Cambridge and the Monetary Theory of Production

This book considers the development of Cambridge monetary and trade cycle theory from the late nineteenth century corpus established by Marshall to the late 1920s, just prior to the ‘Keynesian revolution’. It is only recently that attention has started to turn to the development of the Cambridge Tradition in this period, for example Bridel (1987), and to the question of the forces behind the collapse of the old tradition.

When dealing with the history of thought, in any discipline, there is a tendency to force the actual sequence of events into preconceived theoretical structures either explicitly or, more often, implicitly. That is to say we rationalise a process of evolution which may be, at times, essentially intuitive or even irrational.¹ The process of putting one’s thoughts together, consciously or unconsciously, involves the acceptance or rejection of certain frames of reference. It may be that we normally think or write in a clear, but implicit, framework or orthodoxy. To imagine a problem, is to have already accepted the context. This is the sense of Kuhn’s normal science or of work within Lakatos’s research programmes. Although the framework is potentially questioned with every thought, word or argument, this may not be of primary importance to the historian of thought. However, when the context or tradition begins to be explicitly reconsidered, or inconsistent ideas are maintained, then we must conclude that the orthodoxy itself is under question. In these conditions the writer is, more or less consciously, selecting, creating his precursors.² The retention or rejection of a phrase becomes crucially important. So too may be the precursors, or even the successors, not because ‘it has all been done before’ but because there may be an implied similarity of goals or modes of thought.

The Cambridge School or tradition is already well established in the secondary literature (for example Eshag, 1963; Patinkin, 1974; Moggridge, 1976; Bridel, 1987 amongst many others), but is also apparent from references in the primary material.

This chapter reviews Marshall’s monetary theory under the two main headings of the quantity theory and the trade cycle. This is in preparation for the following chapter which considers the Marshallian research programme. The discussion is concerned only with the domestic economy, and no consideration is given to the international aspects of the theory; for example the determination of exchange rates, the effects of balance of payments disequilibria, and fluctuations in international activity and credit.

From the discussion in Chapter 4, and from the evidence of other reviews of Marshallian thought, it is possible to derive a definition of the Marshallian research programme within the framework suggested by Lakatos. Even outside this methodological framework, the following sections provide, it is hoped, a concise statement of the key elements of what was Marshallian (and pre-1930s Cambridge) economics.

The aim of this chapter is to consider the treatment of risk and uncertainty (in so far as it can be separated from the theories of money and the trade cycle) within the Marshallian research programme. As was outlined in the previous chapter, subjective uncertainty is incompatible with a substantively rational theory. Marshall, as we have seen, attempted to completely objectify uncertainty in microeconomic contexts. This conforms more closely with the present day idea of risk, and it has been suggested that the Classical economists attempted to describe a world without uncertainty but with risk.¹ The same criticisms can be made of the early Cambridge theories. However the type of uncertainty envisaged in the macroeconomic context does not necessarily fit into this category. Therefore we must consider the development of the theory, in particular the strengthening of these more subjective elements of the analysis. This development leads to a more adaptive view of economic behaviour and the possibility that the invisible hand may at times lose its grip. These developments can be seen alongside the increasing tide of criticism of Marshallian microeconomic analysis roughly starting with the Empty Boxes debate in 1922 and continuing with the later attacks on the law of returns and the theory of wages.² Looking forward, some commentators have linked Keynes’s Treatise on Probability (1921i) with the concept of uncertainty claimed to underlie the General Theory (for example Weintraub, 1975).

The development of the Cambridge quantity theory is divided over the next three chapters, and reappears again in Chapter 11 which considers Robertson’s Banking Policy and the Price Level (1926). The subject of the current chapter is the first algebraic presentation of Marshall’s real balances approach by Pigou (1917i)and Keynes’s (1923i)later development of that equation. Hawtrey’s nominal balances approach (as in 1919i) is also considered, since, at least in part, later work in Cambridge can be seen as the coming together of these two approaches. Hawtrey built a theory which could deal with disequilibrium conditions, and a similar analysis was later used by Robertson (1926), but this is properly the subject of a later chapter. Hawtrey also distinguished between consumers’ and traders’ balances, an approach which attracted Keynes in his work after 1923.

Part of the story of the development of the quantity theory in Cambridge in the 1920s is the assimilation of two strands: the Marshall/Pigou approach and the more independent model of Haw-trey. This can be seen as reaching a peak in Robertson’s Banking Policy and the Price Level (1926) which is the subject of a later chapter. The current chapter deals with other elements of the story, specifically Lavington’s and Robertson’s analysis, and in particular their consideration of the role of the banking system. Then, of course, there is Keynes’s Tract on Monetary Reform (1923i) which has been seen both as his most classical quantity theory analysis and as the beginning of his intellectual moves towards the General Theory.

This set of three chapters, reviewing developments in the Cambridge Quantity Theory in the period to 1925, concludes with a consideration of the perceived limitations to the quantity theoretic approach and the concept of the money veil. Once again the main authors reviewed are Keynes, Robertson and Lavington.

With the exception of Hawtrey, the trade and credit cycle theories adopted by the Cambridge writers were broadly based on Marshall’s approach. Hawtrey, however, stressed the monetary aspects of the cycle to a far greater extent and denied the psychological causes underlying the Marshallian approach. This chapter considers some developments of Cambridge trade cycle theory in the period up to 1923. This period is characterised by three phases. The first, pre-war, phase saw some interesting theoretical developments: Hawtrey propounded a model which gave great importance to changes in stocks and quantity adjustments to equilibrium,¹ while Keynes provided a model which concentrated on the amounts of savings and investment. Neither innovation was immediately adopted by the Cambridge School, though from this period through to the late 1920s there was close co-operation and contact between Keynes and Robertson.² The second phase broadly covers the war years; in this period Robertson first expounded his over-investment theory of trade cycles, and there was an increasing awareness of the limitations of the invisible hand. The final phase covers the 1920s (in fact extending beyond 1923) during which time some of the first period developments began to assume greater prominence, and there was continued concern over the limits to the working of the invisible hand.

Since first examining the nature of the Marshallian research programme and suggesting some inherent factors which could ultimately lead to its collapse, in Chapter 5, subsequent chapters have followed the development of Cambridge macroeconomic analysis up until the late 1920s. What this development shows is an increasing, but implicit, rejection of the basis of the Marshallian microeconomic schema (the substantively rational part of the programme). As suggested in Chapter 5, it is to be expected that this should have exacerbated the problems of inconsistency within the overall Marshallian research programme. In reviewing Robertson’s Banking Policy and the Price Level (1926) it has just been shown that when Harrod suggested dropping the concept of justifiable fluctuations Tappan’s defence was to explicitly link this part of Robertson’s theory to the Marshallian marginal analysis. In this sense, as well as in retrospect, Banking took explicitly Marshallian macroeconomic analysis about as far as it could go. If the theory was to make further substantial steps forward, some elements of the Marshallian system had to be abandoned. Of course the Cambridge Tradition did not suddenly end with a blinding flash in the Keynesian revolution, nor was the period from the late 1920s solely concerned with the transition directly to the General Theory.

The main thesis of this book is that the collapse of the Marshallian research programme was due to an inherent split in the very make-up of the programme itself. Marshall’s contribution to the state of political economy in the late nineteenth century was considerable (see, for example, Deane, 1989, pp. 134–41), but it was also essentially a compromise: by drawing together and reconciling many of the different strands of political economy Marshall’s framework helped weld economics into a distinct and separate discipline (the methodological issues being dealt with at much the same time by Keynes’s father, John Neville, in his Scope and Method of Political Economy first published in 1890 — see Deane, 1983). The cost of the compromise was that not all the problems could actually be resolved into a single framework and some cracks were thus papered over.