Thirty years ago, economics saw itself as the emperor of the social sciences, and believed that it was on the verge of subsuming all other disciplines under its mantle. Today, it is so seriously in decline that many academic economists feel like members of an endangered species. What on earth happened?
There are many causes, some of which are not unique to economics. The Dawkins “reforms”, the shift of universities towards vocational training—and of students towards credentialism—all reduced the appeal of “hard” subjects like economics. But most losses by the economics team were the product of own goals.
Firstly, the discipline deliberately turned away from the broad social science it once was in an attempt to recast itself as a science akin to physics. Regardless of whether this was or could be successful, it made economics far less appealing to the majority of potential students. The damage was not limited to a failure to attract students in the first place: the emphasis upon mathematical formalism also caused an even more drastic failure to stop commencing students decamping to other disciplines.
Secondly, this narrowing of focus virtually forced the formation of other specialist disciplines, to teach the issues economics could not or would not consider. Thirty years ago, the only other flatmate in the Commerce household was Accounting. Now, there are Marketing, Management, and many others, and relations between these newcomers and the once dominant resident are far from cordial.
While economists rarely express these sentiments out loud, the truth is that they think most of these other disciplines shouldn't exist. Economic theory tells economists that marketing is unnecessary–consumers already know their preferences, just put the stuff on the shelves and let them shop. Nor is management training necessary: just equate marginal cost to marginal revenue, and “Bob's your uncle”.
Marketing and Management academics beg to differ, and this has cost Economics dearly in terms of student bums on seats. Economists argue vehemently at Faculty meetings that students need more training in economics, and thus that any commerce degree should contain several units of micro and macroeconomics. The other members of the household listen politely, and then vote for less compulsory economics, not more. In 1987, UNSW required six economics units in any Commerce degree. Now there are three–and that is still a very strong position compared to the situation at most other universities today. In some Australian universities, it is possible to get a degree in, say, Accounting, without doing a single unit of economics.
The stealth war against economics wasn't limited to business faculties. Since economists willingly tossed out Marx, Veblen, Schumpeter, and all manner of other woolly thinkers and woolly topics, they were in no position to stop departments of Sociology, Political Science, even Architecture, occupying the space they vacated.
How have economists responded to this crisis? Apart from getting personally and collectively depressed, in general they have stuck to “business as usual”. This lack of innovation from an academic sect which normally champions the innovative spirit would be rather puzzling, except for one thing: economists feel that they shouldn't change because they are right and everyone else is wrong. Why change when you know the eternal truths, and everyone else is misinformed?
Unfortunately, it is the economists who are misinformed, not their critics. Their faith in the ultimate truth of economic theory reflects a lack of knowledge of their own discipline. This allegedly scientific social science practices not science but, as Hugh Stretton put it in his recent book, “scientism”, in which the appearance of scientific behaviour masks a crucially unscientific approach.
Many economists would regard this statement as sheer hyperbole. However, I believe it is easily substantiated using just two of the many issues I consider in Debunking Economics. I apologise in advance to all non-economist readers for the approaching avalanche of jargon, but it is unavoidable if I am to explain to economists and non-economists alike why the foundations of conventional economics are unsound.
Economics argues that a market economy achieves the highest possible social welfare when it allows utility-maximising individuals and profit maximising firms to determine the prices and quantities of all commodities in competitive markets free of both individual and corporate collusion.
Consumers are supposed to maximise their utility by working out the bundle of commodities that gives them the highest level of satisfaction, given the twin constraints of income and given prices. Two key aspects of this theory are that incomes and tastes are independent, and that consumers have pre-existing preferences for all possible bundles of commodities.
Economists represent this model of human behaviour with a graph that shows different quantities of two different goods, a set of smooth curves called “indifference curves”, and a straight line representing relative prices and the consumer’s income. Just as isobars on a meteorologist’s map links points of equal barometric pressure, indifference curves are supposed to link all combinations of goods that give the consumer the same level of satisfaction—so the consumer is “indifferent” between any combination along any given curve. At the same time, the consumer prefers more to less, so the further up this “indifference hill” she can move, the happier she is. The budget line shows how far up the hill the consumer can afford to tread.
This, to put it mildly, is a rather metaphysical way to model human behaviour—since indifference curves are as unobservable as the angels dancing on the heads of medieval pins. However, Paul Samuelson provided an apparent means to put measurement into metaphysics by his concept of “revealed preference”. This was the idea that, though indifference curves themselves couldn’t be seen, they could nonetheless be inferred from actual consumer behaviour.
A crucial outcome of this model of consumption is that, if a consumer prefers one bundle over another at a price ratio where both bundles are affordable (say, she prefers bundle A with “4 biscuits and 1 banana” to bundle B with “3 biscuits and 2 bananas”), then the former bundle must lie on a higher indifference curve than the latter. In this case, there should be no other price ratio at which the consumer’s preferences would reverse when both were affordable: no matter what other price ratio you suggest, so long as both A and B are affordable, she should always pick bundle A. If she chose B over A, then economists would describe her behaviour as “irrational”. Since rational behaviour is a hallmark belief of economists, such “irrational” people must be in the minority in the real world.
Unfortunately, experimental attempts to verify this preconception failed: the subjects breached this and other conditions of the economic model of rational behaviour.
Such a result in a true science may have eventually led to questioning of the underlying notion of what is rational behaviour—since it hardly helps to have a theory of rational behaviour which classifies the entire human race as irrational. But economists have continued on teaching this theory unmodified, because their a priori notions take precedence over the results of experiments.
A very simple Information Systems argument shows that it is not consumers who are irrational, but economists who cling to a computationally impossible model of human behaviour. Colloquially known as “the curse of dimensionality”, it establishes that it is simply impossible for every possible alternative in a decision space to be considered, once you move beyond a trivially small set of alternatives. Here’s how it undermines the economic theory of consumer behaviour:
Consider a consumer deciding what number of each of two goods to buy, where for simplicity she limits the number to be purchased to less than ten of each. That generates 100 possible combinations—everywhere from “zero biscuits, zero bananas” through to “nine of each”. If 90 of these can be ruled out without calculation—because they’re obviously too cheap or too dear given the available budget—there are 10 combinations whose costs have to be calculated (by adding up two numbers) and whose subjective desirability compared. Not a problem!
However, when you consider 3 commodities, the number of possible combinations jumps to 1,000; just 4 different commodities results in 10,000 possibilities, and so on.
By the time you’ve hit 30 commodities, then even if 99.9% of them could be ruled out without calculation, and even if all the necessary calculations (adding up 30 numbers each time now, plus the utility comparison) took 1 billionth of a second, it would take the consumer 32 billion years to reach a decision.
It is therefore somewhat apposite that the French students who have initiated a critique of the dominant strand of economic theory—known as neoclassical economics—chose to call their movement “Post Autistic Economics”—since only an autistic individual with savant skills in calculation could even approximate this still insufficient level of processing power.
In reality, the experimental subjects—who were instructed to try to maximise their utility given their income constraints—took about half an hour to choose varying amounts of between 2 and 4 goods out of 8 possibilities (using a personal computer to do the financial calculations). If we generously assume that the students were actually doing what the economists asked (in fact, their behaviour in general contradicted consumer theory), this implies a processing speed of at least a second per combination. Extrapolating from this, a real person would take a billion times the estimated age of the universe to “utility maximise”, given just 30 commodities—let alone the thousands which exist in the typical supermarket.
It is thus obvious that rational people can’t possibly behave as economists argue they do: “indifference curves” and all the economic paraphernalia associated with them are simply fictions of the economic imagination. But these fictions continue to be taught as rational fact to students of economics, and continue to inform the analysis undertaken by economists when developing theory, and advising on real world issues.
It must be noted, however, that the experimenters themselves concluded that “we should pay closer attention to the limits of this theory as a description of how people actually behave” and “we economists should perhaps be a little more modest in our ‘imperialist ambitions’ of explaining non-market behaviour by economic principles” (Sippel 1997: 1443). Indeed!
The theory of corporate behaviour fares little better. Economists learn very early in their training that the preferred market structure is “perfect competition”. This is because only in perfect competition is the marginal cost of production of a commodity equal the marginal benefit society gains from that commodity (one key mathematical concept on which economics relies is that, for uniformly varying functions, when two “marginals” are equal, the gap between their respective “totals” is as big as possible—thus when marginal cost equals marginal benefit, the gap between total cost and total benefit is as high as possible).
While economists also learn of other market structures apart from perfect competition, and the true gurus of the profession gain intellectual brownie points by working out the impact of these other structures on social welfare etc., “perfect competition” remains the subconscious nirvana towards which economists hope the economy will evolve.
All of this is a pity really, because the “perfect competition” ideal of marginal cost equal to price (which equals the marginal benefit of a product to society) is easily shown to be impossible. The key reason why is another marginal, “marginal revenue”—which represents the change in revenue earned by a firm when its sale price changes. If the price of a commodity falls and the quantity sold rises, the beneficial impact of the higher sales volume is somewhat offset by the negative impact of the lower unit price.
Before I explain the fallacy, an abbreviated explanation of the economic case for perfect competition is necessary.
According to economists, a profit-maximising firm will stop producing when the marginal benefit to it of producing an additional unit equals the marginal cost of producing that unit. In a market with one or very few firms, or where each firm has some degree of market power, this marginal benefit to itself will be less than the price, which represents the marginal benefit to society. Firms with market power will thus produce less than society needs, and charge a higher price for it.
Perfect competition allegedly gets around this problem by having so many firms that each firm, individually, has no market power. Therefore for each firm in a perfectly competitive industry, its marginal revenue equals the market price, so that the marginal benefit to it equals the marginal benefit to society. With each firm’s marginal revenue equal to the market price, collectively marginal cost equals price, and social welfare is maximised.
Well, that’s the theory. The problem is that it is based on a simple but profound mathematical error—that of equating a very tiny quantity with zero. Economic theory assumes that, at the market level, the quantity demanded will rise only if price falls, yet presumes that at the firm level, a competitive producer can sell as much as is feasible without altering the market price. These two assumptions are incompatible: they amount to believing that a long downward sloping line can be broken in to many perfectly flat lines, or that lots of perfectly flat lines can be added up to yield one downward sloping line. Either alternative is mathematically impossible.
The very simple mathematics behind this are explained on my web page; to cut to the chase, the bottom line is that “price equals marginal cost” is not an equilibrium. If any firm is producing at that level, it will increase its profits by reducing its output. The equilibrium for a competitive market turns out to be exactly the same as the equilibrium for a monopoly—which implies on theoretical grounds that the Australian Competition and Consumer Commission’s crusade to promote competition is a waste of time. What economists describe as the “deadweight loss of monopoly” is more correctly described as “the deadweight loss of profit maximising behaviour”.
Many other errors abound in the theory of the firm: the welfare comparison of competitive firms to monopolies contradicts the economic concept of diminishing returns (even if we ignore the mathematical problem outlined above), the “U-shaped cost curves” that economists love drawing are feasible only for trivially small levels of output, and so on. And there are many other technical errors in economic theory, which have been pointed out by all manner of economists—both supporters and critics—over the past 80 years (these are also detailed in Debunking Economics). But still mainstream economics is taught as if it is an unassailable truth.
This is scientism, not science. It is high time that economics abandoned its pre-Galilean approach, and got its hands dirty finding out how the real world actually works.
Fortunately, There are some isolated minorities of economists who have split ranks with the mainstream, and are trying to develop alternative approaches to economics that are dominated by realism rather than a priori theorising. The main realist alternatives are Post Keynesians and Evolutionary Economists.
Post Keynesians, as their name implies, believe that the work of John Maynard Keynes in the 1930s provides a more realistic foundation for economic theory. Their theories work at a more aggregative level than neoclassical economics, and they emphasise the essential role of money and the impact of an uncertain future on economic behaviour.
Evolutionary economists believe that the appropriate foundation for economic theory is not the mechanical statics that underpins neoclassical economics, but evolutionary biology. While their technical repetoire is still underdeveloped, practitioners tend to be highly skilled in modern mathematics and computing—with a knowledge of differential equations, complexity theory and computer simulations which easily exceeds the mathematical knowledge of conventional economists. They are also slowly subsuming the pre-existing school of Institutional Economics in a peaceful takeover.
While both schools are infants when compared to the dominant mainstream, there is some hope that they could become meaningful schools of thought if sufficient intellectual and financial resources were devoted to their development. But that is highly unlikely to occur from within economics departments themselves, given their continuing dominance by neoclassical economists. It may take a complete reversal of the “economic imperialism” that economists once envisaged for the new schools of thought to gain the ascendancy: other disciplines may have to invade the economic domain to bring realism to the fore, and to return those “woolly topics and woolly thinkers” to legitimacy within economic discourse. Such an external reformation would not be necessary if economics could develop the ability to keep its own room in order, but past behaviour gives little room for confidence.
The would-be emperor of the social sciences may one day find his empire subjugated, and proper economics— analysis of the evolutionary dynamics of the complex social system in which we actually live—may ultimately be the winner.
References
Keen, S. (2001). Debunking Economics: the naked emperor of the social sciences, Pluto Press, Sydney.
Sippel, R., (1997). “An experiment on the pure theory of consumer’s behaviour”, Economic Journal, 107:1431-1444.
Stretton, H., (1999). Economics: a new introduction, University of New South Wales Press, Sydney.