increased work and productivity have flowed upwards from workers to managers. As a result, says Harvard law professor Elizabeth Warren, "the middle cla.s.s is under terrific a.s.sault."15 In contrast, the share of the top 1 per cent has more than doubled, and is now greater than that of the bottom 90 per cent. This represents a huge transfer of wealth.

According to the law of supply and demand, the price of a CEO should reflect his or her intrinsic value. As efficient market theory's Eugene Fama said in a 2007 interview: "you're just looking at market wages. They may be big numbers; that's not saying they're too high."16 Studies have shown, though, that the success of a company is best seen as the emergent result of factors such as the state of the market, the contributions of all employees, the internal company culture, and so on. Having a good CEO is an important part of the mix, but far less critical than their pay packages would suggest.17 A summary of the empirical evidence from the International Labour Organisation concluded that CEO compensation has "only very moderate, if any, effects ... Moreover, large country variations exist, with some countries displaying virtually no relation between performance-pay and company profits."18 Indeed, CEO pay in j.a.pan is less than half the levels in the US, but they appear to have some effective companies.

A better explanation for the huge rise in CEO pay is provided by behavioural psychology. For a company in trouble, hiring a new CEO is like a patient looking for a wonder drug. Studies have shown that painkillers are more effective if they are sold in an expensive package than if they are presented as cheap generics.19 Similarly, the cachet of luxury goods is enhanced by an exorbitant price tag. So for CEOs, the "invisible hand" of the market is a helping hand up, lifting them higher and higher. The more they charge, the better they look, and the healthier the company's shareholders feel.

CEO pay in recent years was also probably driven, somewhat perversely, by media reports on executive pay. "Rather than suppressing the executive perks," notes the behavioural psychologist Dan Ariely, "the publicity had CEOs in America comparing their pay with that of everyone else."20 The result, with the help of recruitment consultants, was that the bar kept ratcheting up even higher. CEOs have come to resemble celebrities, like film stars or athletes, whose fame and appeal seem only to increase the more they get paid.

CEOs also benefit from their position as hubs in the business network. Positive feedback network effects mean that the more network connections they make with other business leaders, the more powerful they become, thus improving their position in the network, and so on.21 This positional advantage might make them more effective in their jobs, if they aren't out socialising all the time, but it is primarily a property of the network itself, rather than the individual - the whole rather than the atomistic part. If a CEO retires, the network will produce a replacement or rearrange itself into a new configuration. Conventional economics sees the economy as flat and level as an Illinois cornfield, while in fact the terrain is more like the mountains of Afghanistan. Position matters: the better it is, the easier it is to defend.

Of course, these behavioural psychology and network effects don't occur in quite the same way for, say, greeters at Wal-Mart (average salary, around $10 per hour) or those who actually make the items in third-world sweatshops (tens of cents an hour).22 Low-level employees find it harder to capitalise on luxury cachet, media glamour, or powerful connections. For a typical business, payroll makes up about 70 per cent of their costs. There is therefore tremendous pressure, both from customers and from shareholders, to reduce employee pay. Even if executives are vastly overpriced, their compensation is still a relatively small proportion of the total wage bill, and of course they are the ones deciding how pay should be divided. The pressure is therefore transferred to workers at the bottom, for whom the invisible hand is not a helping friend, but a hand on their ankle pulling them down.

Corporate power.

Another reason cited for the increasing pay of CEOs is that the companies they govern have grown larger and more complex as a result of globalisation. (Oddly, the US military has also grown more complex, but generals still sc.r.a.pe by on about ten times a private's wages. The US government is awfully complicated, but President Obama makes ends meet on $400,000.) It is certainly true that, just as wages in the economy have grown increasingly asymmetric, so have the sizes of companies. These again follow a roughly power-law distribution. At one end of the scale are millions of tiny firms employing only a few people. Looming over them are a much smaller number of huge multinationals, the size of nation-states. At the time of writing, the largest company in the world in terms of revenue is the oil company ExxonMobil, with annual revenues of $390.3 billion. The second is Wal-Mart, at $374.5 billion. Out of the top ten, seven are in the oil and gas business.

The lower panel of Figure 16 shows the 100 largest companies as measured by annual revenue. It is interesting to compare the two plots with Figure 13, which showed the 100 largest price changes in the S&P 500. The similarity is a reminder that power-law distributions, and fractal scaling, apply not just to financial shocks but also to financial opportunities. Just as we don't know when the next crash will occur, so it is impossible to know from the outset which micro-company will grow into a global behemoth, or who will be the next Bill Gates or Warren Buffett. The business model of venture capitalists is to fund an ensemble of small firms in the hope that one or more will hit it big.

The large multinational companies, such as Wal-Mart, rely on global supply chains that source the cheapest suppliers from around the world. This has the beneficial effect of minimising costs for consumers and maximising profits for shareholders. They can exploit economies of scale, and can also invest large amounts of money in improving their products. The pharmaceutical industry, for example, is headed by Pfizer, with total annual revenues in 2008 of $71 billion. Its enormous size means that it can afford a research and development budget of $11 billion.

While there are certainly benefits to having large companies, there are also severe disadvantages. They fuel inequality in the larger economy by increasing the downward pressure on wages for ordinary workers, at the same time as executive compensation balloons. They influence consumers through enormous advertising campaigns. Companies such as banks that become too big to fail can take enormous risks under the implicit guarantee that the government will bail them out if they get into trouble. The companies can also employ armies of lobbyists to exert pressure on government bodies, for example to relax environmental or labour regulations (ironically, while corporations strongly regulate the behaviour of their employees, they oppose any external government regulation of their own behaviour on the grounds that it is inefficient).

Figure 16. Top panel shows wealth (in billions of dollars) of the 100 richest people in 2009 according to Forbes.23 Bill Gates is first with $40 billion, followed by Warren Buffett at $37 billion. Lower panel shows annual revenue in billions of dollars for the top 100 companies. ExxonMobil is top with $390.3 billion, followed by Wal-Mart at $374.5 billion.24 Both plots follow a similar power-law distribution.

When companies are larger than many of the governments they deal with around the world, there is clearly a problem for democracy. The gulf between the average citizen and corporate power has never been greater, especially in developing countries. It's hard to make progress in addressing global climate change when seven of the world's ten largest companies are in the oil and gas business; hard to reform the world finance system when faced with the political clout of firms like Goldman Sachs and JPMorgan; hard to negotiate labour conditions when the party on the other side of the table is Wal-Mart.

Neocla.s.sical economists like Jevons argued that the market gains its power by aggregating the desires of a large number of people, thus increasing overall utility. However, it seems better at aggregating some things than others. As shown over the last few decades, markets can aggregate our power as consumers to force down prices of basic goods, and our power as investors to maximise profits. This restrains pay at the low end. They are much less good at aggregating our desire for a reasonable living wage, or a clean environment, or basic decency and justice. Now more than ever we need strong democratic inst.i.tutions to temper the power of multinational corporations with their unelected and extravagantly paid CEOs. We also need an economic theory to make sense of it all. Unfortunately, mainstream economics isn't it.

Unequal opportunities.

Part of the enduring appeal of orthodox economic theory to our corporate leaders is that it provides a convenient intellectual argument for the programme of deregulation, privatisation of government a.s.sets, and social cutbacks that has been in place in many countries for the past few decades, and that has benefited large corporations at the expense of average workers. These changes have been justified by the myth that the economic system is inherently fair, so people are paid what they are worth. But just as a.s.suming that the market is stable actually makes it more risky, so the a.s.sumption of fairness ends up making the economy less fair.