[ Pobierz całość w formacie PDF ]
.On this matter he cited Stigler s (1945) paper  The cost of subsistence.One canhear an echo of Frank Knight in Friedman s statement that  real wants are for morewants.We see all around us the extent to which we live to work (Friedman,  LectureNotes  Price Theory  , undated, Milton Friedman Papers, Box 76).One can also see elements of Friedman s (1953)  The methodology of positive econom-ics in his lectures for Economics 300A, when he makes the distinction between positiveand normative economics, and also in discussion of the nature of theory as language,that is, theory as a set of filing boxes.Demand and supply are filing boxes for the variousfactors that influence willingness to buy and to sell.The appropriateness of the theoreti-cal framework, that is, filing system, is determined by the extent to which the affectingfactors can be sorted into one box or the other.The behavior of dealers in commodities,for example, does not sort cleanly into demand and supply, for dealers buy in order tosell.A second dimension of the filing boxes is separating short- from long-run phenom-ena, and this distinction is not one that can be made once and for all.Rather, sorting intothe short- and long-run boxes depends on the problem at hand. 16 The Elgar companion to the Chicago School of EconomicsFriedman and Stigler revisitedFriedman and Stigler s deepening friendship left its mark not only on his price theorycourse but on revisions of Stigler s textbook and on the scholarship of both men.Thiscan be seen in their correspondence.For example, we have already referred to the letterwhere Friedman contrasted Stigler s with Marshall s proofs of the Law of DiminishingReturns (or Law of Variable Proportions).Stigler s text discussed two types of proofsfor the law.He labeled the first type a priori and defined it as  attempt[s] to deduce thelaw from self-evident propositions (Stigler 1946, p.118).He labeled the second type ofproof  empirical: no one has discovered any important exceptions to the doctrine (ibid.,p.120).He found a priori proofs unsatisfactory, and noted that they are  essentiallytautological.In arguing in favor of the  empirical line of proof he referred readers toexamples of  quantitative verification in numerous agricultural experiments, which, forexample, showed that applying increasing amounts of fertilizer to a fixed plot of landleads to diminishing returns (ibid., p.120).Friedman considered these statements to be an attack on Marshall s method ofproving the law.Paraphrasing Marshall:If the law of diminishing returns were not valid; that is, if the application of additional units oflabor and capital to a piece of land yielded constant or increasing returns, then individuals wouldhave no incentive to get additional land and we should observe in fact that individuals used andwanted very little.(MF to GS, August 12, 1946, Hammond and Hammond 2006, p.23)Friedman makes his point:Stigler says, in effect, that Marshall is guilty of  question-begging , that his  and similar proofsare essentially tautological ; yet Marshall sounds anything but tautological, he sounds realisticand as if he were basing his results on sound observation.As nearly as I can figure it out, Stiglerhas a sound point; but with little trouble Marshall can be rehabilitated, and, when he is, is farmore convincing than Stigler.(Ibid., p.23)Friedman proceeded to  rehabilitate Marshall by setting up alternative hypothesesunder different assumptions with respect to increasing and decreasing returns to variableproportions.He tested each hypothesis against observed fact, and drew his conclusionsusing the following method,  Our hypothesis leads us to expect a certain result, we findthat result, hence our hypothesis is not contradicted (ibid., p.23).Friedman then posed a rhetorical question:You may ask, why all this fuss when Stigler accepts the law on other grounds, namely, techno-logical experiments [which Stigler had labeled empirical proofs].The reason is that economicempirical evidence of the kind given by Marshall [that farmers are willing to pay for more land]is intellectually far more satisfying to an economist than technological evidence.In addition,part of my purpose is to show that Marshall here as elsewhere, was proceeding on a truly sci-entific basis, not on that tautological, formal basis that enervates so much of modern theory.(Ibid., pp.24 5, original italic)Stigler replied:You say that economic empirical evidence is intellectually far more satisfying than technologicalevidence.I cannot claim even an intuitive understanding of this statement.Diminishing returnsis technological, so you prefer an indirect to a direct proof: The development of post-war Chicago price theory 171.Because you are freed of dependence on non-economic data? Perhaps, but this is clearly amove in the direction of a closed formal system  which you don t like.2.Because it is more efficient? That depends on the case.3 [ Pobierz całość w formacie PDF ]