11 They ignore the role of credit in the economy, an issue that looms very large in my later critique of macroeconomics.

12 The remark may well be apocryphal see en.wikipedia.org/wiki/Let_them_eat_cake; but the sentiment of the wealthy disregarding the fate of the poor certainly played a major role in ushering in the French Revolution.

13 Or returns two values for one input.

14 Mas-Colell's a.s.sumption of a 'benevolent central authority' that 'redistributes wealth in order to maximize social welfare' is probably derived from this ridiculous paper by Samuelson, since he references it as a paper 'For further discussion' (Mas-Colell, Whinston et al. 1995: 118).

15 Say gives a typical statement (reproduced on the Web at Hedonism/Say) of this approach to utility, which denies the ability of anyone to judge or measure the utility any other individual garners from a particular commodity.

16 Kirman's paper is an eloquent and well-argued instance of the phenomenon that those who have constructed the 'high theory' of economics are far less confident about its relevance than more ordinary economists.

17 The 'plus one' rule covers the case of buying no units of one commodity. This isn't an issue in my discrete interpretation of Sippel's experiment.

18 If this sounds extraordinary to you, consider that 10 multiplied by itself 8 times is equal to 100 million.

Chapter 4.

1 In fact, the advanced courses also ignore these more difficult critiques, which means that students who do them, if anything, are even more ignorant than undergraduates.

2 This last paper 'Debunking the theory of the firm a chronology' is freely downloadable from www.paecon.net/PAEReview/issue53/KeenStandish53.pdf.

3 Economists are likely to deflect these critiques by arguing that the theory has moved well beyond the simplistic models taught to undergraduates. However, at the very least economists should stop teaching these models. Secondly, economists still see the model of perfect compet.i.tion as describing the ideal economy. This chapter argues that this ideal is in fact a farce.

4 Astute readers would already see a problem here with the model of perfect compet.i.tion.

5 These numbers come from a mathematical function (a cubic of the form a+bx+cx2+dx3 to be precise), whereas most neocla.s.sical textbooks, if they use numerical examples at all (most don't, and simply use drawings instead), simply 'pluck them out of the air.'

6 If this example seems silly to you surely you would only use workers and machines in the ideal ratio, so if you have just one worker, he works one jackhammer while the other ninety-nine are left idle? then congratulations, you're right! The whole idea that firms vary the ratio of fixed to variable factors as economists a.s.sume they do is a nonsense that we tackle in the next chapter.

7 Average fixed costs start off very high in this example, at $10,000 per unit for the first unit produced and fall uniformly from then on. Variable costs per unit may fall for a while too as productivity rises, but eventually they start increasing as output rises, and marginal productivity falls. The combination of falling fixed costs per unit of output, and rising variable costs, means that average costs are 'u-shaped': they fall while the firm experiences rising marginal productivity, flatten out as diminishing marginal productivity kicks in, and finally rise when marginal productivity has diminished so much that each additional unit costs more to produce than the average to date.

8 In fact, on the two occasions that Smith used this phrase, it was in relation to income distribution, and whether local producers would ship production offsh.o.r.e, not how the market mechanism operated. But the metaphor had a compelling impact on the development of economic theory about the market.

9 There are two other models, known as the Cournot-Nash and Bertrand models, in which firms do react to what they think other firms will do, which also reach the outcome that price equals marginal cost. Though they don't make the same mathematical error as the Marshallian model does, they have other problems that we discuss in Keen and Standish (2010). In a third edition of Debunking Economics, I might add an addendum on this since I'm sure it will be the refuge of those who wish to cling to the neocla.s.sical model but I've left it out of this edition to avoid boring the rest of my audience to death.

10 This a.s.sumption is inconsistent with the a.s.sumption of a 'short run,' during which some factor of production cannot be changed, which is essential to get the phenomenon of diminishing marginal productivity, which in turn generates a rising marginal cost (see Chapter 4). Firms already inside the industry are a.s.sumed to be unable to alter their capital equipment at all, but in the same time period firms not currently in the industry can build a factory and move in on the market? h.e.l.lo? This logic is about as watertight as the script of the average TV soap opera. However, this a.s.sumption plays no part in the standard mathematical model of perfect compet.i.tion, which focuses simply on the impact of the number of firms currently in the industry.

11 If this argument doesn't convince you, good because this is the point at which the economic argument starts to take on that 'flat earth' feeling.

12 There is no specification of time in the standard neocla.s.sical model, so this could be, for example, 135 units per minute.

13 The Cournot and Bertrand theories erroneously argue that the level that maximizes firms' profits is identified by the firms behaving collusively, like a pseudo-monopoly. Instead, in Keen and Standish (2010), we show that the so-called collusive output level is simply the level the firms would produce if they behaved as simple profit maximizers.

14 Sraffa's paper 'The law of returns under compet.i.tive conditions' (Sraffa 1926) critiqued a forerunner to this idea: that economies of scale could be external to the firm, but internal to the industry. This would mean that as an industry expanded, all firms benefited from lower costs, but none benefited any more than any other. Sraffa argued that few, if any, economies of scale would fit in this category: instead most would be internal to a firm, and thus advantage the big firm more than the small.

15 The revised formula, in this 1,000-firm example, is that the firm should make the gap between its marginal revenue and marginal cost equal to 999/1000th of the gap between market price and its marginal cost. The number of firms can be safely ignored and the output level chosen will still be approximately right, whereas the neocla.s.sical formula remains precisely wrong.

16 The then editor of the Journal of Economics Education, Bill Becker, was himself keen to have the paper published, and submitted it to eminent referees to try to improve its chances of being accepted.

17 In addition, to compare compet.i.tive firms to monopolies at all scales of output, then for reasons outlined in Chapter 5 the marginal cost curve must be drawn horizontally.

Chapter 5.

1 Smith and Ricardo allowed exceptions to this rule; neocla.s.sical economics in effect made these exceptions the rule.

2 Money is simply a measuring stick in this a.n.a.lysis, and the monetary unit could as easily be pigs as dollars.

3 As we'll see shortly, this generates just as many problems as aggregation of demand curves did.

4 Significant fixtures like factory buildings are an exception here, though machinery within the factory is not. Kornai's observations about the spare capacity in production, which I note later, supplement Sraffa's critique on this point.

5 Kornai used this last constraint to develop the concept of 'Hard and soft budget constraints.' This has great relevance to the position of banks during the Great Recession (see Kornai, Maskin et al. 2003: 11236; Kornai 1986), but is less relevant here.

6 Marketing expenses cannot be added in to 'rescue' the doctrine, since the true purpose of marketing is to alter the firm's demand curve, and this only makes sense if firms produce differentiated products something the theory of perfect compet.i.tion explicitly rules out.

7 The 'cost curve' for any one firm or industry is the product of interactions between all industries, an issue that is ignored in the neocla.s.sical treatment of a single market. This issue is discussed in Chapter 6.

8 Economics ignores the issue of ecological sustainability, though clearly it must be considered by a reformed economics.

9 Friedman argued that the result that businessmen do not make their decisions on the basis of marginal cost and marginal revenue was 'largely irrelevant' (Friedman 1953: 15).

10 Though empirical work suggests that, in practice, there is little sign of any negative relationship between the quant.i.ty sold and the price and hence little evidence of a 'demand curve' (Lee 1996).

Chapter 7.

1 While this case is most easily made with equations, I'll stick to words here.

2 The same case can be made with respect to the change in the wages bill, but I focus just on profit times capital to keep the argument simple.

3 The ratio of a change in capital to a change in capital is 1.

4 This will apply only when the capital-to-labour ratio is the same in all industries which is effectively the same as saying there is only one industry.

5 Of course, this argument has already been eliminated by the 'benevolent central authority' a.s.sumption derived from the Sonnenschein-Mantel-Debreu conditions.

6 The rule in this example is that 10 quarters of wheat had to exchange for 1 ton of iron, or 2 pigs. These are relative price ratios in which commodities exchange rather than absolute prices in terms of money.

7 At least, not until a 'von Neumann machine' a machine that can both produce output and reproduce itself is invented.

8 This is often all economists know of Sraffa's critique, and they dismiss it immediately by saying that it wrongly ignores the issue of marginal productivity. In fact, there is much more to Sraffa's critique, and Bhaduri's critique establishes the invalidity of the a.s.sertion that the rate of profit equals the marginal productivity of capital.

9 When output is measured in terms of a 'standard commodity,' and when the wage is normalized so that when the rate of profit r is zero, the wage w equals 1.

10 This correspondence is not exact, but it can be made accurate to any level short of 100 percent by continuing the process of reduction for long enough.

11 Approximately because of the irreducible commodity residue left from the reduction process.

12 I'm enough of a wine buff to realize that this example is practically impossible, but it will do as an ill.u.s.tration.

Chapter 8.

1 I have hardened my opinion on this front since the first edition, when I was willing to describe economics as a science, though a rather 'pathological' one.

2 I first heard this joke in a public debate between my then professor of economics and a physicist. I now appreciate the irony that physicists are turning their attention to economics and in general being horrified by neocla.s.sical economic theory.

3 I am grateful to my student Marchessa Dy for suggesting this very evocative a.n.a.logy.

4 The a.n.a.lysis below is a brief summary of Imre Lakatos's concept of competing 'scientific research programs.' The philosophy of science is today dominated by more 'postmodernist' concepts. I will leave exploration of these newer strands to the interested reader to pursue.

5 This reference to physics is now seriously dated, since this empirical observation has now been corroborated see the Wikipedia item on the 'Accelerating Universe' for a brief discussion.

6 Ironically, Austrian economics, an alternative school of thought that is very closely related to neocla.s.sical economics, differs by singing the praises of capitalism as a disequilibrium system (see Chapter 18).

7 Equilibrium in turn has been endowed with essential welfare properties, with a 'Pareto optimal equilibrium' being a situation in which no one can be made any better off without making someone else worse off.

Chapter 9.

1 If you have ever taught a child to ride a bike, you would know that this lesson is the most difficult one to grasp that a moving bike balances itself, without the need for training wheels or other props which would keep it upright when it was stationary.

2 This a.n.a.logy is apt in more ways than one. The art of balancing a stationary bike requires great skill, and anyone who has mastered it is likely to 'show it off' at every opportunity, regardless of how impractical it might be. Similarly, economists who have mastered the difficult mental gymnastics involved in equilibrium a.n.a.lysis take every opportunity to parade their prowess regardless of how irrelevant this skill might be to the art of managing a real economy.

3 Only the gold market in London even approaches this structure, and even that is a market at which only one commodity is traded, rather than 'all commodities' (O'Hara 1995).

4 The main theorem is the Perron-Frobenius theorem on the eigenvalues of a positive matrix. See en.wikipedia.org/wiki/Perron%E2%80%93Frobenius_theorem for an explanation.

5 Debreu used a notation that allowed for negative prices and negative quant.i.ties. However, this was a convenience only, and has no impact on the a.n.a.lysis in this section.

6 My discussion of the instability of general equilibrium above was with respect to a production economy, where the nature of the input-output matrix makes stability impossible. There is no inputoutput matrix in an exchange-only economy because there is no production!

7 The actual equations were: 'the rate of change of x with respect to time equals the constant a multiplied by (yz); the rate of change of y with respect to time equals x multiplied by (bz) minus y; the rate of change of z with respect to time equals (x multiplied by y) minus (c multiplied by z).'

8 I use chapter and section references for Marx, rather than page numbers, since his work is now freely accessible via the Internet from the site www.marxists.org/archive/marx/.

9 The two equations are linked, because workers' wage demands depend on the rate of employment, while investment which determines the rate of growth depends on income distribution (a higher workers' share means lower profits, and hence lower investment).

10 For more details, see the Wikipedia entries en.wikipedia.org/wiki/Functional_flow_block_diagram, en.wikipedia.org/wiki/Transfer_function, http://en.wikipedia.org/wiki/State_s.p.a.ce_(controls) and en.wikipedia.org/wiki/Systems_engineering.

Chapter 10.

1 he became Fed chairman in February 2006, having briefly served as chairman of the president's Council of economic advisers before that.

2 more strictly, a market demand curve can have any shape that can be described by a polynomial equation. This rules out a curve that returns two or more prices for the same quant.i.ty, but allows curves that return the same price for many different quant.i.ties.

3 keynes lumped what we today term neocla.s.sical economists with those we today call the cla.s.sical economists. While they are distinctly different schools of thought, keynes was correct to group them together on this issue, since they concurred that a general deficiency of aggregate demand was impossible.

4 Surprisingly few books give this argument in full, given the extent to which it is a core belief in economics. Two that do are Baird (1981: ch. 3) and Crouch (1972: ch. 6).

5 his actual procedure was to argue that, when employment increased, demand for consumer goods would increase by less than the increase in employment, and that equilibrium would be achieved only if investment demand automatically took up the slack. This confusing argument is equivalent to the simpler case set out here.

6 as milgate observed, 'received opinion, that keynes's General Theory is a contribution to "disequilibrium" a.n.a.lysis, was stamped indelibly upon the collective consciousness of the economics profession at an early date by critics and converts alike' (milgate 1987).

7 I explain how marx derived this result in Chapter 17.

8 For those of you for whom mcCarthyism is ancient history, see en.wikipedia.org/wiki/mcCarthyism. Though mcCarthy was out of the picture by the late 1950s, the influence of that period continued for many years.

9 I have found that many people find this confusing on the basis that, if debt has financed a purchase, wouldn't that already be recorded in Gdp? There are two reasons why this is not the case. First, part of spending is on pre-existing a.s.sets which are not a component of Gdp. Secondly, in our demand-driven economy, the demand comes first before the supply and demand can be sourced either from previously earned income, or an increase in debt where this debt reflects an increase in the money supply by the private banking system, as I explain in Chapters 12 and 14. The debt-financed demand for commodities does later generate production of more commodities, and this turns up in Gdp but the debt precedes the supply. This relationship is thus best thought of in 'continuous time' terms: aggregate demand at a point in time equals income at that time, plus the change in debt at that time. aggregate supply (and the sale of existing a.s.sets) follows slightly later.

10 Sometimes they do, of course, but in order to clarify his argument Schumpeter considers the case where an entrepreneur does not have pre-existing money and must therefore borrow to finance his venture.

11 marx's theory of value is normally regarded as the labor theory of value, which is criticized in Chapter 13. I argue that his theory of value is something quite different.

12 There is an interesting parallel in research into producing robots that can walk. The first attempts designed a robot that always kept its center of gravity directly above the foot in contact with the ground resulting in a robot that was always in gravitational equilibrium, but which could walk only in straight lines with five seconds between steps. To enable fluid motion, the researchers found they had to put the center of gravity in continuous disequilibrium: then it could walk as naturally as we humans do. See world.honda.com/aSImo/history/e0.html and world.honda.com/aSImo/technology/walking_02.html for details.