The Future of Everything

May 17, 2021

Shifting the Economyths

Filed under: Economics, Quantum, Talks — David @ 3:29 pm

This is the text for my contribution to the online conference Beyond the False Dichotomy: Shifting the Narrative

I wrote Economyths a little over ten years ago, and in part the book was my response to the financial crisis. The thesis was that mainstream economics is based on a number of what I called economyths which were defined as beliefs or stories that have shaped economic thought. There were ten in total but to give you an idea here I will just mention four:

One said that the economy is made up of independent individuals. This is the idea that people are like the classical picture of a self-contained atom, and there is “no such thing as society” as Thatcher said.

Another is the idea that the economy is stable and self-correcting. This is the main story of economics from Adam Smith’s invisible hand, to modern equilibrium models.

A third is the idea that the economy is rational and efficient. The assumption that people make rational decisions to optimise utility is connected to the ideas of stability and efficiency. Corporations are defined to be an incarnation of rational economic man.

And then consistent with these is the next economyth which is that the economy is balanced and symmetrical. The issues of power and distribution are viewed as “soft” and peripheral to the subject, and therefore tend to be ignored. As one economist said in an interview, “economists are not good at what’s fair, right?”

Economists have always insisted that their theories are far more complex than these economyths would suggest, but the reality is that these assumptions have been extremely influential and in particular form the basis of the mathematical models used to simulate the economy and make policy recommendations.

In 2017 I did a revised version of the book which included an additional economyth, which was the idea that the economy boils down to barter. This is the myth which in many ways justifies the others, because it means that money isn’t important. Adam Smith for example focused on the “real” economy of labour and commodities and saw money as a kind of veil or a distraction. Economists since then have treated money as just a metric or an inert medium of exchange, and ignored its confounding properties – its dual real/virtual nature, its ability to entangle people through debt, its inherent instability, its tendency to cluster and create inequality, and its psychoactive effects on the human mind.

The drawbacks of omitting things like money and banks from the model became evident after the financial crisis of 2007, when it turned out – as one central banker explained ten years later – that “In the prevalent macro models, the financial sector was absent, considered to have a remote effect on the real economic activity.” And today, the continuing problems of financial instability, social inequality, and environmental degradation can all be largely traced to the money system, which is unstable, unfair, and is reliant on continuous growth to pay off debt.

One reason I called these ideas economyths is because, like myths, their legacy goes back a long way. We could start with post-war economists like Milton Friedman; or go back further to the neoclassical economists who first tried to establish economics as a kind of social physics in the late nineteenth century; or to Adam Smith, who didn’t try to quantify economics but was inspired by Newton. But I would argue that we can actually go back much further. My reason for saying this is that the ten economyths were based directly on a list of opposites, divided into good and evil, from the philosopher Pythagoras, who believed that the universe was based on number, and who lived at a time when coin money was changing the world of commerce. As he said, “number is all” and money is a way of putting numbers on the world.

In order to understand the narrative appeal of this model, it is interesting to compare it with another model which was very influential for a long time, and also reflected the Pythagorean ideals of symmetry and mathematical elegance, namely the Greek model of the cosmos. This model incorporated two main assumptions. The first was that the celestial bodies moved in circles, which were considered the most perfect and symmetrical of forms. The other assumption was that the circles were centered on the Earth. In Aristotle’s version, the planets and stars were thought to be encased in crystalline spheres which rotated around us at different speeds. The fact that planets did not follow perfect circles around the Earth, but sometimes tended to loop back on themselves, was handled by adding epicycles – circles around circles.

This geocentric model was complemented by a theory of physics. According to Aristotle, all matter consisted of the five elements Earth, Water, Air, Fire, and Ether which was reserved for the heavens. His theory was less a theory of motion, than a theory of stability: each element sought its own level, following the same order with Earth on the bottom and Ether on top, and in fact would do so instantaneously in a vacuum. Aristotle deduced from this that a vacuum could not exist: nature abhors a vacuum.

The Greek model lasted for well over a thousand years – it was adopted by the Church, and was not finally overturned until the Renaissance. How did it manage to last for such a long time? And what lessons are there for economics?

One reason for its durability was that it could make accurate predictions of important events such as eclipses. Another reason, though, is related to aesthetics and the fact that, as Aristotle put it, man is a political animal. There was a strong parallel between the perceived order of the cosmos and the order of society. Greek society was structured as a well-ordered hierarchy, with slaves at the base, followed in ascending order by ex-slaves, foreigners, artisans, and finally the land-owning, non-working upper class. These men alone could be citizens, and oversaw everything from above, like the stars in the firmament (women did not take part in political life and took their social class from their male partner). A model of the universe which suggested that everything has its natural place in a beautiful, geometrically-governed cosmic scheme therefore supported the status quo. For this reason it would certainly have appealed to the male leisure class that ruled ancient Athens, and later to the Catholic Church.

The first cracks in the model appeared in 1543, when Copernicus proposed that the Earth might go around the Sun, rather than vice versa. European astronomers began to observe comets which passed between the planets, so if Aristotle’s crystalline spheres had actually existed, they would have broken through them. Finally, in the late seventeenth century, Isaac Newton derived his three laws of motion and the law of gravity. The static circles of classical geometry were replaced with dynamical equations, which had a different but equally powerful aesthetic appeal. Sometimes it takes a model to defeat a model.

So what does this have to do with economics? Well, there is an obvious parallel between the Greek circle model, and our modern model of the economy, because it too pictures the world as rational, stable, ordered, and efficient, and therefore favours the elite. Indeed mainstream economics, in its obsession with rationality and efficiency, sometimes sounds like the PR wing of the financial sector. The model can’t make predictions of things like crises, or indeed the economy in general, but it goes a step further by predicting that it can’t predict, as per efficient market theory. The main difference is that while Aristotle thought that a vacuum could not exist, because otherwise things would find their equilibrium immediately, efficient market theory assumes that prices do reach equilibrium instantaneously – so a vacuum does exist, and it is the market.

Both the Greek circle model and the economics model also show the strength of a mathematical model. Even something like the financial crisis only made a dent. As Paul Krugman wrote in 2018, “Neither the financial crisis nor the Great Recession that followed required a rethinking of basic ideas.” So like the Greek model, the orthodox model of the economy is incredibly resilient.

I think if there was a specific point where this narrative became compelling in economics, it was when economists changed their theory of value, while leaving the rest of the theory mostly the same. The switch was announced 150 years ago by William Stanley Jevons, in the second paragraph of his 1871 Theory of Political Economy, where he wrote that “Repeated reflection and inquiry have led me to the somewhat novel opinion that value depends entirely upon utility.” Here utility was a kind of energy-like quality which roughly equated to happiness. Classical economists such as Smith had followed a labour theory of value which acknowledged the role of power. Utility created a new narrative which flipped this on its head. Economics was now about pleasure and good times.

Of course utility couldn’t be measured directly. Another approach though was to simply assume that utility is reflected by price. Or as Jevons put it: “just as we measure gravity by its effects in the motion of a pendulum, so we may estimate the equality or inequality of feelings by the decisions of the human mind. The will is our pendulum, and its oscillations are minutely registered in the price lists of the markets.”

A side-effect of this emphasis on subjective utility, ironically, was that by reducing value to a number, subjective things like emotion or social power or dignity or ethics were usurped by the theory and thus stripped of all weight. As Tomáš Sedláček wrote for example in his 2011 book The Economics of Good and Evil, “The issue of good and evil was dominant in classical debates, yet today it is almost heretical to even talk about it.” Another thing which also of course didn’t fit into this rational utility approach – it doesn’t compute – was money, with its economyth-defying properties of entanglement, instability, and irrationality.

But still, why is it that more than a decade after the crisis economics still remains rooted in the past? Why is the orthodox narrative so resistant to change?

One reason again is that as with Aristotelian physics, the economyths form part of a connected structure. The story they tell is consistent and self-reinforcing – markets are efficient, stable, rational – and you can’t change one part without changing them all.

Instead, what happens is that economists attempt to fold in new ideas from other areas such as complexity or behavioural psychology, without changing the structure too much. Behavioural economics for example has in my view won acceptance exactly because it can be incorporated in this way, and viewed as an epicycle that can be wheeled out for particular situations. Economists also try to add in so-called “frictions” to their equilibrium models, but assume the equilibrium exists in the first place. Paul Krugman again: “We start with rational behaviour and market equilibrium as a baseline, and try to get economic dysfunction by tweaking that baseline at the edges.” A few more epicycles and it will be perfect, the thinking goes.

The story also offers a powerful restoration narrative, because it says that crises and upsets are caused by external events, and economic forces bring the economy back to this imagined equilibrium. It therefore taps into the human desire for order which is important in politics but also in fiction. And the idea that markets are rational and efficient also justifies the powerful position of the financial sector and the wealthy, while at the same time distracting from the workings of power and the role of money.

To change the narrative it isn’t enough to modify details of the model, instead we need to go back as the neoclassicals did to the fundamental idea of value and its relation with price – in other words, the question of how much something is worth, its numerical cost. There are a number of ways of going about this, but I would argue that a good place to start is with the math.

This might seem counter-intuitive. After all, it is something of a cliché to say that economics is too mathematical. As one recent book put it, “today mainstream economics follows a path of great mathematical rigor that . . . does not make much room for other accounts of economic life.” However rigor isn’t useful if you are using the wrong kind of mathematics in the first place. And while Keynes wrote that “practical men … are usually the slaves of some defunct economist” we could also say that those economists are themselves slave to beliefs about number and value that are inherited in large part from a mathematical model.

Orthodox theory for example is based on the core idea that markets are stable and price is a measure of inherent value. It is best represented by the X-shaped figure of supply and demand. This plots supply as a line which increases with price, and demand as a line which decreases with price. The point where they match in an X represents the stable equilibrium where the market clears. This diagram appears in all introductory economics textbooks, but it is also there in the mathematical models used to simulate the economy.

An odd feature of the graph is that price appears on the vertical axis, and is assumed to be determined uniquely and passively by unknown forces of supply and demand which are in balance, so cancel out. While neoclassical economics is often described as Newtonian, it assumes equilibrium and has no real concept of mass or dynamics or force. The problem though is we never observe supply or demand independently, we only observe transactions. Like the crystalline spheres, these are imaginary constructs. Indeed the whole idea of representing a complex dynamic system as the stable intersection of two lines is very strange and is not done in other areas such as biology, where the only things that are at equilibrium are dead.

To make the diagram more scientific, a first step is to flip it around so that price is the independent variable on the horizontal axis. This seems a trivial change but is actually key to the whole story. Instead of utility, we can then plot the propensity curves for the buyer and seller, which represent the probability of transactions as a function of price. The fact that price is somewhat arbitrary and decisions are subject to effects such as context means these curves are inherently probabilistic. The X-shaped lines of supply and demand from the classical diagram are therefore replaced by probabilistic waves which only collapse down to a particular price during a transaction. This model therefore acknowledges that value is a soft and fuzzy concept, while price is a measure which is subject to intrinsic uncertainty.

The next step is to acknowledge that people are not separate atoms who only communicate by bouncing off one another, instead they are entangled beings who talk to each other, and share things like social norms. An example is the prisoner’s dilemma game, another staple from the textbooks: in classical theory everyone defects and rats out on the other person, but experimental results show people choose to collaborate between a third and a half of the time, which suggests a high degree of entanglement.

We also need to address the fact that people make decisions based on a mix of objective and subjective factors which are entangled in the mind and may interfere with one another. And above all we need to include the dynamics of money, which behaves more like a kind of information than a classical physical object.

The correct mathematical framework for this theory has already been developed by a group of radical thinkers over a century ago. Unfortunately this mathematics has until recently been reserved for the esoteric area of subatomic particles. I’m talking of course about the quantum formalism. For our purposes this refers, not to quantum physics, but to a kind of logic and probability which allows for effects such as interference, entanglement, and the idea of a measurement procedure. The word quantum is from the Latin for “how much” which applies naturally to the economy, where prices are measured through transactions. The point is not that there is a direct map between subatomic particles and humans, but that we can use similar mathematical tools to model each, which is a subtle but important distinction.

While quantum ideas have been around for a while, they are only now starting to reach critical mass in the social sciences. This new quantum narrative is under construction, in economics and finance but also in other areas such as psychology and political science. For example the Carnegie Foundation is funding a series of quantum bootcamps for social scientists at Ohio University starting this summer. There is a new anthology coming up this year from Oxford University Press on Quantizing International Relations. Danah Zohar has been bringing quantum principles into management theory for some time. The area which is seeing the most rapid adoption of quantum ideas is finance, because of quantum computing. As far as I know you can’t take a university course in quantum finance, but you can get a job in it right now in financial centers such as Paris, New York, Toronto, and so on. The development of classical computers in the post-war era changed the way we model and think about systems including the economy, and quantum computers – which have entanglement built in – are now doing the same thing.

What counts for the purposes of today’s discussion though is not the math, but the story told by the math. To summarise, the core narrative of mainstream economics is that people behave like classical atoms: hard, independent, stable. The economy can therefore be modelled as an equilibrium system. The main message of quantum economics is that people are entangled: with their own subjective feelings, with other people, with what they read in the news, and above all through the money system. The economy is a complex, dynamic, living system which can be modelled using a mix of techniques, such as ones from complexity science or systems dynamics, so long as they respect the indeterministic and entangling nature of both mind and money; and particularly the ability of the money system to scale up cognitive and financial entanglements to the societal level.

So as a one word description of the new narrative I would choose quantum or maybe entanglement. If you don’t want to do quantum mathematics, which is understandable, it doesn’t matter because what counts is the idea that the economy is best seen as a complex system which is entangled through a mix of financial and social effects. In practical terms this means that all the so-called “soft” ideas such as subjectivity, emotion, social dynamics, power, money, value, fairness and ethics that have been exiled from economics are now back in play. Debtors and creditors are entangled, shareholders and stakeholders are entangled, and we are all entangled with the climate system. Obviously the quantum approach doesn’t have all the answers, but its built-in emphasis on uncertainty can be liberating, and encourages a pluralistic response. Perhaps it is the model which teaches us to sometimes at least let go of models, because they can’t capture enough of the complex reality. And my hope is that the quantum approach and the idea of entanglement resonates with some of the other ideas and narratives discussed today.

Again, the idea that a new narrative should begin with our system of logic and probability may seem strange but history shows that mathematical models have great influence. And sometimes, as mentioned, it takes a model to beat a model. A new narrative which marks the next evolutionary step of capitalism is going to need a new mathematical framework, if only to better define its language, and help to do an audit on what ideas one may inherited, perhaps unconsciously, from the classical model. It is ironic that we live in an age characterised by volatility, uncertainty, complexity, and ambiguity but our economic theory assumes a deterministic state of placid equilibrium. It is therefore well overdue for an upgrade.

Further reading: quantum economics resources

May 10, 2020

The quantum coin trick

Filed under: Economics, Quantum, Talks — Tags: — David @ 3:17 pm

Upcoming free talk at the CQF Institute on July 1:

Quantum Economics and Finance: The Quantum Coin Trick

Quantum economics and finance uses quantum mathematics to model phenomena including cognition, financial transactions, and the dynamics of money and credit. This talk takes a particular route into the subject, through a discussion of the quantum coin. Unlike a classical coin toss, which can be either heads or tails, a quantum coin can be – like Schrödinger’s cat – in a superposition of states. This gives it intriguing properties which can be used to simulate everything from the prisoner’s dilemma, to the credit relationship, to the pricing of options. The talk is based on material from the book Quantum Economics and Finance: An Applied Mathematics Introduction.

Slide 1



March 29, 2019

Talk for Quantizing IR panel at ISA 2019

Filed under: Economics, Physics, Quantum, Talks — Tags: — David @ 6:49 pm

This is an edited version of my contribution to the panel discussion “Quantizing IR I: Physicists, Meet Social Theorists!” at the International Studies Association conference in Toronto on March 29. The session was chaired by James Der Derian (University of Sydney) and the other participants were Alexander Wendt (Ohio State University), Shohini Ghose (Wilfrid Laurier University/Perimeter Institute), Kathryn Schaffer (School of the Art Institute of Chicago), Michael Schnabel (University of Chicago), and Genco Guralp (San Diego State University).

One nice thing about quantum is that it looks different to people who come at it from different backgrounds and take different paths. My background is in applied math, and my own interest in applying quantum methods to social questions came about several years ago when I was researching a book on the history of money. And I think money serves as a particularly simple and illustrative example of a quantum social phenomenon, so I will give a quick description of that before getting on to more general points.

The word “quantum” is Latin for “how much” and the money system is a way of answering that question – or “quanto costa” in Italian which makes the quantum connection more clear – which when you think about it is not an obvious thing to do since value is a quality not a quantity. Because of this fundamental incompatibility at its core, and because it is related to the transfer of information rather than of classical objects, the money system turns out to have its own quantum properties including indeterminacy, duality, entanglement, and so on.

The most trivial of these is that money moves discontinuously, in sudden quantum leaps. Schrödinger once said “If we have to go on with these damned quantum jumps, then I’m sorry that I ever got involved” but with money the same thing happens every time you tap your card at a store. There isn’t a little needle that shows the money draining out, it just jumps.

Money is fundamentally dualistic, presenting both as a physical object, like a coin, or a virtual object, like a credit transaction, while still retaining the properties of each. A bitcoin for example is a virtual currency but it exists on a real hard drive.

The money system is indeterminate: the price of something like a home is fundamentally uncertain and is only settled at the time of purchase. Transactions therefore act as a measurement process on value.

Money is entangling: here entanglement means indeterminate but coupled at the same time, an example being the state of a loan between two parties. We can model each participant as being in a quantum superposition of cognitive states. If the borrower defaults that acts as a measurement which collapses the state of the loan, even if the other party only finds out later. Of course this is a simpler version of entanglement than the kind seen in physics (there is only one axis, namely default or no default), but an advantage is that you don’t need sophisticated statistical experiments to tease it out.

So money does not behave classically which is one reason it has traditionally played a small or even non-existent role in economic models, which as many commentators have noted is one reason the financial crisis wasn’t predicted. Conventional models didn’t include a banking sector, let alone the financial entanglements represented by a quadrillion dollars worth of derivatives. The fact that money was left out of the picture seems a remarkable omission given its obvious importance not just to economics, but to everything from marital relations to international relations, but fits with the classical view that money is just an inert medium of exchange.

It was only later, through the work of people like Alexander Wendt, that I connected this with the broader areas of quantum cognition and quantum social science, which of course add many completely new dimensions. One way to think of money is as a kind of prosthetic that extends the dualistic properties of quantum mind, as we mentally collapse value down to price.

I agree with the idea that quantum processes are likely to play a role in human cognition. In itself though I don’t think that this will show or prove that society is quantum or that we are best seen as wave functions, because living systems can’t be reduced to their components. Quantum processes are believed to play a role in avian navigation, but this doesn’t really change the way we think about birds. At a trivial level, we are quantum because the universe is quantum, but what counts is the emergent behaviour. The question from my perspective is whether social systems can be usefully modelled as quantum.

The quantum methodology allows scientists to model things like indeterminacy, interference between incompatible concepts, and entanglement, all of which characterise human relations. Bohr’s theory of complementarity for example was inspired by psychology, and our ability to hold two incompatible ideas in the mind at the same time. This position is a little different from the physicalist argument, because it says that something like money is quantum not because it inherits these properties from subatomic particles, but because it is a quantum system in its own right, and I would argue this holds for other social institutions as well.

In this sense I’m not sure that economics needs a model of consciousness, but the problem is that it already seems to have one, which is that people are lifeless automata. This approach is epitomised by the old and, as economist Julie Nelson points out, rather gendered concept of rational economic man, who makes decisions to optimise his own utility based on preferences that are fixed and known. It would be a great improvement to shift to the connected, fluid, and indeterminate idea of quantum economic person.

It might still sound here that quantum is being used as just a metaphor. But the idea of a metaphor is to explain something complex in terms of something that is concrete and immediate, as in “all the world’s a stage”. A wave function is many things but it is not concrete or immediate. For one thing it involves imaginary numbers, and we have no idea what wave function collapse means, with many different interpretations, which is pretty humbling as Kathryn Schaffer noted.

Instead it makes more sense to go the other way, take human experience at face value, and use it as a metaphor to understand the physical world. One reason we don’t do that obviously is because quantum was applied to physics first.

We also need to distinguish between the system and the model of the system. Quantum models are not reality, even if they appear to give exact results in calculations. Instead they are tools adapted from mathematical techniques.

The original quantum idea came from someone, Max Planck, trying to fit a mathematical model to some strange looking data, for black-body radiation. One of the main tools in quantum mechanics is Hilbert space which is a generalisation of normal Euclidean space. This was developed in the early twentieth century and later used by physicists to formalize quantum mechanics. In general, quantizing a system is a fairly clunky mathematical procedure which converts classical equations into quantum ones.
These mathematical methods were adopted in physics not because anyone liked them but because they worked. Social scientists are permitted to do the same thing (models can be applied to different systems and at different scales).

For example if someone decides to quantize some aspect of the economy, the hope is that the resulting model captures the essence of the underlying system, addresses shortcomings in traditional models, is parsimonious in terms of parameters, and can make useful predictions. Such decisions are the prerogative of the modeller and should be made based on sound principles of mathematical modelling.

Of course there has been a lot of pushback from both physicists and economists to this use of the quantum approach, which is all to the good, but I can address a couple of points that often come up. One common objection is that quantum processes don’t scale up (Bohr’s correspondence principle), so what happens at the micro level doesn’t affect us at the macro level because it all washes out. But quantum processes do scale up, especially through the use of technology. In physics we have the atom bomb (which doesn’t wash out), or a laser pointer for that matter, in society we have the financial system which can be viewed again as a kind of quantum social technology that can be used for good or ill.

Another objection is that quantum social science is the ultimate example of “physics envy” – and there is some danger of that. But as someone pointed at another one of these events, that horse already left the barn. An example is the efficient market hypothesis, which is a central theory of economics and finance. The idea that market prices follow a random walk, and the emphasis on probability, was directly inspired by quantum theory, and was developed in part by the many nuclear physicists who switched to finance after the war. However this was a sanitised version of quantum that picked up on stochasticity but omitted its other features; and the theory was widely misused to justify the financial instruments which played a key role in the crisis.

While there is no shortage of model abuse in economics, physics envy is not the main problem. Instead it is institutional pressures that encourage the use of models that look good based on the aesthetic criteria of mechanistic science but have so many parameters and moving parts that they can give any answer you want. This is what economist Paul Romer called in a paper “the trouble with economics”. And in fact you see exactly the same issue in physics – Romer’s paper was named after Lee Smolin’s book The Trouble With Physics.

All this raises a couple of questions. One is that if something like the money system can be described as quantum, then does that tell us something useful about how we should interpret physics, for example the role of information (no idea). Perhaps more relevant from a sociological perspective at least is, why has it taken a century for these ideas and methods to feed into the social sciences.

Of course there is the worry that quantum ideas can be dangerous nonsense when applied outside physics. As physicist Sean Carroll wrote in 2016, “No theory in the history of science has been more misused and abused by cranks and charlatans”. However I would argue that the theory misused to the greatest effect in social science is the idea that we are like classical machines: inert automata, slave to the mechanisms of cause and effect. This has done far more damage than anything like “quantum healing”. Why is there so much controversy about social scientists possibly abusing quantum models, when they have been abusing mechanistic models for years?

One reason is that quantum ideas represented a challenge to our traditional scientific world view, and in some ways we still haven’t got used to them. Einstein for example said quantum reality reminded him of “the system of delusions of an exceedingly intelligent paranoiac, concocted of incoherent elements of thoughts”. Yet these concepts such as duality, indeterminacy, and entanglement seem quite reasonable when applied to our own thought patterns. And I think Einstein’s comments would apply quite well to the social world of finance.

As the Torontonian Marshall McLuhan wrote in 1992: “I do not think that philosophers in general have yet come to terms with this declaration from quantum physics: the days of the Universe as Mechanism are over”. So it is exciting that 25 years later that is starting to change.

April 27, 2015

Money is the Message – 2015 McLuhan Lecture

Filed under: Economics, Talks — Tags: , — David @ 3:25 pm

McLuhan lecture video, from Transmediale in Berlin

December 15, 2014

transmediale Marshall McLuhan Lecture

Filed under: Talks — David @ 11:44 am

Looking forward to giving the Marshall McLuhan Lecture at the transmediale festival in Berlin next month. The talk will be called “Money is the Medium” and will follow up on McLuhan’s description of money as a social medium.

June 13, 2012

Climate Change and its Impact on Preservation Management of Archaeological Sites, Athens

Filed under: Talks — Tags: , — David @ 1:40 pm

Talk on climate change and cultural heritage, Athens

Talk at roundtable on climate change and cultural heritage in Athens, April 2012.  Watch video

Breakthrough economics

Filed under: Talks — Tags: , — David @ 10:28 am

This is an article based on my talk at the Breakthrough Capitalism Forum in London, May 29, 2012. Versions available in Polish and Spanish.


Economics can be viewed as a mathematical model of the world. Such models are interesting because they encode a kind of story about reality. The models – and especially the assumptions on which they are based – therefore reveal much about the way we see the world and we see ourselves.

An early example was the Greek model of the cosmos. This model incorporated two main assumptions. The first was that the celestial bodies moved in circles, which were considered the most perfect and symmetrical of forms. The other assumption was that the circles were centered on the Earth. In Aristotle’s scheme, the planets and stars were thought to be encased in crystalline spheres which rotated around us at different speeds.

Of course, as we now know, both of these assumptions were wrong. But that didn’t seem to matter, for the model persisted for well over a thousand years – it was adopted by the Church, and was not finally overturned until the Renaissance. How did it manage to last for such a long time?

The main reason for the success of the Greek model of the cosmos was that it could make accurate predictions of important events such as eclipses. In a time when human affairs were believed to be influenced by the movements of the celestial bodies, this was an impressive demonstration of the power of mathematics, and gave a sense of order and stability to the universe (our desire to predict the future is coupled with a yearning for a comprehensible world).

Another reason, though, is related to aesthetics and the fact that, as Aristotle put it, man is a political animal. There was a strong parallel between the perceived order of the cosmos and the order of society. As with the model of the cosmos, Greek society was structured as a series of concentric rings, with slaves at the base, followed in ascending order by ex-slaves, foreigners, artisans, and finally the land-owning, non-working upper class. These men alone could be citizens, and oversaw everything from above, like the stars in the firmament (women did not take part in political life and took their social class from their male partner).

A model of the universe which suggested that everything has its natural place in a beautiful, geometrically-governed cosmic scheme therefore supported the status quo. For this reason it would certainly have appealed to the male leisure class that ruled ancient Athens, and later to the Catholic Church.

The first cracks in the model appeared in 1543, when Copernicus proposed that the Earth might go around the Sun, rather than vice versa. In 1577, the Danish astronomer Tycho Brahe observed a comet which passed between the planets, so if Aristotle’s crystalline spheres had actually existed, it would have broken through them. Finally, in 1687, Isaac Newton combined Kepler’s theory of planetary motion, with Galileo’s study of the motion of falling objects, to derive his three laws of motion and the law of gravity. The static circles of classical geometry were replaced with dynamical equations, which had a different but equally powerful aesthetic appeal.

Newton’s work laid out a template which scientists have continued to follow until the present day. To understand and predict a system, you break it down into its constituent parts, find the mathematical equations that govern their behaviour, and solve.

This mechanistic approach has been very successful in areas such as chemistry and physics. Newton was less confident that it would apply to other fields. After losing most of his fortune in the collapse of the South Sea bubble, he warned that ‘I can calculate the motions of heavenly bodies, but not the madness of people’. In the nineteenth century, though, economists decided to forge ahead anyway. The result, known as neoclassical economics, was directly inspired by Newton’s ‘rational mechanics’. As a mathematical model, it may have had as great an effect on society as the Greek model did in centuries past.

Efficient markets

In order to mathematicise the economy, economists of course had to make many simplifying assumptions, which imposed a kind of symmetry on the system. No one thinks these assumptions are completely true, but they have certainly been extraordinarily influential, and have shaped the way we see the economy – and therefore the economy itself.

The theory assumed that individuals and firms, who were the atoms of the economy, acted independently, rationally, and with similar power to maximize their own utility. This led to the famous caricature of rational economic man, or homo economicus. The ‘invisible hand’ (usually attributed to Adam Smith, though the idea predates him) then drives the economy to a stable equilibrium. The result of all this was supposed to be societal happiness, or what economist Francis Edgeworth called in 1881 ‘the maximum energy of pleasure, the Divine love of the universe’. As with the Aristolian scheme, everything is driven towards its correct place in the natural order.

In the same way that geocentric models of the cosmos were used for forecasting the positions of the celestial bodies, the aim of economic models was to make accurate predictions. However in this case the models did not seem to work so well. Predictions of quantities such as oil prices or gross domestic product are famously unreliable, and even today are not much better than random guessing.

The problem was apparently solved in the 1960s when the ‘efficient market hypothesis’ was floated as an explanation for the inaccuracy of forecasts. This was a physics-inspired theory which assumed that markets magically attain a stable and optimal equilibrium, and any changes are random perturbations that inherently cannot be predicted. However it should be possible to calculate risk based on statistical methods such as the normal distribution (bell curve), or variants thererof. The normal distribution is mathematically similar to a square (the word ‘normal’ is from the Latin for square) or circle in that its properties can be specified by a single number. The risk of an asset could be measured by the standard deviation of its past fluctuations, just as the size of a circle is given by the diameter.

This elegant theory, with its emphasis on stability, symmetry, and rationality, was the backbone of what Alan Greenspan described to the US Congress as ‘the whole intellectual edifice’ forged by Nobel laureate economists that ‘collapsed’ during the recent financial crisis – and indeed helped cause it by creating a feeling of false confidence. So why is it that economic storms still come as a surprise? Perhaps it is because concepts such as the self-centered homo economicus have even less relationship to the movements of money, than Earth-centered circles do to the movements of planets.

Not normal

The theory assumes that the ‘invisible hand’ drives the economy to a stable equilibrium. But the price of an asset – say gold – is often very unstable. The reason is that we buy gold because we hope it will go up in value, and if the value is going up we get excited and buy more. This positive feedback drives the price up further. The same thing happens in reverse on the way down, resulting in an unpredictable series of booms followed by busts. The same effect is seen for other assets such as oil, housing, currencies, and so on.

Now, this kind of unpredictability is superficially consistent with the efficienct market hypothesis, and is often wheeled out as support for the theory. However unpredictability does not imply efficiency – snowstorms are unpredictable, but no one calls them efficient. The theory also assumes that price variations are random, independent, and therefore normally distributed. But as we know markets are susceptible to sudden changes including catastrophic crashes. Their statistics are not ‘normal’, as predicted, but are similar to those of earthquakes in that they follow a highly asymmetric ‘power-law’ distribution – most price movements are small, just as the Earth is constantly experiencing small tremors, but there is the ever-present possibility of extreme events. Orthodox theory sees the economy as fundamentally stable, but perturbed by random pieces of news, while in fact the economy is a lively and dynamic system, dominated by nonlinear feedback loops and prone to sudden changes.

The model assumes that people act independently and make rational decisions to optimize their own utility (the proof of market equilibrium actually assumed infinite computational capacity, and the ability to plan into the future). But as behavioural economists or for that matter advertisers know, emotions such as trust and fear play a vital role in markets. So does communication between market players, as shown by the rapid financial contagion that often follows a seemingly contained local crisis.

The original idea of neoclassical theory was to optimise happiness. But while the economy has grown enormously in recent decades, reported happiness levels have remained static or even declined slightly. Perhaps one reason is that we have internalised these values of rationality, independence, and utility optimisation; while happiness apparently has more to do with connectivity, communication with other people, and a sense of community. The model, it seems, does not correctly predict what makes people happy – and could in fact be making us unhappy.

New model

To summarise: our orthodox theory, which has long dominated economics, is based on ideas of stability, symmetry, rationality, and mathematical elegance which go back to the ancient Greeks. We have modeled people as if they were rational and can look into the future. We model the economy as if it obeyed the harmony of the spheres. Like the Greeks, we have imposed our ideas of order and logic onto the universe. But there is one important difference, which is that the Greek model could make useful forecasts of events like eclipses – they could predict when the lights were going to go out. Our models don’t have that degree of empirical validity. The financial crisis which began in 2007, and came close to shutting down the global financial system, went unpredicted by every major forecasting institution including the IMF, the OECD, and the US Federal Reserve.

Motivated in part by the crisis, we are now seeing the development of a viable alternative to mainstream economics, that exploits new tools from applied mathematics (e.g. complexity theory, network theory, nonlinear dynamics), as well as from other fields (e.g. psychology, ecology, ethics), and tempers them with a healthy dose of humility (models are incomplete patches that are only useful in certain domains and contexts). As argued in my book Truth or Beauty, this development is part of a larger shift in science from seeing the world as a Newtonian machine to seeing it as a living system.

Given that its drawbacks have been known for some time – so-called heterodox economists have been criticizing it for decades – how did the neoclassical model manage to persist for so long? In this case the answer has nothing to do with prediction. That leaves aesthetics and the status quo. As with the Greek model of the cosmos, the idea that the system is the stable, rational, efficient, and optimal outcome of a mathematically-governed process is clearly one that is favored by the small elite – that layer of distant stars in the financial firmament – which derives the most benefit from the current arrangement.

Once again, we need a new model of the world – and one that is based on the messy realities of people and the environment, rather than elegant circles. Could the financial crisis be the comet that shatters the crystalline spheres?



April 20, 2012

Video of TEDx talk on prediction at Park Kultury, Moscow

Filed under: Talks — Tags: — David @ 1:02 pm

Theme of this event was Future 2112


Create a free website or blog at