Hicks's IS LM Interpretation of the IS-LM Model in J M Keynes's Chapter 15 of the General Theory: Partly Wrong in 1937 and Partly Wrong in 1981

Abstract

Hicks’s 1937 interpretation of Keynes’s IS-LM model in chapter 15 of the General Theory was inferior to Champernowne’s 1936 interpretation which correctly incorporated variables, denoted by Champernowne as Q and Q’, that stood for “The State of the News”, investor nervousness, uncertainty, or the confidence a decision maker had in his expectations. These variables appeared in both the IS and LM functions as independent variables that could shift both curves simultaneously. Only Champernowne’s work in 1936 incorporated these variables. No economist in the time period 1936 -2017 has acknowledged Champernowne’s unique and far sighted accomplishment although his article has been cited repeatedly since 1936. Hicks deliberately removed these variables from his own 1937 paper in Econometrica because it would have prevented him from creating what he called the general, general theory. In 1981, after wasting much of his readers time talking about different time periods and fix prices (money wages) versus flex prices (money wages), Hicks finally admits, at the very end of his 1981 recantation-retraction of his 1937 IS LM paper in an article published in the Journal of Post Keynesian Economics, that the problem with IS LM is that it does not incorporate Keynes’s concerns with uncertainty into the LM curve. Nowhere is the reader told that it was Hicks himself who had deliberately removed the uncertainty variable explicitly when presenting his interpretation of Keynes’s IS – LM curves from chapter 15 of the General Theory. Hicks’s 1981 paper’s citation of Shackle on uncertainty demonstrates that he was still confused and did not understand that Keynes’s definition of uncertainty, on p.148 of the General Theory, was that uncertainty was a unique function of Keynes’s weight of the evidence variable only. Keynes integrated uncertainty explicitly into decision theory with his indeterminate, interval valued probability approach, based on Boole, in chapters 3, 5, 10, 15, 16, 17, 20 and 22 of the A Treatise on Probability, and in his conventional coefficient of weight and risk, c, from chapter 26 of the A Treatise on Probability. Keynes dealt with imprecise probability in chapter 29 with his suggested use of Chebyshev’s Inequality as a lower bound. Since the Liquidity Preference LM curve is built on Keynes’s decision theory from the A Treatise on Probability, Hicks needed to have stated that he had no idea about what Keynes meant by uncertainty in either 1937 or 1981, but that uncertainty, as a function of the weight of the evidence, needed to be reintegrated into the IS LM model in both the IS curve and the LM curve.