Macroeconomic Foundations of Macroeconomics

Alvaro Cencini

• Publisher: Routledge
• Year: 2012
• ISBN:
  9781134382231
• Edition: 1
• Topic: Business
• Subtopic: Business General
• Index: Business

Contrary to common belief, macroeconomics is not merely a theory of aggregates, and cannot be constructed from individual behaviour. Both nationally and internationally, there are economic laws that are logically independent of economic agents' behaviour. These are the macroeconomic foundations of macroeconomics.

Presenting cutting-edge material, Alvaro Cencini explores these foundations, and shows that the introduction of money entails economics being interpreted conceptually not mathematically. His innovative book provides the elements for a new approach by applying the most recent results of monetary analysis to the study of national and international economics. It covers recent progress in monetary theory, provides the reader with a greater understanding of the subject, and will be essential reading for economic students as well as a valuable resource for economists.

Part I

Macroeconomics versus microeconomics

1 Neoclassical, new classical, and new business cycle economics

Walras and general equilibrium analysis: the triumph of microeconomics

Let us be clear. Walras was one of the greatest economists of all times and his attempt to make a rigorous science of economics deserves our unreserved admiration. Like Jevons and Menger, the French economist believed that pure economics is a branch of mathematics and that it is thanks to algebra that the rational method may be introduced into economics. ‘[G]iven the pure theory of economics, it must precede applied economics; and this pure theory of economics is a science which resembles the physico-mathematical science in every respect’ (Walras 1984: 71). According to Walras, the pure theory of economics is the theory of value, and ‘the theory of exchange and value in exchange, that is, the theory of social wealth considered in itself, is a physico-mathematical science like mechanics or hydrodynamics’ (ibid.: 71). The founding father of general equilibrium analysis goes as far as to claim that the value in exchange ‘partakes of the character of a natural phenomenon, natural in its origins, natural in its manifestations and natural in essence’ (ibid.: 69).

Walras's desire to build economics on scientific and mathematical foundations was so strong that he did not hesitate to claim that the price of any given commodity (wheat in his example) is determined ‘objectively’, by scarcity or usefulness – ‘both of these conditions being natural’ (ibid.: 69). Hence, if wheat is worth 24 francs a hectolitre, ‘[t]his particular value of wheat in terms of money, that is to say, this price of wheat, does not result either from the will of the buyer or from the will of the seller or from any agreement between the two’ (ibid.: 69). Unfortunately, Walras never really attempted to explain value in exchange independently of the will of the buyer or the seller or of any agreement between the two. He endeavoured instead to show that relative prices are determined through exchange in a competitive market where supply and demand are subject to individuals’ preferences. ‘Value in exchange, when left to itself, arises spontaneously in the market as the result of competition. As buyers, traders make their demands by outbidding each other. As sellers, traders make their offers by underbidding each other’ (ibid.: 83).

Walras's revolution with respect to the theory supported by some great classical authors such as Smith, Ricardo, and Marx consisted mainly in switching from the analysis of production and absolute values to that of exchange and relative prices. The necessary rejection of the metaphysical concept of embodied labour led Walras to investigate the exchange between commodities, in search for a principle defining value as a mere relationship. Concentrating his attention on relative exchange, the author of Elements of Pure Economics reduced production to a subsidiary role, classifying it under the heading of applied economics. ‘Thus the theory of economic production of social wealth, that is, of the organization of industry under a system of the division of labour, is an applied science. For this reason we shall call it applied economics’ (ibid.: 76). The consequence of this choice is that pure economics is limited to the mathematical attempt to explain how supply and demand may determine relative prices and at the same time reconcile the maximization of individuals’ well-being with that of the society as a whole. A difficult task indeed, which led Walras to transform the market into a deus ex machina, and which set the stage for any future microeconomic analysis based on the principles of general equilibrium.

The widespread acceptance of Walras's GEA marked the triumph of microeconomics over the classical authors’ attempt to provide economists with a macroeconomic theoretical framework. Individual agents’ behaviour became the key factor in an essentially subjective interpretation of the ‘natural elements’ of economics. Supply and demand were (and still are) considered the social forces through which individuals’ preferences exert their action on the market in a continuous search for an equilibrium that, because of its very nature, can never be stable. Far from determining a set of logical rules economic agents have to conform to, neoclassical economists expect economic agents to find the best possible equilibrium allowing for the optimization of their individual utilities. Concerning the system taken as a whole, the state of the economy is thus seen as the result of a constant adjustment involving each individual agent in his relation with the others. No true macroeconomic criteria are envisaged and economics itself is conceived of as nothing more than a secondary branch of mathematics.

If economics were indeeed of a mathematical–physical nature and if relative prices could actually be determined by the simultaneous solution of a system of independent equations of supply and demand, Walras would be right in claiming that ‘the theory of value in exchange is really a branch of mathematics which mathematicians have hitherto neglected and left undeveloped’ (ibid.: 70). The triumph of microeconomic analysis is therefore closely related to the possibility of interpreting economics mathematically and, in particular, to the logical possibility of explaining economic events using mathematical formalism. Since Walras's publication of his Elements, the host of economists approaching economics from a mathematical viewpoint has never stopped swelling. Today the great majority of economists believe mathematics to be an indispensable analytical tool and make regular use of it. This simple observation is enough to give a clear idea of how far Walras's influence reaches and how deeprooted microeconomic foundations are. Despite important differences, the main schools of economic thought share Walras's belief in mathematics as well as his microeconomic approach. Apart from a handful of heterodox thinkers, in fact, economists still reason in terms of adjustment, looking for a more or less perfect equilibrium in more or less flexible markets.

It is a fact that Walras's general equilibrium analysis led to a radical change in economic thinking, and that even Keynes's revolution did not succeed in overthrowing it. Indeed, Keynes's original contribution was submitted to a neoclassical interpretation that was implicitly, if not explicitly, accepted even by a vast number of Keynes's followers. As we shall argue at length in the following two chapters, both Keynesian and new Keynesian economists share the neoclassical point of view with respect to the need of explaining reality through mathematical modelling. Now, the use of mathematics rests on the possibility to determine a positive number of functional relationships between economic ‘variables’ (where the choice of the term ‘variable’ is itself evidence of the fact that economic magnitudes are axiomatically considered to be functionally related). Except for the particular case of identities, equations are conditional equalities, that is, equalities that are verified only for a given numerical value of the variables they relate to. Within GEA economic reality is represented by a series of models based on conditional equalities (the most famous being, of course, the equations relating supply and demand). Despite their notorious differences with the neoclassical models, Keynesian and new Keynesian models are also constructed on the same logical grounds. This explains why the search for microeconomic foundations has attracted so much interest. At the same time this clearly shows how deeply Walras's conception of economic analysis has influenced the history of our science in the last hundred and thirty years.

New classical economics: the microfoundations of econometric modelling

New classical economics: a short introduction

The first new classical models were elaborated in order to propose a valid alternative to Keynesian econometric models and in order to explain how monetary factors could generate business cycles, without abandoning the main features of GEA. As a first approximation, it may be said that new classical models reproduce the Walrasian general equilibrium system except for the introduction of imperfect information regarding prices. For example, Lucas (1972, 1973) showed that by dropping the assumption of perfect information the Phillips curve is compatible with the general equilibrium framework, so that monetary shocks can indeed have an influence (albeit only a temporary one) on real variables.

As claimed by Laidler, although the monetarists’ criticism of the Keynesian analysis of inflation and unemployment jeopardized the reliability of particular functional relationships in the economy, monetarism ‘constituted no radical theoretical challenge’ (Laidler 1997: 335) to Keynesian orthodoxy. In fact, Keynesian econometric models did already incorporate a demand for money function and ‘could easily enough be accommodated’ (ibid.: 335) in order to account for a more stable relationship. Likewise, ‘large-scale Keynesian econometric systems proved easily able to absorb monetarist ideas’ (ibid.: 335), so that monetarism did not really provide a theoretical alternative to Keynesianism. What monetarism failed to do was successfully carried out – so we are taught – by new classical economics, which provided an analytical setting clearly distinct from the Keynesian framework. The central feature of the approach proposed by Lucas and his followers is the reintroduction of the main principles of general equilibrium analysis, in particular those of price flexibility and competitive market clearing. The originality of new classical models with respect to traditional general equilibrium models lies in the assumption that economic agents ‘do not have full information about the structure of relative prices when they engage in trade’ (ibid.: 338). The rational expectations hypothesis introduced by new classical economists has thus the advantage of accounting for equilibrium while leaving room for temporary disequilibrium. Uncorrelated random errors are in fact possible when agents form their expectations, but must necessarily reduce to zero if expectations are rational. Hence, output and employment fluctuations are perceived as ‘voluntary responses to misperceived price signals’ (ibid.: 336) occurring when the economy is subjected to random exogenous shocks. Agents may mistake changes in money prices for changes in relative prices and consequently modify their supply of goods. Yet, this change in real supply is not supposed to last. Economic agents will soon be aware of their mistake and modify their behaviour in accordance with the actual evolution of relative prices, thus re-establishing equilibrium at its proper level.

Reliance of new classical economics on price flexibility and market clearing is a sign of how closely related it is to microeconomic principles. It is not a mystery, of course, that new classical economists claim to derive macroeconomic models of reality from individual maximizing behaviour. What is particularly relevant here is the claim that new classical economics allows ‘clearly defined links to be established between individual and market experiments without recourse to empirical laws’ (ibid.: 340). It thus seems possible to avoid the traditional obstacle of aggregation facing Keynesian macroeconomic theories. Derived from a Walrasian theoretical framework, new classical economics builds its macroeconomic system directly on it. The postulates of perfect competition and rational behaviour facilitate the transition from micro to macroeconomics, the distinction between the two being a matter of size and not of substance. It is clear that, in these conditions, macroeconomics is reduced to microeconomics, the state of the economy taken as a whole being the direct result of the decisions taken by individual agents. The claimed superiority of new classical economics over its rivals rests precisely on the principles of GEA, the logical coherence and rigour of which it shares. In particular, new classical economists are well aware of the fact that mathematical modelling has its natural origin in the neoclassical system of general equilibrium, which is still the most rigorous setting to date, at least as far as mathematical economics is concerned.

The new classical approach has been widely criticized, both outside and within mainstream economics. Akerlof, for example, in his inaugural lecture at the London School of Economics maintained that the continuous market clearing hypothesis is generally incompatible with the principle of profit maximization. Having claimed that profit-maximizing firms tend to follow standard business practice, which ‘normally prohibit[s] firms and industries from paying the market-clearing wage at all times and in all places’ (Akerlof 1979: 231–2), Akerlof goes on to prove that ‘the firm which tries, contrary to the standard practice, to behave in the manner suggested by the market-clearing model will not in general maximise its profits’ (ibid.: 232). Violation of widely accepted norms and of standard business practices will in fact prove costly, often too costly when compared to the advantages deriving from the adoption of market-clearing wages. Now, Akerlof's criticism is much less harmful than it might appear. Norms and business practices are historically determined and tend to evolve. Practices that are costly today may well become the norm tomorrow. We can openly disagree with the political and ethical implications of the new classical theory, of course. But this is not the point. What really matters here is whether or not the new classical macroeconomics depicts the logical workings of an ideal economic system.

New classical economists claim that their models are the only models that are logically self-consistent, and that the introduction of the continuous market clearing and rational expectations hypotheses is a logical requirement for a rigorous theory to be worked out at all. The free competition framework resulting from it would therefore be the unavoidable and ‘objective’ consequence of choosing the only empirical setting truly compatible with the sole theoretical model of the economy. If today's institutions do not entirely conform to the ideal model, this is no proof that the model itself is wrong. If we are to reject it, it must be on logical grounds, and not simply because its consequences do not fit our ethical or political choices.

A good economic theory must be able to explain the real world. The new classical theory is no exception to this rule. The fact that markets do not clear ‘at all times and in all circumstances’ (ibid.: 233) is apt to undermine the new classical approach, as is the fact that economic agents are far from acting rationally ‘at all times and in all circumstances’. Yet, this discrepancy between theory and empirical evidence might be simply a matter of gradual adjustment. If no logical inconsistency were to be found in the new classical approach and if it were possible to show that its foundations rely on empirical axioms derived from the workings of the economic system, then new classical economists could indeed maintain that their theory has the twofold capacity to explain the present malfunctioning of the economy and to provide a model of how the economy should work in order to be consistent with its own logical laws (as determined precisely by the theory).

As we have already seen, the assumption of continuous market clearing and the postulate of price flexibility are a central tenet of new classical economics. While Keynesian macroeconomics rests on the assumption of price rigidity, new classical analysis states that a shift in demand is always matched by a change in prices allowing for the clearing of markets. If fluctuations in output and employment nevertheless occur in the real world, this is said to be due to misperceived price signals. Because of a lack of information about the meaning of money price variations, economic agents may mistake changes in money prices for changes in relative prices and react by modifying their real supply of goods and services. If this happens, output and employment fluctuate in the short run, giving rise to a real business cycle. Now, as observed by Laidler, the continuous market clearing hypothesis is apparently in contrast with individual agents getting their expectations wrong. Errors are costly, so that agents ‘have every incentive to make their expectations as accurate as possible, and to use all available information in order to do so’ (Laidler 1997: 31). This led new classical economists to resort to the rational expectations hypothesis. It is to this assumption that we shall now turn our attention.

The rational expectations hypothesis

Expectations were already taken into account by Friedman (1968), whose expectations-augmented Phillips curve suggested the existence of a relationship between the expected rate of change in real wages and unemployment. Yet, new classical theorists were not happy with Friedman's assumption since it privileged adaptive rather than rational expectations. In 1961, Muth put forward the idea that forward-looking agents form their expectations on a rational basis, which can be mathematically formalized. As noted by Gerrard, one great advantage of the rational expectations hypothesis is that it ‘offers a more acceptable assumption in choice-theoretic terms since rational expectations display the twin properties of unbiasedness and orthogonality, thereby ruling out systematic expectational errors’ (Gerrard 1996: 58).

According to the rational expectations hypothesis ‘agents form expectations “as if” they were fully informed about the structure of the economy in which they operate, and make mistakes only to the extent that the economy is subjected to random exogenous shocks’ (Laidler 1997: 34). Thus, models proposed by new classical theorists are greatly dependent on the way agents are expected to react to economic changes given their rationally determined behaviour. If they have insufficient information to distinguish relative from absolute changes in prices, monetary shocks may cause short-run fluctuations. However, if they behave rationally, preannounced monetary policies will be ineffective, tax-financed and debtfinanced fiscal expansions will be equivalent, and time-inconsistent policies will be less effective because they are less credible.

The rational expectations hypothesis was introduced as the principle required ‘to reconcile the price distributions implied by the market equilibrium with the distributions used by agents to form their own views of the future’ (Lucas 1980: 707). This reconciliation is a necessary requirement for the contingent-claim interpretations of a competitive equilibrium model suggested by Arrow and Debreu, and considered by Lucas as ‘a powerful model-building apparatus specifically designed to help us deal with problems involving choice under uncertainty’ (ibid.: 708). Yet, as observed by Buiter, the rational expectations hypothesis ‘appears to be in danger of being consistent with any conceivable body of empirical evidence’ (Buiter 1980: ...

Table of contents

• Front Cover
• Macroeconomic Foundations of Macroeconomics
• Routledge frontiers of political economy
• Title Page
• Copyright
• Dedication
• Dedication1
• Contents
• List of figures
• List of tables
• Acknowledgements
• Introduction
• Part I Macroeconomics versus microeconomics
• Part II The macroeconomic analysis of national economics
• Part III The macroeconomic analysis of international economics
• Part IV Conclusions and prospects
• Bibliography
• Author index
• Subject index