## The betrayal of the intellectual

from **Robert Locke **

It is very difficult for historians to establish any set of ideas when confronted by people who are historically ignorant. Recently I wrote, in a the RWER Blog that economics has never established a scientific paradigm in its discipline. So the idea that neoclassical economics did and that it is now being challenged is false. It never did achieve paradigm status. Twenty-five years ago (1989) I said the same thing in the first two chapters of Management and Higher Education Since 1940 (Cambridge University Press). In the first chapter, “The New Paradigm,” I wrote about the attempt

“It is hardly a coincidence…that the rise of mathematical economics accompanied the development of marginalism. Stanley Jevons..used mathematics. He was the first consciously to seek, according to Leon Walras,”to apply mathematics to economic theory.” Walras’ own Elements d’economie pure, which “presented the ensemble of economic theory as essentially a mathematical theory, where are the important propositions could be stated in equations” was published in 1874. (His achievement a system of simultaneous equations which expressed general equilibrium theory, has been hailed as the great accomplishment of modern economics (Nogaro). He, Alfred Pareto, and a handful of others were the pioneers; literary economics faded into the background.Neoclassical economics, however, was still of marginal interest…because the mathematics it employed proved to be inappropriate to the entrepreneur’s problems. (pp.14-15). … Because calculus could not deal realistically with the “allocation problem,” Erich Schneider, that great admirer of the neoclassical mathematical economists, admitted “is not to be solved with the method of infinitesimal calculus. Subsequently, John von Neumann and Oskar Morgenstern observed in their famous book Theory of Games and Economic Behavior (1944), “the concepts of economics are fuzzy but even in those parts of economics where the descriptive problem has been handled more satisfactorily, mathematical tools have seldom been used appropriately. Mathematical economics has not achieved very much.” (16). Since this came from a great twentieth century mathematician, it was a serious rebuff. Neo-classical economics was, from a decision-making perspective, not relevant.

… [T]he economists required an algorithm that could make their theories practice relevant. They found it after the war principally in linear programming. George B. Dantzig and his Rand Corporation associates developed the simplex linear programming algorithm for the United States Air Force (18)…. The mathematics it required was…if not new to mathematicians, natural scientists, and engineers, so unfamiliar to economists that when explaining the new algorithm to their colleagues, economists included special sections on vectors and matrices in their books. (EX. Robert Dorfman, Paul A. Samuelson, and Robert M. Solow,(1958) Linear Programming and Economic Analysis (New York) Economists were drawn to linear programming…because it did not undermine the body of economic theory which had been built up so painfully during the twentieth century. Linear programming was, the economists insisted , just a special case of marginal analysis and, hence, quite compatible with neoclassical economic theory. (Schneider …, 454) p. 19

Although the most important algorithm, linear programming was but one of the new activity analysis algorithms to which economists were attracted. Leontief’s input-output algorithm, first formulated in 1938 but not given a full exposition until 1941, provided an analytic framework for determining the effect in the economy of altering production goals, or the feasibility of a production program given available resources and possible methods of production. The Leontief model, which dealt with national economics and with industrial sectors, was of limited use at the level of the firm. Moreover, it did not allow for alternative production mixes to be taken into consideration, as did linear programming. But it did provide economists with an operationally oriented system. Even more important in this respect was the development of game theory. The mathematician Emile Borel introduced it in 1921 in order to make predictions about games of chance. Although von Neumann developed further the modern approach to problems of competitition and cooperation in the “Theory of Parlour Games” (Zur Theorie der Gesellschaftsspiele,” Mathematische Annalen 100 (1928): 295-320), the economic implications of game theory were not spelled out until von Neumann and the economist Oskar Morgensterm published Theory of Games and Economic Behavior in 1944. Game theory drew a straight line from modern mathematics (because von Neumann used algebra, matrix, theory and probability theory in his calculations), through economics, to management to show how entrepreneurs in conflict situations, could, under certain assumptions, act so as to be guaranteed at least a certain minimum gain (or maximum loss) by following the algorithm. Dorfman, Samuelson, Solow.

Thus…linear programming, topology, and probability theory replaced the diagrams of the distant past and the calculus of more recent times to give scientific status to economic knowledge . (p.20) … There was an excitement in the air, after the war. … Postwar economists for the first time could talk about actually managing an entire economy instead of just aspiring to understand the principles of eco (p.21) nomic behavior. With the same tools, the same ambitions were extended from the macroeconomic policy sciences to the level of the firm. And the confidence exuded in America spread into Europe. German economists and business economists, for instance, voiced the general optimism at a two-day joint meeting, June 18, 1963:

‘In the last three decades a tranformation in the work methods of the economic sciences is clearly recognizable. An analysis of the economic structure and the behavior of its components, which can be characterized accurately as natural science, has replaced the old backward looking descriptive and intellectually conceived perspectives. In all branches the theory of economic circulation has led to important insights. A complete and purposefully oriented statistical science and the adoption of mathematical methods from engineering have afforded us new possibilities to verify our working hypotheses.’ Der Volkswirt (1963), 17:23, 131-32.”

The second chapter of the 1989 book, entitled “The New Paradigm Revisited” begins

“Yet there were critics right from the very beginning. Among them were the institutionalists from the old descriptive school who distrusted the mathematicians.” (p. 30). The doubts, moreover, crept up among those whose very occupational raison d’etre arose from the creation of the new paradigm. As the 1960s turned into the 1970s, the opposition even within OR [whose techniques so influenced economists] was neither exceptional nor inconsequential. … (p. 31) In 1981 Dando and Sharp evaluated the mood of operations research as reflected in the pages of the Journal of the Operational Research Society by looking at the issues published in 1963, 1968, 1973, and 1978. Up to 1968 when “optimism about the future of OR” reigned, there was “almost a total lack of criticism and debate in the journal.” (p. 31). In 1973 the papers reflected considerable doubt about the practical effectiveness of OR, a doubt which by 1978 was being voiced in about one quarter of the major papers appearing in the journal. The essays of the late 1970s were, therefore, a culmination of a decade of ever-increasing and deepening criticism at the very center of the new paradigm. … OR models, can never be a “perfect representation of a problem,’ Russell Ackoff observed. They leave out the human dimension, the motivational one indeed Ackoff an OR pioneer complained that the successful treatment of managerial problems deserves “the application not only of science with a capital S but also all the arts and humanities.” (“The Future of Operational Research is Past, JORS 30 (1979), 93-104), p. 34.

Finally, disappointing OR results have been obtained from macro-economic analysis. Anybody familiar with the popular as well as the semischolarly and scholarly press knows about the lost credibility of the economists. “Twenty-five years ago,” Robert Kuttner observed in 1985, “the age old problem of boom and bust seemed to have been solved. [But, s]ince 1970 an outpouring of serious and ideologically diverse articles and books has pronounced that economics is in a state of severe, perhaps terminal, crisis. (“The Poverty of Economics,” Atlantic Monthly 255 (Feb. 1985), 74-80, 75.) …[Kuttner] was concerned…with the failure of mathematized econometric models to provide fruitful policy guidance. The era of macroeconomic prognostication got under way in the mid1950s. The models were relatively successful as long as the future resembled the past, but such a requirement for success hardly inspires confidence. It means that models are useful when they are not of any particular interest, as long as things remain the same. But when the future does not resemble the past, spectacular failure to predict rates of economic growth, business profitability, inflation rates, private consumption (p. 35) levels, employment, etc., can and have resulted. Even if models did predict successfully, they were not instructive because the few that did coexisted with many that did not and there was no way to know in advance which prognosis would be correct. The conclusions of W. Friedrichs and K. Kuebler about the reliability of German econometric models seem to apply to the entire macroeconomic exercise in model building: “Neither the econometric, nor the naive prognosis, nor the judgmental forecasts could satisfactorily predict future economic development.” “Die Leistungsfaehigkeit oekonomistrische Prognose-System,” Operations Research Verfahrens 26 (1977); 814-26.)

The solutions that the new mathematics provided, to replace those produced by the insufficiencies of the mathematics employed by neoclassical economists, have proved themselves, from the macroeconomic management perspective, to be inadequate. Indeed, a well-known economist Kenneth Boulding, has called the whole mathematical enterprise in economics a mistake. “Perhaps the real villain,” he wrote “is the discovery of seventeenth century mathematics some two hundred years later by Cournot, Jevons, and most of all Walras, whose influence and brilliance set economics on a path that increasingly has become a dead end.’” “What Went Wrong with Economics.” The American Economist 30:1 (Spring 1986): 5-12.” (p. 35)

The point about the Chapter “The New Paradigm Revisited” is not that literary and institutional economics resisted the new paradigm, but that the scientific-mathematical toolkit economists imbibed from statisticians, engineers, scientists, and operations researchers, which was integral to the new paradigm, could in the end not be used to justify neo-classical claims to have established a science. I wrote the chapter more than a quarter century ago; since then there is no evidence that the usefulness of the scientific toolkit has improved – witness the fiasco of model-building in financial economics that Mandelbrot and Taleb tried vainly to get the professors to understand before their models induced financial collapse. (Locke & Spender, 169-70)

In his 1927 book, *The Treason of the Intellectuals*,[1] Julien Benda observed that “from the time of the pre-Socratics, intellectuals, considered in their role *as* intellectuals, had been a breed apart. Thanks to such men, he wrote, “humanity did evil for two thousand years, but honored good. This contradiction was an honor to the human species, and formed the rift whereby civilization slipped into the world.” Mainstream economics’ failure to discuss the shortcoming of their discipline dishonors it and blackens the reputation of the universities in which they reside.

Bertrand Nogaro, “Questions theoretiques: Les mathematics considerees comme logique formel et la mise en equation des problemes economiques,” Revue d’economic politique, 54, 1940, 467-83.

Robert R Locke and J.-C. Spender. (2011) Confronting Managerialism. (London: Zed).

Erich Schneider, “Der Weg der Betriebswirtschaftslehre in den Letzen 25 Jahren,” in Volkswirtschaft und Betriebswirtschaft. 452-57.

Agree that we have no scientific paradigm. But we need to construct one.

Neoclassical economics is the prevailing paradigm because it is in basic textbooks, presumed to be consensual, core economic knowledge, used to discuss and applied to form economic policy.

The paradigm is scientific in appearance only, with mathematics, statistics etc, but not in substance. To say that the neoclassical paradigm, the Keynesian paradigm, or for that matter, various versions of their hybrid are not scientific, it is necessary to define what is science:

http://www.asepp.com/what-is-science/