# R. Lucas's Research Program: Far Too Narrow with Too Many Special Assumptions

42 Pages Posted: 26 Sep 2017 Last revised: 18 Oct 2017

See all articles by Michael Emmett Brady

California State University, Dominguez Hills

Date Written: September 24, 2017

### Abstract

R. Lucas’s research program is based on advancing Jeremy Bentham’s research program of using optimization theory at the micro level (Maximize Utility and Maximize Profits) to describe outcomes at the macro level. No further analysis is needed. There is no macroeconomics for Bentham. de Vroey’s new book, "A History of Macroeconomics from Keynes to Lucas and Beyond", provides valuable evidence of what happens when a “modern” economist, who is wedded to Benthamite Utilitarian ethics and has no idea about the technical, theoretical connections between Keynes’s 1921 A Treatise on Probability and his 1936 General Theory, attempts to evaluate Keynes’s contribution in the General Theory. An earlier contribution in this field with similar problems was published by Giles Dostaler in 2007 as" Keynes and his Battles."

Five points will be covered in evaluating Lucas’s research program.

First, Lucas is simply ignorant of the technical analysis done by Keynes in Part II of the A Treatise on Probability, which has never been read by any macroeconomist, which established that indeterminate and imprecise probabilities (Keynes used the term uncertainty to stand for indeterminate probabilities in the General Theory. Uncertainty was defined to be an inverse function of the weight of the evidence) are the general case and precise probability is a special case only. A modified version of Boole’s 1854 The Laws of Thought upper and lower probability approach was used by Keynes to develop an interval valued probability approach to cases where the weight of the evidence, w, defined on the unit interval from 0 to 1, where a w=1 means that there is no uncertainty, was less than 1. A weight of 1 is required before a decision maker can used numerically precise probabilities or claim that he knows what the probability distributions are, be they objective or subjective.

Keynes’s indeterminate probabilities, as well as his imprecise probabilities approach developed in Part V of the A Treatise on Probability, are non (sub) additive in nature. They do not sum to 1 as do all of the precise probabilities in all of the articles and models published by Muth, Lucas, Sargeant, and Real Business Cycle theorists, etc.

Keynes also incorporated non linear probability preferences. Keynes’s conventional coefficient of weight and risk, c, the first decision weight approach ever proposed in history for decision making use, demonstrates that Keynes’s approach, using imprecise and indeterminate probabilities to model uncertainty while using precise probability to model risk, is the general case, while Lucas’s rational expectations claim, that decision makers know all of the relevant probability distributions over time, is a very special case only of Keynes’s general theory of decision making. For instance, De Vroey never mentions anywhere in Part I of his new book what the decision theoretic foundations are for the General Theory. The foundation is Keynes’s interval valued probabilities developed in Part II of the A Treatise on Probability.

Second, Keynes, not Hicks, Meade, Harrod, Lange, Reddaway, Champernowne, Hansen or Modigliani, is the originator and developer of the IS-LM model, which he would refer to as the IS-LP (LM) model, where LP stood for liquidity preference.It does not make any sense to talk about IS-LM, as de Vroey does in chapters one to four of his 2016 book, if the reader is not informed that Keynes’s entire IS-LP (LM) model is provided in chapter 21 in Section IV on pp.298-306. The only original contribution made by Hicks in 1937 was to draw the IS-LM (LL) diagram.

Third, Keynes made in clear in Chapter Two of the General Theory that the existence of involuntary unemployment will show up in the Labor market as a disequilibrium. It shows up in the Output market as one of a number of possible multiple equilibria, only one of which can be a full employment equilibrium. The mathematics is in chapter 20. A simpler version is contained in footnote two on pp. 55-56 where Keynes develops the Aggregate Supply Curve, which is a locus of all possible equilibriums.

Fourth, Keynes pointed out to both Harrod and Hicks that the existence of a Keynes – Neoclassical synthesis was possible because his theory merged with the Neoclassical theory once a full employment, equilibrium level of output had been attained. Neither Hicks nor Harrod responded to Keynes in his lifetime.

Finally, all of the work done by Lucas and his associates in the Real Business cycle approach assumes rational expectations. Rational expectations requires that the decision makers know all of the relevant probability distributions over time before any decision has to be made. This was the same type of claim made by Jeremy Bentham in 1787.

The main difference between Bentham’s assumption of rationality through the maximization of utility with known, additive uncertainties (probabilities) about the future consequences of all actions and Lucas’s rational expectations is that Bentham argued that all individual uncertainties were known and additive, but he did not talk about known probability distributions. A huge body of experimental evidence, going back to Preston and Beretta, Ward Edwards, Herbert Simon, Daniel Ellsberg and Amos Tversky and Daniel Kahneman, shows that there is simply no empirical, experimental, or positive evidence to support basing macroeconomics on the supposed ability of consumers and producers to make and base all of their decisions over time using precise probability.

Keywords: Lucas, Keynes, Bentham, rationality, rational expectations, Samuelson, optimization theory Max U, Max \$

JEL Classification: B10, B12, B14, B16, B20, B22

Suggested Citation

Brady, Michael Emmett, R. Lucas's Research Program: Far Too Narrow with Too Many Special Assumptions (September 24, 2017). Available at SSRN: https://ssrn.com/abstract=3042089 or http://dx.doi.org/10.2139/ssrn.3042089