Critique of Austrian Economics
Part III: Conclusion
Austrian economists (by nationality – they were not a school until Hayek’s famous lectures) were already weak by 1930. Menger was good. His 1871 Principles of Economics (1981) is one of the most important economics book ever written. But, with the publication of Böhm-Bawerk’s 1889 Positive Theory of Capital (1959), Austrian economists split into two branches. Menger did not find a worthy successor until Mises and together they laid the groundwork for the Axiomatic School founded by this author (1999). Meanwhile, Böhm-Bawerk was laying the groundwork for what would become the Hayekian School. This branch has grown progressively weaker all the way up to 1990. We found only problems with it while the legacy of Mises contains some good ideas.
Hayek lost his debate with Keynes because he saw that “it would be open to us to deal with the difficulties by the aid of higher mathematics” (1967, p. 43) but chose the easy way out instead. When confronted with his confusion between stock and supply, he chose to retain “the simplest assumption of this kind that [one] could make” (1967, p. xi) and, within a year, his followers had abandoned him to embrace Keynes’ General Theory. Since then, the Austrian’s only attempt at a diagrammatical exposition (Garrison 1978) did nothing to improve their reputation. And Skousen can hardly claim to be a better mathematician either. His book (1990) would benefit tremendously if all the hand-drawn APS graphs were replaced with computer-generated printouts of the DWCS, Are-rt, with its mean, 1/r, marked. The values of the parameters A and r should also be clearly labeled so we can see exactly what he thinks changed as a result of the government policies described in the text. All too often it is not clear to the reader (or probably to Skousen himself) if only r changed or if A was also affected.
Sechrest writes, “Graphical analysis can produce definite benefits. It forces one to identify the variables and parameters involved in whatever relationship one is examining” (2001, p. 83). No, it does not. Mathematical analysis does. If an author draws one Hayekian triangle and then another – or even if he superimposes them – how are we to know if the areas underneath them are supposed to be the same or different? Only if he tells us what the areas are. And that means math, not graphs. Graphs illustrate mathematical functions; they are not a substitute for them.32
Garrison writes, “The choice of a linear construction [for the APS] over an exponential one maintains a simplicity of exposition without significant loss in any other relevant regard” (2001, p. 46). This author disagrees. Math is easier when one does it right. If we define the APS to be Are-rt, then the output of consumer goods is C = f(0) = Ar. The output of investment goods, I, is Gross National Output, A, minus C, that is, I = A(1-r). Consider the Production Possibilities Frontier, PPF, with C on one axis and I on the other. Clearly, r = 0 implies that C = 0 and I = A; r =1 implies that C = A and I = 0. So, as the interest rate advances from zero to 100%, we move along the PPF from a situation of all investment and no consumption to the other extreme of all consumption and no investment.
Now let us define the APS as a traditional Hayekian triangle with consumption, C, one leg of the triangle; time since the beginning, B, the other leg of the triangle; gross output, A, the area under the triangle; investment, I, gross output minus consumption, A - C; and the interest rate, r, the slope of the hypotenuse, C/B. Clearly, r = 0 implies that C = 0 and I = A. But what must the interest rate be for C = A and I = 0? One hundred percent? Or some other percentage? I know the answer but, since Garrison has made such an issue of telling us that the choice of a linear construction over an exponential one maintains a simplicity of exposition, I want to see him derive the maximum interest rate for us.
Keen observes “The one barrier which stands in the way of today’s neoclassical economist transmuting into tomorrow’s Austrian is the Austrian insistence that there is little, if any, role for mathematics in economic analysis” (2001, pp. 304-305). He is right. This issue must be addressed if the Austrians are to survive.
The introduction of the DWCS has greatly strengthened the Hayekian position. If they follow through with these ideas they can have a viable theory. In return, I ask that they stop describing their theory as deductive and based on the platitudinous action axiom. The use of words like “deduction” and “axiom” is inappropriate for people who have not provided a clear and concise statement of their postulate set. Kolmogorov had it easy – mathematicians were already familiar with axiomatic systems like geometry, so the only question put to him was whether his particular postulate set was an adequate foundation for the theory of probability. But with Debreu’s breathtaking dismissal of the real world on one hand and Hoppe’s catch-22 on the other, most of my time is spent banishing misconceptions about what “axiomatic” means.33 The first step is a clean split between the Misesians and the Hayekians.
To adopt this division, the economists known as “Austrians” (e.g. Garrison and Skousen) must take the less ethnic name of “Hayekians.” Keynesians do not call themselves “English” and Sraffians do not call themselves “Italian,” so why do Hayekians call themselves “Austrian?” Skousen has recorded (2001, p. 434) that Anna Schwartz refused to contribute to Feminist Economics and, for the same reason, I have grave doubts about identifying economists by their ethnicity. The four leading Austrian-born economists of this century, Mises, Hayek, Strigl and Morgenstern, are associated with four different schools of thought. There is no more an “Austrian” economics than there is a Feminist or a Black economics.
It would be more accurate to consider Menger and Mises forerunners of this author’s Axiomatic School while making Hayek the founder (and Böhm-Bawerk the forerunner) of the Hayekian School. Menger, Mises and this author are the only truly subjectivist economists.34 Böhm-Bawerk’s average period of production demonstrates that he was still mired in the labor theory of value. Hayek’s backwards triangle and his use of the terms “earlier stages” and “later stages” is no better. Skousen gives plenty of lip-service to subjectivism but is belied by his instructions for compiling the APS (1990, pp. 184-185), which depend on remembering the date of an item’s manufacture and when its costs of production were paid. Making everybody save their receipts is not that much different than Böhm-Bawerk’s trying to remember the one hundred working days that were expended in the production of a consumption good.
Thus, for this reason and because Mises’ praxeological method and his regression theorem somewhat inspired this author’s postulate set, I insist on claiming Mises as my forerunner and on asking the economists now called “Austrians” (e.g. Garrison and Skousen) to call themselves Hayekians.
By 1978, the headlong retreat from Mises and the praxeological paradigm had begun in earnest.... Rothbard clearly recognized that the rapid spread of the Lachmannian strand of nihilism and the calculated apotheosizing of the methodologically tolerant Hayek and shunting aside of the allegedly “dogmatic” Mises, which was fostered by the new institutional arrangements, was leading to decay and retrogression.... Lew Rockwell gave us the hard-core institute; Murray Rothbard gave us the hard-core journal. With these institutional means at our disposal we have achieved our goal of putting the modern Austrian revival back on track (2002, p. 121-125).
This is a remarkable thing for the editor of the QJAE to say only a year after devoting an entire issue (Fall 2001) to apotheosizing the methodologically tolerant Garrison and shunting aside Mises’ dogmatic insistence that economics be deduced from a few general axioms. Garrison writes, “Lachmann’s ideas about expectations and the market process served as an inspiration for many of my own arguments... the reader will not fail to notice Hayek’s influence in virtually every chapter – and in virtually every graph – of this book” (2001, p. xiv).35 He refrains from speculating, with Rothbard, on whether this might lead to decay and retrogression.
Salerno writes that Rothbard “seized upon [QJAE] as the main instrument for reclaiming Austrian economics from those who had stripped it of its essential Misesian content in search of acceptance by mainstream economists” (2002, p. 124). I too would like to see the essential vision, though not necessarily the content, of Mises reclaimed. However, I believe that the only way to vindicate Mises’ vision of economics as a deductive science is the discovery of an axiomatic system that actually works. Like mine.36
A word to the wise: There is no such thing as acceptance, only submission. What QJAE contributors need is not to look more like mainstream economists, but a bigger stick to beat them with. If it was not clear in the seventies, it is now: Austrian economics is inadequate. But it is inadequate because the axioms were chosen badly, not because economics cannot be deduced from a few general axioms. While Mises does not deserve the ostracism that he gets in the mainstream journals, neither is he worthy of the adulation that he gets from QJAE contributors. Leadership of an “outsider” organization should go to one who does not value followers who worship him so much as enemies who fear him.
32 If A is independent of r, then the Production Possibilities Frontier, PPF, is a straight line from (0,A) to (A,0). Skousen (1991, pp. 20-27) claims that all investment and no consumption is unrealistic, while Garrison (2001, pp. 40-45) draws the PPF as a slightly convex line from (0,A) to (A,0), but without specifying A(r) such that it decreases as it approaches either extreme. If Skousen wishes to resurrect his 1991 argument, he must make A an explicit function of r and challenge Garrison to justify the shape of his PPF. In the following two paragraphs, since neither man has specified any function A(r), I will assume that A is independent of r and the PPF is a straight line from (0,A) to (A,0).
33 The Random House College Dictionary offers three definitions: “axiom, n, 1. a self-evident truth. 2. a universally accepted principle or rule. 3. Logic, Math, a proposition that is assumed without proof for the sake of studying the consequences that follow from it.” This author (1999, p. ix) employs definition #3. The problem with #1 is that, lacking a “burning bush” experience, nothing ever appears sufficiently self-evident. The problem with #2 is the same one encountered when ordering pizza: Everybody will go hungry if they must wait for universal agreement on which toppings they want.
34 Mises writes eloquently: “Neither acting man himself nor economic theory needs a measure of the time expended in the past for the production of goods available today. They would have no use for such data even if they knew them. Acting man is faced with the problem of how to take best advantage of the avail-able [stock] of goods. He makes his choices in employing each part of this [stock] in such a way as to sat-isfy the most urgent of the not yet satisfied wants. For the achievement of this task he must know the length of the waiting time which separates him from the attainment of the various goals among which he has to choose. As has been pointed out and must be emphasized again, there is no need for him to look backward to the history of the various capital goods available. Acting man counts waiting time and the period of production always from today on. In the same way in which there is no need to know whether more of less labor and material factors of production have been expended in the production of the products available now, there is no need to know whether their production has absorbed more or less time. Things are valued exclusively from the point of view of the services they can render for the satisfaction of future wants. The actual sacrifices made and the time absorbed in their production are beside the point. These things belong in the dead past” (1966, p. 494).
35 One cannot fail to notice the absence of Mises’ influence; check the index for “deduction,” “axiom,” etc..
36 Moise, in a section titled “A Historical Comedy,” has written “Saccheri was dissatisfied with the situation of the parallel postulate; he believed that this statement ought to be proved as a theorem... he undertook to ‘vindicate Euclid of every blemish’ by showing that the parallel postulate was a consequence of the other postulates of synthetic geometry.... The irony is that if Saccheri’s enterprise had succeeded in the way he thought it had, no modern mathematician would have regarded his book as a vindication of Euclid.... In the nineteenth century, two fundamental questions were settled. First, it was shown that the postulates of synthetic geometry, including the parallel postulate, were consistent.... It was shown further that the parallel postulate is independent of the others. This was done, in the only way it could be done, by the discovery of ‘geometries’ in which all the synthetic postulates except the parallel postulate were satisfied. These two developments were the real vindication of Euclid from a modern viewpoint” (1990, pp. 158-159).
There is a somewhat parallel comedy in economics with Mises as Euclid, Rothbard as Saccheri and this author as Lobachevsky. Mises initiated the idea that economics should be deductive but, by 1982, “a lot of younger Austrians... had given up basic Misesian praxeology, that is: that Austrian theory is deduced from a few general axioms” (Salerno, 2002, p. 124). Rothbard vowed to vindicate Mises of every blemish and, towards that end, he (and Hoppe) treated me like a pariah when I showed them my book (1999). Why? Because I had my own postulate set, not Mises’ “action axiom.” Yet it is I, not Rothbard, who vindicated Mises, in the only way that it could be done, by the discovery of a postulate set that actually works.
But Mises was no Euclid. Of Moise’s two questions, the first one failed for Mises because his “action axiom” is just a platitude. Thus, only his vision of economics as a deductive science is vindicated.