Epilogue
Reception of Debunking Economics
After reading Debunking Economics, I doubt that you would be surprised to learn that the book provoked very different reactions from different classes of readers. Those for whom I wrote the book—people who were dissatisfied with conventional economic theory but did not know the theory well enough to articulate their objections to it—embraced the book enthusiastically. On the other hand, committed neoclassical economists hated it. I was of course pleased with both reactions!
My publishers were not disappointed either. A print run of 1200 is considered successful in Australia for a book like Debunking Economics. It has sold 4,000 copies to date in Australia, and a similar number overseas (mainly in the USA and UK). It is still selling strongly 4 years after its publication.[[1]]
My main ambition for Debunking Economics was not to achieve enormous sales, but to disturb the hegemony of neoclassical economics on the way the world thought about economic issues. My main contribution here was to make already established critiques accessible to non-economists and new students of economics. However while drafting the book, I realised a flaw in the neoclassical theory of the firm that took shape in the chapter “Size Does Matter”.
Since this was a new critique, I expected to attract the most animosity from neoclassical economists—and I was not disappointed. Almost all of the formal and informal criticism that neoclassical economists have made of my book have been directed at that chapter. Fortunately those critiques have not weakened my case, but strengthened it.
Fallacies in the neoclassical theory of the firm
As you will recall, the key argument in “Size Does Matter” is that the demand curve for the individual firm has to have the same slope as the market demand curve: it cannot be “horizontal” while the market demand curve is downward sloping. At the time I wrote Debunking Economics, I had realised this logically, but had not done the mathematics to support my case. After finishing the book, this was the first research project I undertook (and I was ably assisted by many colleagues, especially Russell Standish from the University of New South Wales John Legge from La Trobe University, Trond Andresen from the Norwegian Institute of Technology, and Geoff Fishburn from the University of New South Wales).
Possibly the most remarkable thing I learnt in this process was that this argument was not only not new, but had been made almost half a century earlier by one of the most trenchant defenders of neoclassical economics, George Stigler (Stigler, G.J. 1957. “Perfect competition, historically considered”. Journal of Political Economy 65: 1-17). Stigler used what is known as the “chain rule” in calculus to show that the slope of the demand curve facing the individual firm was the same as the slope of the market demand curve, multiplied by how much market output changed if a single firm changed its output. The latter quantity is simply 1 for a market consisting of many independently managed firms: if one of them in isolation chooses to change its output, then total output rises by that amount. Therefore the slope of the individual firm’s curve equals the slope of the market demand curve.
In a telling sign of the intellectual dishonesty of neoclassical economics, there is, to my knowledge, no neoclassical textbook that mentions this fact: all still continue to teach that a downward-sloping market demand curve is perfectly consistent with a horizontal demand curve for the individual competitive firm.[[2]] Nor has so-called learned literature—academic economic journals—treated this proposition any better: Stigler’s 1957 paper has been cited only once in the past 20 years, and then in a Law journal rather than an economics one.[[3]] Stigler’s straightforward, accurate piece of calculus seems to have died a death of ignorance in economics simply because its truth is inconvenient.
There are some neoclassical economists who are aware of Stigler’s argument, but they take solace in Stigler’s own attempt to circumvent his own logic. Stigler tried to show that, even though the slopes of the market and individual firm demand curves were identical, profit-maximizing behavior was consistent with perfect competition. He reworked the formula for marginal revenue for the single firm to show that this was equal to price plus a second term which was divided by the number of firms in the industry. He then argued that “this last term goes to zero as the number of sellers increases indefinitely” (Stigler 1957: 8), so that “in the limit”, profit-maximizing behavior meant that marginal revenue would equal price and price in turn would equal marginal cost.
While the mathematics was correct, this didn’t make sense to me: something had to be wrong with the argument. When I investigated the issue carefully, I was delighted to realise that indeed there was something wrong: the mantra that so many students of economics have learnt, that “profits are maximized by equating marginal revenue and marginal cost” is wrong.
The error is simple. The technique of maximizing the gap between two functions by finding where their rates of change are equal is mathematically valid so long as you include all factors that cause the two functions to change. Under the conditions of neoclassical analysis, changes in the costs faced by a single firm are caused only by changes in its output. But for the same firm, changes in its revenue occur not only because of what it does, but also because of what all other firms do. This is especially because the single firm can’t control its competitors. The firm is like an oarsman who can’t control what the winds and currents do, but must nonetheless row his boat according to the winds and currents if he is to get to his destination.
Technically, this makes working out the profit-maximizing output level for a single firm a problem in what is known as total differentiation, whereas what economists call “marginal revenue” is an exercise in partial differentiation—seeing how the firm’s revenue changes as a result only of changes in its own output. This is like trying to row across a river to a particular spot while ignoring the impact of the wind and current on your direction.
The solution to the properly defined maximization problem had a delicious irony. Neoclassical economics is obsessed with equilibrium, and the intersection of curves as determining everything in economics. However, the profit maximizing formula for a single firm involves not the intersection of two curves, but the location at which the gap between one pair of curves (marginal revenue and marginal cost) is a fraction of the gap between two others (price and marginal cost).[[4]] The neoclassical rule held only in the case of a monopoly, because only then was revenue for the single firm exclusively a function of the behavior of that firm.
The mathematical analysis also confirmed that, as argued verbally in Chapter 4, supply and demand analysis is not possible: if neoclassical assumptions are accepted, then in the aggregate all market structures behave like a monopoly. Therefore market equilibrium occurs not where price equals marginal cost (which is where the demand and “supply” curves intersect), but where marginal revenue equals marginal cost.
The mathematics behind this argument is both simple and incontrovertible, and in a true science this would have made publication and promulgation of the argument quite straightforward. Not in economics, however: my attempts to have this argument published have also inadvertently proven how resistant neoclassical economics is to logical criticism.
My argument has been published in four publications to date: the interdisciplinary physics journal Physica A; a special issue of the American journal Utilities Policy which was devoted to discussing the debacle that flowed from the deregulation and privatisation of California’s electricity industry; a new European annual New Economic Windows; and Edward Fullbrook’s edited collection A Student’s Guide to What’s Wrong with Economics. But it has been rejected by several mainstream economics journals, including the Economic Journal, the American Economic Review, and the Journal of Economic Education.
The grounds given for rejection are instructive of the intellectual blinkers economists are willing to wear in order to hang on to core beliefs. In several cases, referees simply failed to comprehend—or perhaps even read—the essential propositions in the paper (this was the case for the American Economic Review). In others, the referees rejected a key premise—the proposition that the slope of the firm’s demand curve had to be the same as the market demand curve—on little more than a whim.
One referee for the Journal of Economic Education, for example, stated that “Stigler's many attempts to save neoclassical theory have always caused more problems than they have solved. His version of the chain rule is contrary to the partial equilibrium method and thus is irrelevant”. Another tried to argue that the demand curve was horizontal from the point of view of small firms because prices are stated in currency units. Therefore the demand curve is not a smooth function but a series of very long flat steps, with the height of each step being the minimum currency unit (a cent in America for example).[[5]] This objection is spurious and easily refuted, but comments like it have to date kept the paper out of mainstream journals.
This experience has confirmed something that I already knew: economics is extremely resistant to criticism, and is not likely to “reform its own house”: outside pressure will be needed if economics is ever going to become a science. Fortunately, outisde pressures have risen since the initial publication of Debunking Economics.
Subsequent developments in Economics: PAECON
At much the same time as I was writing Debunking Economics, a group of French PhD students were becoming increasingly dissatisfied with the barrenness of their education in economics. While instruction in economics is bad enough in most of the Western world, it appears that the situation is particularly bad in France. The curriculum for economics is centrally determined there (which makes it very difficult for my counterparts in France to provide students with a critical perspective on economics), and it emphasises solving neoclassical mathematical puzzles.
These students took the politically brave step of rebelling against this approach to economics, which they labelled “autistic”, and they published a manifesto of their beliefs (the permanent website for the movement, http://www.paecon.net/, reproduces their manifesto in full). Their petition, first published in June 2000, commenced with the statement that:
We, economics students of the world, declare ourselves to be generally dissatisfied with the teaching that we receive.
It then gave three broad reasons for their dissatisfaction:
1. We wish to escape from imaginary worlds! 2. We oppose the uncontrolled use of mathematics! 3. We are for a pluralism of approaches in economics!
Finally, it ended with an appeal to their teachers:
4. Call to teachers: wake up before it is too late!
The reaction to their political bravery was remarkable, and came from all sides. The French newspaper Le Monde reported on their petition, and interviewed several non-neoclassical academic economists who had been feeling just as oppressed as their students. At the same time, neoclassical economists condemned them, and attempted to establish a counter-movement. The French media took up the students’ protest, while over 200 French academic economists signed the PAECON petition. A public debate led to a government investigation of their claims, established by the French Minister of Education.
The report, written by Jean-Paul Fitoussi, President of the l'Observatoire Français des Conjonctures Economiques, broadly supported the students, and the French Minister of Education has promised that changes will be made (however for the moment economics training continues as per usual in France).
The expression of dissatisfaction in France led to similar reactions by PhD students in Cambridge England, and many other parts of the world. A “PAECON” movement—Protest Against Autistic Economics—was born. The movement is still in its infancy, but there are groups of students and professional economists meeting and criticially debating economics in many countries around the world—including, I am delighted to say, China. These groups maintain websites that are regularly updated with news on economics from a non-neoclassical perspective. The main website (maintained by Edward Fullbrook in the UK) is:
Www.paecon.net
and the Chinese PAECON website is:
Http://pf.nccu.edu.tw/Faculty/calin/index.htm
One important activity of the PAECON group is producing accessible guides to problems with economics. The prolific Edward Fullbrook produces a regular post-autistic economics review (now up to its 30th issue) that has over 8,000 subscribers around the world (you can subscribe for free at the PAECON website), and has also published several books of readings—most recently A Guide to What’s Wrong with Economics (Anthem Press 2004).
This movement shows that dissidents against neoclassical economics are many and widespread. If you feel that economics is out of touch with the real world, you are far from alone! The movement and the community of critical scholars it has helped create also keep alive the dissident tradition in economics, which has existed for as long as neoclassical economics has been dominant. But for several reasons, I do not think it will lead to substantial change within economics. If this were the only force against neoclassical economics, it would continue to dominate economic thinking and be only trivially troubled by the non-believers within its ranks.
However, there are also non-believers outside its ranks, and these I think may prove to be far more dangerous to the hegemony of neoclassical economists: physicists.
Subsequent developments in Economics: Econophysics
When I wrote Debunking Economics in 2000, I concluded with a far too brief survey of alternative schools of thought, and noted the existence of the “complexity” school. This has since developed substantially and is now known as “econophysics”. The vast majority of practitioners in this area have PhDs in physics, and are applying the advanced mathematical and experimental techniques of that discipline to issues in economics. The first International Econophysics conference was held in Bali in 2002, the journals Physica A and Quantitative Finance are largely devoted to their interests, and researchers are producing literally thousands of advanced technical papers each year. There are many web sites devoted to econophysics, with the key site being:
http://www.unifr.ch/econophysics/
The majority of econophysics researchers are interested in the behavior of finance markets, simply because the enormous volume of accurate data concerning financial transactions. In contrast, most economic time series are inaccurate and pitifully short. For example, statistics on inflation in the USA are provided monthly and go back only till 1913, so that there are only about a thousand data points in the series. The series itself compresses and therefore distorts a huge amount of information, using indices to derive an average rate of inflation from the rates of change of prices in a huge number of very different commodities.
Finance data, on the other hand, is voluminous and comparatively accurate. There is data on every transaction (known as “tick” data), it records the trading of well defined securities, and the sheer volume of data is enormous: time series with tens of millions of observations are the norm.
Why the obsession with data? Because, since econophysicists have been trained in a real, experimental science, they are not interested in constructing “thought experiments” or models based on a priori reasoning. They want to explain what exists, and in order to know what exists they must have data. Only after they have analysed the data do they venture to develop abstract models that might explain it.
What they have found in this data to date is both fascinating and in strong contrast to what economists alleged existed. For example, you will remember the “efficient markets hypothesis” (EMH) I discussed in chapter 10, which constructed a model of finance from the a priori notion of a rational investor—where “rational” was redefined to mean what is called “prophetic” in ordinary speech. One of the predictions of this model is that movements in share prices should be reasonably defined by an average daily increase, with variation around this that is statisically normal.
Instead, physicists found that share price movements followed what they describe as a “Power Law”. This says that the number of events of a given size (say the number each century of 10% daily movements in the Dow Jones stock exchange index) is roughly equal to that size raised to some constant power. This is a very different picture to the one predicted by economists. The most crucial difference is that a Power Law predicts that stock market crashes—really big movements of, say, 6 per cent or more in one day—will occur, and furthermore that crashes of any size can occur: even crashes of 20 per cent or more in one day.
|
|
Figure 1: Log-log plot of size and frequency of stock price movements
Figure 1 illustrates the difference between these perspectives with a plot of the daily movements in the Dow Jones over the 20th century versus a Power Law fit, and a fit based on the belief that the movements in prices were simply random. The values on the axes are logs of the relevant numbers, so -2 on the horizontal axis means 1 divided by 10 squared, which is 0.01 or 1 per cent. 4 on the vertical means 10 to the power of 4, or 10,000.
Each dot on the graph represent a percentage movement and the number of times a movement of that magnitude occurred on one day during the 20th century. The first dot on the far left represents how many times the Dow Jones index moved 1% or less in a day during the 20th century; the second is how many times it moved 2%; and so on.
Obviously for the low percentages, the random movements (EMH) prediction—the line with the square symbols—is pretty accurate. For example, the “EMH” prediction[[6]] is that the market would move as much as 3% 719 times last century;[[7]] in fact it moved that much 780 times. But for larger percentages, the “EMH” prediction becomes widly inaccurate. There were 106 days on which the market moved as much as 5%; the EMH predicts that movements this big should have occurred only 3 times last century. For really large movements, the EMH prediction is a joke: it predicts that the odds of an 8% movement were less than one in a million in the last century when in fact there were 11 such movements; it gives odds of less than one in a billion for a 9% movement when there were 3.
The physicists’ alternative of a Power Law, on the other hand, is very accurate for larger movements. The Power Law predicts that the last century should have had roughly 70 daily movements of 5%; in fact there were 106; it predicts 11 movements of 8%, and indeed there were that many.
The physicists’ statistical portrayal of the stock market is a far more accurate characterisation of its behavior, and it has led to a far more accurate abstract model: the “minority game”. The EMH assumes that all investors agree about the underlying value of all stocks, that therefore investors don’t speculate against each other, and the market is modelled as if it is stable and approaching equilibrium. The minority game instead assumes that only a minority can be right, that players do speculate against each other, and since only a minority can be right, the equilibrium cannot be stable.
As they have developed this and other models, econophysicists have been simply mystified by the theories and statistical approaches of neoclassical economics. This is both highly ironic, and also quite dangerous for neoclassical economics.
It is ironic because the founders of the neoclassical school believed they were bringing the rigour of physics to economics when they began their program to make economics a mathematical discipline in the late 19th century. It is dangerous because in the past, neoclassical economists have tended to dismiss their critics as mathematical simpletons. Now they have a set of critics who are far more mathematically sophisticated than they are. They are quite capable of taking economics over from the outside. I wish them every success.
Reader feedback
I have been very heartened by the feedback I have received from readers over the past three years. I have received literally hundreds of emails from readers, and their comments have confirmed that, despite its blemishes, the book lived up to the ambitions I had for it. Of the many comments I received, one from Valerie Radford in the USA was possibly the most gratifying. I have seen far too many intelligent people falsely question their own intelligence because they couldn’t “get” neoclassical economic theory, when in fact it was the theory that was at fault and not them:
I just wanted to send a note to personally thank you for your thought provoking and inspiring book. I am an "ex-" graduate student of Economics who, in the throws of endless mathematical computations, became thoroughly disillusioned with the relevance of economics during her graduate education. Your book put it all in perspective for me and I am forever grateful. Now let me start out by saying that I am not in the 95% of folks who thoroughly dislikes Math -- so I am not another person that is going to bemoan the fact that Economics has gone "Math." In fact, I actually like the subject. I earned my undergraduate degree in Industrial Engineering precisely because I like Math. And, more importantly, I like applying Math to solve problems which on some level is Engineering. Part of my undergraduate coursework included taking a few introductory course in Economics (of all the engineering disciplines Industrial Engineering prides itself in being most "business-like" and thus all students are required to take core business courses) and I got hooked on Economics. I thought that at last I found a discipline where I could apply my analytical techniques to solve issues that business faced. My plan was to finish my engineering degree, work a little, but then move on to graduate study in Economics so I could get my degree and then, armed with all this knowledge and understanding, consult and make the business world a better place.
So I did just that ... but as soon as I started taking graduate classes the dream started to unravel. I knew graduate study was going to be difficult, I was warned of the long hours of endless reading and studying. What I wasn't prepared for was just how little reading (Varian and Henderson and Quandt could hardly be considered reading!) and just how much studying or shall I say computing I had to do. My entire graduate life could be characterized by me endlessly backsolving algebraic problems for answers that were given as divine truths, but never quite made mathematical sense. There was never any reading, there was never any discussion about the different schools of classical thought nor less the alternative schools, nothing. We barely touched Keynes, nor less Marx -- and forget about Sraffa (who by the way I never knew even existed until your book!). A graduate level economic history course was not even offered. But even given that I persevered thinking that once I learned the basic analytic tools, I would be free to go off and apply them to whatever problem I so chose.
Well it just got worse -- the Math became more confusing and theory more illogical the more I progressed. There I was a former-engineering student who fancied herself as being pretty good at Math and I couldn't even get above a C in an exam because I couldn't do the Math -- it just didn't add up to me! Nothing made sense to me. I managed to struggle to complete my master's degree supplemented with a few business courses so that I could actually get a job. Frustrated and humiliated, I announced my defeat and exited to corporate America. The relevance that economics affords to business was clearly above my head and I was suitably brought down to size.
Now, 10 years later, I have finally come to terms with my graduate education after reading your book. I realize now that it wasn't just me. The numbers really don't add up. There are inconsistencies in the logic. I should have had the opportunity to take courses in economic history and philosophy -- including the alternative schools of thought. I should have spoken up! So thank you -- thank you for showing me that it was not just me and for making me remember why I chose to study the subject in the first place. And lastly, for re-igniting in me the spark that made me want to study Economics in the first place.
I could ask for no more gratifying feedback than that.
Where to from here?
If Debunking Economics has ignited your interest in a non-neoclassical approach to economics, there are now many resources out there to further your education away from this barren mainstream. My website has grown over time, and there are now many presentations (almost all of which are in the form of Microsoft Powerpoint slide shows) that delve more deeply into issues covered in the book. The PAECON movement’s websites and books provide a wealth of reading, and link to many active developers of new approaches to economics. The econophysics movement is developing apace, and there are numerous papers and books available—though the mathematical level of these is often daunting.
From my point of view, there are two tasks to be undertaken: developing my own non-neoclassical economic models (based mainly on the work of Schumpeter, Keynes, Richard Goodwin, Hyman Minsky and the Circuitist/endogenous money school of thought), and keeping my critique of neoclassical economics up to date.
At some stage, the latter task will necessitate a second edition of Debunking Economics. In part this will be necessary to embellish the arguments of the first edition. For example, the critique of the theory of the firm neglected the substantial empirical research that shows that marginal costs fall for the vast majority of firms; my critique of the neoclassical theory of the firm has also become much stronger; and I’ve added to my critique of consumer theory. In part, it is necessary to address issues that were left out of the first edition: international trade theory, the concept of “Pareto optimality” as a way of sidestepping the inability to aggregate individual utility, game theory, and so on.
Clearly, there is very much to do if we are to rid ourselves of an illusory explanation of the market economy and finally develop a realistic model of economics to guide the development of human society.