7. Great Controversies

This chapter is partially based on Marchionatti (2001). For a survey of the Marshallian debate, see also Marchionatti (2006) and Hart (2012).

In fact, the 1926 article had a “tremendous impact”, as Kahn (1984) wrote, on the Marshallian Cambridge environment, much more than the earlier 1925 Italian article had had on Italian academic circles. In fact, Sraffa came from a country where Marshall’s doctrine was well-known and enjoyed considerable circulation, but not dominant—“Marshall was not Economics” in Italy, where it was subject to criticisms and the limits of partial equilibrium analysis were widely recognized.

Sraffa’s criticism here was not new: as Sraffa himself wrote in the 1925 article, he accepted Enrico Barone’s conclusions (Barone 1894) on the inadmissibility of the increasing supply curve of a commodity for the production of which factors are used which are also employed in other productions.

Sraffa’s proposal to follow Cournot’s approach had been undervalued by economists in the past decades, though not totally neglected: in one of his Economic Essays (1906) on Cournot, Dmitriev (see Vol. I, Chap. 5.4) had maintained that the French economist was the only author capable of constructing a complete theory of competition. More recently, in Italy, the Paretian Luigi Amoroso (Amoroso 1921) had favorably reassessed the analysis Cournot conducted in his Recherches. Sraffa’s positive suggestions had been partly anticipated by Knight, who wrote in his 1921 book that the competitive theory may be a good analytical device, a first approximation, but it does not exist in the real world, and must thus be supplemented to make it more applicable. Similarly, in his 1921 article, Knight wrote: “One of the most serious oversights in the discussion of decreasing cost is the neglect of the mixture of competition and monopoly which is a general characteristic of the type of business supposed to exhibit this type of cost function … The correct approach to the explanation of price in the case of partial monopoly would seem to be to apply the theory of monopoly, not that of competition” (Knight 1921, p. 332). On the other hand, these ideas were not completely new in the United States at that time. First, Moore had noted that “actual industry … to a large extent, is in a state intermediate between perfect monopoly and perfect competition” (Moore 1906, p. 215). Then, Viner (1921) anticipated some of the essential ideas underlying the theory of imperfect competition and J. M. Clark (1923) emphasized that in the contemporary American economic world the majority of markets lay in the intermediate zone between theoretical competition and theoretical monopoly. Some years later, in 1927, Chamberlin laid the foundation of The Theory of Monopolistic Competition in his dissertation at Harvard (under Young’s supervision).

Pigou (1928) holds that the firm’s expansion is offset by an exactly equal contraction on the part of the other firms in the industry, whereas for Viner (1931) expansion in the industry takes place as a result of an increase in the number of firms. As Chipman (1970) demonstrated, these are artificial assumptions which need not be introduced.

Shove’s suggestions were ignored at that time. His statement was made again 25 years later by J. N. Wolfe (1954): in his article, inspired by Robertson’s and Shove’s contributions, the American economist suggested that the representative firm should be treated as an “abbreviated notation” for the distribution of firms by size, thus making the result a process of chance. This concept of equilibrium can be represented mathematically, by employing, as Newman and Wolfe (1961) do, the technique of non-homogeneous Markov chains. The rehabilitation of Marshall’s life cycle theory by Negishi (1989) can be also considered a variation of the theme sounded by Newman and Wolfe.

Young’s view was to be theoretically fertile: its influence can be seen in Stigler (1951), Downie (1958), Richardson (1975), and Kaldor (1981) among others, and to some extent in Romer (1986, 1987).

We should also mention Friedrich von Wieser, who in Der Natürliche Werth (1889) maintained that the method of value calculation is essentially the same under communism and under capitalism.

Hutchison (1953) cites the criticism of socialist economics by the German liberal-social-reformist Albert Schäffle (1831–1903). He notes that Schäffle in some works—principally Die Quintessenz des Sozialismus (1874) and Bau und Leben des sozialen Körpers (1881)—on the one hand emphasized the dangers to freedom of choice in goods and jobs in a collectivist economy (a judgment also stressed by the member of the historical school Lujo Brentano) and on the other hand “tentatively sketched out a scheme for a liberal-socialist economy with a centralized authority, plentifully supplied with statistical intelligence, and manipulating a system of taxes and subsidies, which would retain the freedom and stimuli of the competitive market. Schaeffle’s idea was essentially that later developed by Barone, Lange, and others, but without the precise mathematical formulation and the background of Walrasian and Paretian analysis” (p. 294).

Eminent figure of German society in the tragic years between the war and the early postwar period, Walther Rathenau, industrialist, writer and politician, was one of the main organizers of the German economy during the war, and after the war was minister of reconstruction and then foreign minister of the Weimar republic. He was assassinated in 1922 by members of a right-wing movement.

Hayek (1935) claimed that Max Weber had also arrived independently at the same conclusion in the second chapter of Wirtschaft und Gesellschaft, which was published posthumously in 1921: “Like Professor Mises, [Max Weber] insisted that the in natura calculations proposed by the leading advocates of a planned economy could not provide a rational solution” (Hayek 1935, p. 33). In fact, as Hutchison (1953) writes, Weber advanced an analysis similar to Mises at some points, “but quite different”: “Weber agrees that ‘monetary calculations’ ensure, and are necessary for, a certain purely formal rationality in economic activity, but points out that this formal rationality may, and usually will, conflict with a substantive rationality in the direction of economic activity” (Hutchison 1953, p. 301).

Two important contributions also appeared in Germany in these years: Carl Landauer’s Planwirtschaft und Verkehrswirtschaft (Planned Economy and Exchange Economy) (1931) and Cläre Tisch’s doctoral dissertation Wirtschaftsrechnung und Verteilung im zentralistisch organisierten sozialistischen Gemeinwesen (Economic calculation and distribution in the centrally organized socialist community (1932), discussed at the university of Bonn with Schumpeter as tutor). Carl Landauer (1891–1983) was born in Munich and received his doctorate at Heidelberg. Member of the Social Democratic Party from 1912, he was Professor at the Commercial University in Berlin up to 1933. He then left Germany and joined the Department of Economics at Berkeley. Cläre Tisch (1907–1941) studied economics in Bonn and Berlin and became Spiethoff’s assistant, but she was forced to leave university under Nazism and in 1941 was deported to the Minsk ghetto where she was murdered. Landauer and Tisch’s works were contributions to the theory of market socialism. As regards Tisch’s dissertation in particular, Hagemann (2019) considers it an important contribution, as it dealt with the socialist calculation problem in terms of the Walrasian general equilibrium system in Gustav Cassel’s simplified version.

See Caldwell (1997) for a survey of Hayek’s life-long contributions on socialism.

It is worth mentioning that Lionel Robbins had anticipated many of Hayek’s comments one year before in Chapter seven of his The Great Depression (Robbins 1934), criticizing the American New Deal. This reveals that the place where these new Austrian positions developed was, not surprisingly, the LSE (see Chap. 3).

Hayek also cites W. Crosby Roper Jr’s The Problem of Pricing in a Socialist State (1931), but this book is simply a summary of the debate until then.

As Schumpeter notes, in the formulation of his theory of the socialist market, Lange (together with Lerner) is the representative of a phenomenon that had developed since the 1920s, viz. that “the scientifically trained socialist is no longer a Marxist except in matters of economic sociology” (Schumpeter 1954, p. 884). He adds that Lange made the position in question very clear in his paper on “Marxian Economics and Modern Economic Theory” published in the Review of Economic Studies (Lange 1935).

The chapter is partly based on Garrone and Marchionatti (2007) and Garrone and Marchionatti (2009).

Keynes’s critique of Tinbergen’s first League of Nations study is considered to have sparked the debate about the role of econometrics (Hendry and Morgan 1995). Keynes's criticism remains controversial, but the long-prevailing view is that Keynes was an a priori anti-econometrician. Samuelson (1946) maintained that he was technically incompetent. Klein (1951, p. 450) called his review “one of his sorriest professional performances”. Stone (1978) maintained that Keynes had little or no awareness of the econometric literature. He wrote that his review was “a model of testiness and perverseness” (Stone 1978, p. 61), largely because of his temperament.Since the end of the 1970s, the relevance of Keynes’s criticism has been acknowledged. It was Patinkin (1976) who first found it “somewhat depressing to see how many of [Keynes’s criticisms of the use of correlation analysis to estimate equations] are, in practice, still of relevance today” (Patinkin 1976, p. 1095). Hendry (1980, p. 396) wrote that “[Keynes’s] objections make an excellent list of what might be called problems of the linear regression model”. Pesaran and Smith (1985) and Rowley (1988) went beyond the simple consideration of technical issues. The former scholars recognized that Keynes was right on both the technical and logical arguments: on the technical level, “the problems that Keynes raised were real, despite his occasional technical confusion” (Pesaran and Smith 1985, p. 144). At a logical level, “econometric inference … is insupportable” (ibid., p. 147). Rowley (1988, p. 25) maintained that “Keynes’ criticisms have been diluted, forgotten or mis-stated rather than absorbed into the prevalent orthodoxy”. He regretted that “we have waited too long for econometric methodology to come of age and address its logical bases” (ibid., p. 30). Actually, it is in this wider context that Keynes has been considered in the 1990s. McAleer (1994, p. 332) writes that some of Keynes’s criticisms of Tinbergen’s econometric methodology “remain relevant to this day” and that his implicit research program “subsequently led to the development of numerous econometric techniques that are now widely used in applied econometrics” (ibid., p. 334) (see also Dharmapala and McAleer 1996). Similarly, Keuzenkamp (2000) maintains that Keynes’s skeptical attitude remains substantially justified. Summing up, it is now recognized that Keynes’s criticism of Tinbergen was sound in many points. However, it is considered overly harsh and Keynes is blamed for throwing the baby out with the bath water.

Keynes received a proof copy of Tinbergen’s book from the American economist and diplomat Royall Tyler (1884–1953), expert in the Economic and Financial Section of the League of Nations in Geneva, in order that he might comment on it.

In the same year, at the invitation of the editors of the Review of Economic Studies, Tinbergen also wrote a paper “to go into some more detail concerning the method” of analysis (Tinbergen 1940b). It offers a restatement of the method and integrates Tinbergen’s reply to Keynes. In particular, Tinbergen emphasizes his method’s flexibility.

On some points Keynes’s critique shows his limited knowledge of the developments of the econometric literature in the previous two decades (despite the fact that Keynes had been on the editorial board of Econometrica since 1933) and a few misunderstandings of technical issues. This fact is well known and widely emphasized (see, e.g., Hendry and Morgan 1995).

Keynes also cites the inadequacy of statistics—an “obvious” difficulty: “These many doubts are superimposed on the frightful inadequacy of most of the statistics employed, a difficulty so obvious and so inevitable that it is scarcely worthwhile to dwell on it” (Keynes 1973c, p. 317).

The question of whether testing can prove a theory to be correct is not controversial. Both Keynes and Tinbergen agree that testing cannot prove the correctness of a theory, whatsoever amount of empirical evidence is available. As noted in Keuzenkamp (1995), the idea that scientists cannot prove a theory but may be able to falsify it was “a common sense notion in the statistical literature since (at least) the turn of this century” (Keuzenkamp 1995, p. 240).

Pesaran and Smith (1985) re-estimated some of Tinbergen’s relations by the OLS method using the original undetrended series. (Their purpose was to examine the effect of de-trending on Tinbergen’s results.) They found that “the un-detrended OLS results suffer from a significant degree of residual autocorrelation which sheds considerable doubt on the size and the statistical significance of the estimated regression coefficients … The method of de-trending employed by Tinbergen can, and often does, deal with the problem of residual autocorrelation. But … its application can also introduce erroneous dynamics into the relation and its residuals” (Pesaran and Smith 1985, pp. 141–142). The presence of residual autocorrelation can be due to the factors stressed by Keynes: omitted variables, functional form misspecification, structural change and a host of other factors “all of which are highlighted in Keynes’s review” (p. 143).

For an analysis of the early reception of Keynes’s criticism by the econometricians, see Garrone and Marchionatti (2009).

The correspondence with Roy Harrod is published in Keynes’s Collected Writings; the (un-published) correspondence with Ragnar Frisch, concentrated in the period 1932–1936, is kept at the National Library of Norway, Oslo Division. A discussion of the latter is in Louçã (1999).

Tinbergen himself, reviewing the recent business cycle theories in 1935 (Tinbergen 1935), devoted great attention to the parts of Keynes’s Treatise of Money “which give very pertinent remarks on the business cycle problem” (p. 266). Tinbergen classifies Keynes’s theory as a semi-mathematical one and argues for its mathematical treatment.