Keynes and the Neoclassical Synthesis

Einsteinian versus Newtonian Macroeconomics

Dario Togati

• Publisher: Taylor & Francis
• Year: 1998
• Print ISBN: 9780415183963
• eBook ISBN: 9781134670284
• Edition: 1
• Topic: Business
• Subtopic: Business generale
• Index: Business

This remarkable volume provides a critical assessment of Neoclassical Synthesis, long regarded as the standard interpretation of Keynes. Taking issue with this orthodoxy, the author offers a unique interpretation of the foundation of modern macroeconomics, arguing that the subject derives from the conflict between two research programmes inspired b

Part I

THE TWO BASIC PARADIGMS OF MACROECONOMICS

1

HICKS’S VALUE AND CAPITAL

Hicks’s work in the 1930s opens the way to the basic result of the Neoclassical Synthesis: namely, the reduction of the General Theory to a special case of Classical theory. Hicks reaches the same conclusion in all his major works during this period:

• 1 In ‘Mr. Keynes and the “Classics”’ (1937), where the General Theory is linked to the special case of the liquidity trap.
• 2 In Value and Capital (1939), where Keynes’s analysis is presented as a special case of his temporary equilibrium model with fixed money wages.
• 3 In ‘A Suggestion for Simplifying the Theory of Money’ (1935), which provides the background for his later view that Keynes’s liquidity preference theory is a special case of Cambridge theory and may be analysed using the tools of Neoclassical value theory.

Hicks, arguably, draws this conclusion quite simply because he relies on the general equilibrium method, which leads him to rule out the crucial features of Keynes’s approach—such as the use of aggregates and expectations as autonomous variables—which account for the autonomy of his macroeconomics. My justification for basing my analysis on Value and Capital is thus relatively straightforward. In this book Hicks provides the most complete description of a full-blown general equilibrium approach to macroeconomics—what he calls ‘pure theory’. As I stressed in the Introduction, Hicks’s approach would appear to be inspired by the mechanistic model. To grasp this concept fully it might be useful to refer to Einstein’s precise description of the model in the book he wrote in collaboration with Infeld:¹

The great results of classical mechanics suggest that the mechanical view can be consistently applied to all branches of physics, that all phenomena can be explained by the action of forces representing either attraction or repulsion, depending only upon distance and acting between unchangeable particles.

(Einstein and Infeld 1938:67)

Here Einstein stresses three basic features of the standard model:

• • the existence of unchangeable particles (atoms) whose position and speed can be accurately stated;²
• • the view that these particles constitute the simplest elements to which all physical phenomena should be reduced (the compositive or constructive method);
• • the existence of simple forces acting directly between the particles (direct forces).

I would suggest that each of these features finds a precise counterpart in the method of Value and Capital.³ More specifically, it is possible to establish analogies between:

• • the particles of classical physics and the individual agents of ‘pure theory’ (for a similar analogy see Fisher 1926:85–6; Hodgson 1993:234–5; Lawson 1997:90);
• • the constructive method used in the mechanistic model and that adopted by Hicks;
• • direct forces, such as Newton’s gravitational attraction force and the universal properties of human nature, such as the self-interest which Hicks emphasizes (for a similar analogy in Adam Smith’s contribution, see e.g. Deane 1978:11; Gordon 1989:133).

Before going into these analogies in detail, two points need to be emphasized. The first is that, insofar as it is the clearest expression of the Newtonian paradigm in macroeconomics, Hicks’s ‘pure theory’ provides the benchmark for our analysis.⁴ The second is that I am fully aware that pure theory as we normally understand it does not correspond to macroeconomics. It includes, for example, microeconomics and a full-blown disaggregated general equilibrium model with n firms and consumers. However, Hicks claims that in Value and Capital a bridge exists between this standard notion of pure theory and macroeconomics (or what I refer to in this book as ‘pure theory’). As we shall see, this bridge is represented by the device of the representative agent, which allows a smooth transition from micro- to macroeconomics.

Atomism

Analogies between Hicks’s ‘pure theory’ and the mechanistic model

It is possible to underline a few significant analogies between the atomism underlying the mechanistic model and that underlying Hicks’s ‘pure theory’. In the first place, just as the mechanistic model defines the position and speed of individual particles, so Value and Capital is built on axioms which concern the behaviour of individual agents. A close link exists between Hicks’s axiomatic approach and that of Hayek and Robbins, who were with him at the LSE in the early 1930s. In the first part of Value and Capital, Hicks develops what Hayek (1937) calls the ‘pure logic of choice’, dealing with the basic principles of standard value theory and the maximizing behaviour of consumers and firms. According to Hayek, insofar as it consists of deductions from a set of self-evident propositions based on universal facts of experience derivable from simple introspection—for instance, that people have preferences and can arrange them in order—this is the truly a priori part of his economic analysis. As a set of self-evident propositions, the pure logic of choice is only subject to internal consistency and is not directly applicable to the explanation of real-world phenomena. As Hayek points out:

The logic of choice applies only to persons who have to allot limited means among a variety of ends. And for the field so defined our propositions would again become a priori true. But for such a procedure we should lack the justification which consists in the assumption that the situation in the real world is similar to what we assume it to be.

(Hayek 1937:46)⁵

Let us now turn to the second analogy between the two models. Just as the atoms of classical physics are independent from the ‘environment’, in the sense that there is no interaction between matter and space (the latter has autonomous existence with respect to the former, see e.g. Einstein 1952: 310), Hicks regards agents’ preferences as being private in nature and autonomous or context-independent elements. This view accounts for the conception of rationality which underlies the logic of choice. This logic implies that the individual agent follows the canons of ‘perfect’ or ‘absolute’ rationality: no problems of imperfect knowledge and computational ability can ever arise in the rational calculation or the adaptation of means to ends due to the tautological nature of the analysis which rules out a priori contamination with realworld phenomena.

A third analogy between the atomism of the classical model and Hicks ultimately encompasses essentialism. As Popper (1972:167–8) notes, this is nothing other than the view that atoms are unchanging bits of matter which represent a fundamental level of reality and the basis of any ultimate explanation. It is important to stress that this view implies a clear-cut distinction between the primary and secondary qualities of particles as well as large bodies. Primary qualities are extension, persistence, impenetrability and so on, which constitute the only proper object of analysis.⁶

Hicks’s reference to the standard concept of homo economicus also, arguably, implies a sort of essentialist view. Accepting the principles of methodological individualism, he regards the individual agent, characterized by unchanging preferences and purposes, as the bedrock, the very foundation upon which economics has to be built (e.g. Hodgson 1989:53). Just like atomism in physics, Hicks too relies on a clear-cut distinction between the primary and secondary qualities of agents. For example, in his timeless logic of choice, he focuses exclusively on what can be regarded as the primary quality of individual agents—namely, the pursuit of self-interest governed by context-independent preferences in conditions of certainty and stability— thus neglecting the other (secondary) qualities of agents, such as the ones typical of a ‘real-world’ context (for example, the need to form expectations and interact with other agents).

In my view, Hicks’s retention of essentialism is not called into question by the fact that he, like Pareto and Robbins, develops the ordinalist approach to consumer behaviour, attempting to separate the axioms of individual rationality from the concept of cardinal utility which underlies the first versions of marginalist theory.⁷ This ordinalist move is, admittedly, to some extent in contrast with the old paradigm. Indeed, by relying on the axiomatic method and dropping the reference to the metaphysical concept of utility as a substance or essence of value, Hicks is in tune with the modern epistemological developments which began with the discovery of non- Euclidean geometry in the mid-nineteenth century and gave rise to a vast anti-mechanistic movement.⁸ It seems fair to say, however, that these new developments lead Hicks not to make major conceptual changes with respect to the original general equilibrium model and its atomistic core, but simply to provide a more rigorous version of the same model, purged from many of the realistic claims or unnecessary concessions to utilitarianism made by Walras.⁹ In particular, it can be argued that Hicks’s innovation is that he changes the way of representing the individual’s choices (e.g. the rejection of cardinal utility and the adoption of indifference curves), not the central role that the individual plays in the analysis.

Implications for the interpretation of Keynes

As far as the interpretation of Keynes is concerned, Hicks’s atomist view is relevant insofar as it accounts for his rejection of the aggregate concepts used in the General Theory. In line with the views of Robbins and Hayek (e.g. Deane 1978:146; Mirowski 1989:337; Popper 1961:103, 122, 1966:297), Hicks points out that the only legitimate approach to economic analysis is ‘pure theory’ which is based on concepts directly relating to individual behaviour and aims at logical precision: ‘In order to get clear-cut results in economic theory, we must work with concepts which are directly dependent on the individual’s scale of preference’ (Hicks 1946:177).¹⁰ In spite of their familiarity, macroeconomic aggregates such as income, saving and investment should not, on the contrary, be used in ‘pure theory’ since, being the product of ordinary language and everyday business life, they are not logically precise:

There is far too much equivocation in their meaning. At bottom, they are not logical categories at all; they are rough approximations, used by the business man to steer himself through the bewildering changes of situation which confront him. For this purpose, strict logical categories are not what is needed; something rougher is actually better.

(Hicks 1946:171)

Another reason which Hicks stresses for avoiding the use of aggregates is that they have no direct significance for the conduct of the individual agent —the sole concern of ‘pure theory’.¹¹ This much is clear from Hicks’s analysis of the concept of income, especially when he passes from a consideration of individual income to one of social or national income. The former is defined in ex ante terms; as a subjective concept, dependent on the particular expectations of the individual in question. It is ‘the maximum value which he can consume during a week, and still expect to be as well off at the end of the week as he was at the beginning’ (Hicks 1946:172). However, if the expectations of different individuals are not consistent, the aggregate of their incomes has little meaning. Hicks thus stresses that the only way to rescue the notion of national income is to consider it as an ex post magnitude:

Income ex post equals the value of the individual’s consumption plus the increment in the money value of his prospect which has accrued during the week; it equals Consumption plus Capital accumulation. This last very special sort of ‘Income’ has one supremely important property. So long as we confine our attention to income from property; … Income ex post is not a subjective affair…it is almost completely objective… [and] can be directly calculated. Since the income ex post of any individual is thus an objective magnitude, the income ex post of all individuals composing the community can be aggregated without difficulty.

(Hicks 1946:178–9)

However, Hicks also emphasizes that while the concept of national income in ex post terms may prove to be very useful in applied work, it should not be extensively used in economic theory:

Ex post calculations of capital accumulation have their place in economic and statistical history; they are a useful measuring-rod for economic progress, but they are of no use to theoretical economists, who are trying to find out how the system works, because they have no significance for conduct… The income ex post of any particular week cannot be calculated until the end of the week… On the general principle that ‘bygones are bygones’ it can have no relevance to present decisions.

(Hicks 1946:179)

On these grounds, Hicks draws the conclusion that ‘the concept of income [is] one which the positive theoretical economist only employs in his arguments at his peril. For him, income is a dangerous term, and it can be avoided’ (1946:180).

Hicks’s constructive method

Analogies between Hicks’s ‘pure theory’ and the mechanistic model

A few significant analogies can also be found between the constructive or compositive method underlying the mechanistic model and the one followed by Hicks in Value and Capital. While atomism in classical physics stresses the existence of unchangeable particles whose position and speed can be accurately stated, the constructive method shows instead that these particles constitute the simplest elements to which all physical phenomena should be reduced.¹² Two aspects of this method need to be stressed. The first is its universal applicability. As Einstein pointed out

The great achievements of mechanics in all its branches, its striking success in the development of astronomy, the application of its ideas to problems apparently different and non-mechanical in character, all these things contributed to the belief that it is possible to describe all natural phenomena in terms of simple forces between unalterable objects.

(Einstein and Infeld 1938:57–8)

The second aspect is that the method attempts to explain all phenomena on the grounds of extremely simplified models resulting from the aggregation of the elementary particles. In particular, the laws of the mechanistic model are formulated with respect to ‘good’ or ‘well-behaved’ frames of reference, that is, inertial frames of reference. Hicks’s approach seems to be in line with both of these aspects. On the one hand, it tries to account not only for remote stationary conditions but also for the most significant phenomena of the ‘changing, progressing, fluctuating economy’ or ‘the real world in disequilibrium’ (Hicks 1936:86). On the other hand, it does so by setting out from the highly simplified model of the logic of choice described in the first part of Value and Capital.¹³

The transition issue: Hayek’s view

In this book, Hicks seems to believe in the possibility of making a relatively smooth transition from the pure logic of choice to the analysis of real-world phenomena such as those dealt with by Keynes. When discussing the transition from the static theory of the first part of his book to the more dynamic theory of the second, he argues for instance: ‘There is a way of reducing the dynamic problem into terms where it becomes formally identical with that of statics. Thus the results of static theory can be used after all’ (Hicks 1946:119). However, the transition turns out to be much more problematic than Hicks’s words would appear to suggest. As Hayek (1937) pointed out, some serious problems arise in ‘pure theory’ when the focus is no longer on isolated individuals as in the logic of choice, but on the interaction of agents as in the dynamic part of the theory. One major problem is that the analysis of competitive equilibrium implies the focus on time and individuals’ plans and expectations. In particular, Hayek point...

Table of contents

• Front Cover
• Keynes and the Neoclassical Synthesis
• Routledge Studies in the history of Economics
• Title
• Copyright
• Dedication
• Contents
• List of figures
• Preface
• Introduction
• Part I The two basic paradigms of macroeconomics
• Part II The Neoclassical Synthesis
• Part III Microfoundations
• Conclusion
• Notes
• References
• Index