The New Economy and Macroeconomic Stability

Dario Togati

• Publisher: Routledge
• Year: 2013
• ISBN:
  9781134302789
• Edition: 1
• Topic: Economics
• Subtopic: Macroeconomics
• Index: Economics

The past decade has seen many leading economies, especially the US, undergo profound structural transformations. Departing from the standard theories employed to explain this phenomenon, here author Togati provides the first broad analysis of the New Economy. In this book, the first to look at the new economy from a post-Keynesian / post-modern perspective, he focuses on its macroeconomic implications, presenting a more balanced view than that provided by orthodox neoclassical analysis, and studying the interaction of key variables such as:

* information technology
* globalization
* the increasing significance of intangibles and financial markets.

This ground-breaking book utilizes a 'neo-modern' perspective drawing on complexity theory to advance the study of the stability and dynamic behaviour of economic systems. Togati utilizes the Calvino labels to identify new empirical evidence, and examines the implications for global stability based on New Classical Macroeconomics and Keynsian theory.

The analysis developed in this book has important practical and policy implications for the New Economy, making this book essential reading for students, academics and practitioners in this field.

Part I

Alternative approaches to stability

Part I of this volume examines a number of approaches to the analysis of stability. As noted in the introduction, it is not enough for our purposes to rely on mere observation of macroeconomic indicators, for these incorporate policy moves and, thus, may obscure structural change. We have also argued that a fully satisfactory approach to stability issues has to include structural analysis concerning the sources of qualitative change arising from the interplay of a number of key phenomena, while operating within the context of a relatively weak notion of macroeconomic equilibrium.

What follows is an attempt to determine how well the existing approaches fulfil this ideal. As shall be seen, none of them is entirely successful, despite the valuable insights they provide. Neoclassical theory, dealt with in Chapter 1, adopts a strong notion of equilibrium but neglects to consider structural change. Instead, the relatively heterodox theoretical perspectives discussed in Chapter 2 combine a concern for certain forms of instability with weaker notions of dynamic equilibrium. Finally, the neo-Schumpeterian approach, discussed in Chapter 3, stresses structural change without equilibrium.

The overview provided in Part I is not intended to be exhaustive. Other approaches to instability, such as those by Minsky and the theory of transformational growth (e.g. Nell 2002), have not been included because they can be seen as akin or complementary to the Keynesian perspective defended in this book. They will therefore be addressed more explicitly in later sections, along with a discussion of Keynes' theories.

1 Equilibrium without structural change

In this chapter we focus on how neoclassical theory, with its strong roots in the tradition of general equilibrium economics, treats the issue of stability. Since analysis of the NE inevitably implies reference to some notion of equilibrium, it seems only natural to assume that the Standard concept of general equilibrium intrinsic in much of current macroeconomics might serve as a good starting point for the study of stability. However, this would be a very hasty conclusion. Given that it considers the system to be inherently stable, both dynamically and structurally, Standard general equilibrium theory is clearly not suited for the task.

Key aspects of neoclassical methodology

It is worth noting from the start that the assumption of stability underlying neoclassical theory does not derive from sheer neglect of complexity. Nor does it stem from the a priori ruling out of particular variables and their interrelations. For example, neoclassical theory acknowledges the existence of a range of factors capable of influencing economic growth; moreover, in its analysis of the NE, it readily admits the existence of factors other than IGT, such as globalization or finance. Rather, the premise of stability in neoclassical theory is the inevitable outcome of how general equilibrium theory deals with this multitude of variables and their interrelations. Above all, it is a result of the reductionist approach adopted by Standard theorists that leads them ‘to remove anything that could create inherent instability from their models’ (Tvede 2001: 164). This is accomplished by adopting a number of key assumptions (e.g. linearity, perfect competition and negative feedback mechanisms) and analytical tools (e.g. production functions and representative agent devices) as the basis for the macroeconomic and econometric models used to study fluctuations and growth.

Let us take a closer look at the particular features of Standard models that allow any assumption of instability factors to be ruled out. First of all, theory within the general equilibrium tradition is taken to be a purely deductive and self-contained sphere or ‘closed system’ (see e.g. Lawson 1997; Dow 2002; Chick 2003). One implication of this stance is that Standard models admit a clear-cut definition of exogenous and endogenous variables and accept the ‘economicist’ view that ‘the boundaries of the economic sphere are objectively defined’ (Freeman and Louça 2001: 110). Having defined the maximizing behaviour of atomistic agents as the canon of rationality and the ‘first principle’ of economic analysis, Standard theory holds that one can endogenize, that is, explain in terms of rational behaviour, certain factors such as state intervention, social institutions or cultural characteristics. Indeed, according to Gary Becker's approach, for example, phenomena such as education and skills, crime, the household and even addiction can be ‘treated as if [they were] the outcome of rational (utility maximizing) behaviour by individuals, albeit in what often are non-market contexts’ (Fine 2003: 214).

Second, Standard theory takes the view of economies as cumulative and linear processes. The assumption of linearity allows a facile solution to the complexity of interrelations. Standard models tend to focus on a small number of variables and to group related factors together under some exogenous residual random term. For this purpose, these models ‘assume linearity somehow and then brush off the remainder as statistical noise’ (Freeman and Louça 2001: 109; Tvede 2001: 186). In particular, production functions and growth accounting exercises treat technological change essentially as an exogenous shock or residual. Technological change is, therefore, regarded as having simple, linear effects on the economy under artificial conditions (perfect or unbounded rationality, free competition, full availability of perfect information, etc), introducing the concept of what can be referred to as ‘technological determinism’.

Third, neoclassical theory is based on a mechanistic view, according to which it is possible to reason in terms of atomistic relations, or relatively isolated subsets of the complex socio-economic system. Thus, for example, Standard theories rely on the ceteris paribus condition, following Mill's approach. While they do take into consideration a wide variety of factors, these are dealt with separately, as though independent. The individual causal factors analysed in isolation from all other factors can then be – ideally – combined with them once all theorizing is complete. In other words, analysis is carried about before synthesis (see Viskovatoff 2000: 151). A mechanistic view also implies that it makes sense to isolate pure market forces from their institutional context, which can be taken as a given. The implicit assumption is that markets tend to perform the task of allocation no matter what kind of shocks or institutional context are at work in the economy.¹ For example, despite the enormous institutional differences between countries, competition is assumed to bring about certain outcomes – such as international price uniformity – as implied by the law of one price or the purchasing power parity theorem.

A fourth factor is that Standard models used in the study of growth and cycles are intrinsically rooted in the concept of equilibrium – in the sense of a permanent dynamic property prevailing in the markets – and the rocking-horse metaphor introduced by Wicksell and Frisch (see e.g. Louça 1997; Freeman and Louça 2001: 60; Tvede 2001: 149–50). According to the equilibrium paradigm

the economic system tends spontaneously to equilibrium; cycles are exogenous perturbations produced by random shocks which trigger an endogenous propagation mechanism with stabilising properties. This provides the rationale for separating growth and fluctuations, that is for decomposing the movement of an economic system into trend and cycle. Trend is conceived as the loci of equilibria – a moving centre of gravitation – while cycle is restricted to the analysis of the stochastic error term and to the properties of the equilibration mechanism.

(Reati 2001: 109)

It is important to note that detrending – namely the view that the trend (the growth of the economy) and the cycles (the acceleration and deceleration of growth) can be dealt with as if they were unrelated phenomena – implies ruling out the effects of cycles on the trend. As a result, structural factors that are likely to influence the longer-term evolution of productive forces come to be defined as merely exogenous. This, in fact, is a crucial aspect of Solow's Standard growth model, which assumes that the natural rate of growth depends on the growth of the labour force and labour productivity (as determined by technical progress), and that these are exogenously determined, as implied by the assumption of diminishing returns to capital (see e.g. Thirlwall 2002: 20–8).²

It is also worth noting that emphasis on equilibrium implies a corresponding emphasis on stationarity and the irrelevance of the passage of time. This means that economic variables are expected to stabilize around some imaginary permanent level or constant rate of growth. In other words, Standard theory relies on the ‘ergodic assumption’, resting on the belief in long-run equilibrium independent of initial conditions (see e.g. Davidson 1982–3), the neglect of uncertainty and, thus, acceptance of the deterministic nature of empirical phenomena.

As a result, the descriptive potential of equilibrium models is limited to recurrent phenomena characterized by a high degree of quantitative regularity (see e.g. Vercelli 1991: 141). Neoclassical macro theory thus concerns itself with ‘universal laws’ conceived as ‘event regularities’, that is, stable patterns among data series identified with the aid of econometric techniques, such as the link between money and prices in Friedman's version of the quantity theory of money (see e.g. Lawson 1997, 2003; Dow 2002: 138).

Stability as an article of faith

The stability assumption can also be seen as a counterpart to economic mechanisms that, for neoclassical theory, grant the attainment of equilibrium. As for dynamic stability, neoclassical theory focuses on the role of market-clearing prices. The efficiency of the price mechanism is guaranteed by the existence of negative feedbacks, which are thought to return the system to a unique point of equilibrium whenever it has deviated from it. In particular, the presumed universal presence of diminishing returns assures that the system will gravitate towards a unique and stable equilibrium (see e.g. Prasch 2000a: 220).

It should be noted that this conclusion, which underlies much of current macroeconomics, does not follow from scientific demonstrations based on dynamic accounts (involving out-of-equilibrium processes) of how equilibrium comes about. Strictly speaking, the competitive static equilibrium underlying Standard theory implies an a priori view concerning stability:

All economics has been able to do for the last fifty or one hundred years is to look at systems with very strong attractors, not even talk about how an equilibrium point is reached but simply point out that there is an equilibrium and that if we were there, there would be a tendency to stay there.

(Arthur 1994: 64)

This is no mere coincidence. It has been clear for some time that attempts to demonstrate dynamic stability have failed, not only in the context of Standard general equilibrium models, but also within the context of more recent developments in ‘pure theory’, including game theory (see e.g. Vercelli 1997: 290).³ Indeed many theorists have concluded that

the two main classes of models which were consecutively taken to embody the most ‘fundamental’ economic theory – general equilibrium theory and game theory – suffered from the same basic problem: they could not demonstrate the stability or uniqueness of the equilibrium which their logic seemed to require.

(Viskovatoff 2000: 129–30)

Rather than dropping the assumption of general equilibrium because of lack of proof, many mainstream macroeconomists seem to take stability almost as dogma. This can be seen in the following statement by a prominent economist (reported by Fisher): ‘[T]he study of stability of general equilibrium is unimportant, first, because it is obvious that the economy is stable and second, because if it isn't stable we are all wasting our time’ (1983: 4; quoted in Vercelli 1991: 30–1).

In terms of structural stability, commitment to the uniqueness of equilibrium forces general equilibrium theorists to assume that the structure of the economy is fixed and immutable, that it is a given prior to the interaction of agents within the market. As underlined by Prasch, this ‘assumption allows prediction and removes market processes from the study of economics’ (2000a: 218); in particular, the market process, the price relations, must not feedback on the structure of the economy, that is, either demand or supply schedules, otherwise the possibility of multiple equilibria cannot be ruled out.⁴

Limitations of the stability assumption

Strictly speaking, the ubiquitousness of the stability assumption in Standard macroeconomics does not necessarily lead to the ruling out of negative outcomes, such as depressions or sluggish growth. Claims about the possible instability of real-world economies can be found in the literature, in textbooks and research articles alike. However, the lack of attempts at stability analysis in Standard macro theory undermines the quality of its explanations and makes it subject to a number of serious limitations. First of all, it leads the theory to dismiss the scientific relevance of unstable equilibria, which are considered to be ephemeral and unobservable states (see e.g. Vercelli 1997: 287).

Second, Standard theory regards negative outcomes as being ultimately generated not by the internal workings of the economy but by anomalies, exogenous disturbances to the private sector (e.g. productivity shifts and institutional rigidities) or through sheer error (e.g. confusion about relative prices or between permanent and transitory prices or policy mistakes, such as unjustified changes in the money supply of the central bank) (e.g. Tvede 2001: 160–1). In other words, it can be argued that neoclassical theorists rely on the distinction between external impulse, or shock and propagation (the inherent equilibrating mechanism) made by Frisch (1933) with his 'stick and rocking horse' metaphor:

The fact that neoclassical theorists assumed that markets were rational and efficient did not mean that they pretended that business cycles didn't exist. It would be absolutely coherent to assume that markets are efficient and people rational and at the same time observe business cycles if these were caused by series of external shocks. The shocks were like Ragnar Frisch's ‘stick’. The basic assumption was that shocks that were external to the market place itself caused the fluctuations.

(Tvede 2001: 163)

Third, it provides an explanation of long-run instability emphasizing supplyside factors. Following its use of production functions to analyse growth issues, neoclassical theory holds that those countries that face sluggish growth of income or productivity have an insufficient endowment of productive factors, such as labour force or various types of capital.

In the end, the stability assumption invites many macroeconomists to accept a peculiar sort of dichotomy between short-run and long-run analysis. This occurs when the notion of stability of equilibrium is only accepted with reference to the long run, as is the case with many Old and New Keynesians. While NCM insists that the economy is always in perfect market equilibrium, this influential group of Keynesians accept that the economy may be in disequilibrium in the short run (i.e. at variance from full employment equilibrium), with the adjustment process taking place only rather slowly due to various types of market imperfections.

Blanchard's recent macroeconomics textbook is a good instance of this approach. Several times throughout the book, Blanchard mentions the current lack of visible endogenous adjustment mechanisms in many European economies. However, he explains that in his view this is either a short- or a medium-run phenomenon and does not call into question the ‘economies of the long-run’ based on the stability assumption (see Blanchard 2003). This approach reveals an internal consistency problem, as it amounts to suggesting that alternative principles are relevant in explaining the same circumstances. Old and New Keynesian macroeconomics also fail to completely harmonize short- and long-term equilibrium, leaving a ...