Open Economics

Economics in relation to other disciplines

Richard Arena, Sheila Dow, Matthias Klaes

• Publisher: Routledge
• Year: 2009
• Print ISBN: 9780415746571
• eBook ISBN: 9781134045679
• Topic: Business
• Subtopic: Business generale
• Index: Business

Economics has developed into one of the most specialised social sciences. Yet at the same time, it shares its subject matter with other social sciences and humanities and its method of analysis has developed in close correspondence with the natural and life sciences. This book offers an up to date assessment of economics in relation to other disciplines.

This edited collection explores fields as diverse as mathematics, physics, biology, medicine, sociology, architecture, and literature, drawing from selected contributions to the 2005 Annual Conference of the European Society for the History of Economic Thought (ESHET). There is currently much discussion at the leading edges of modern economics about openness to other disciplines, such as psychology and sociology. But what we see here is that economics has drawn on (as well as contributed to) other disciplines throughout its history. In this sense, in spite of the increasing specialisation within all disciplines, economics has always been an open discipline and the chapters in this volume provide a vivid illustration for this.

Open Economics is a testament to the intellectual vibrancy of historical research in economics. It presents the reader with a historical introduction to the disciplinary context of economics that is the first of its kind, and will appeal to practising economists and students of the discipline alike, as well as to anybody interested in economics and its position in the scientific and social scientific landscape.

Part I
Economics in relation to the humanities and social sciences

1 The social science of economics

Brian J. Loasby

… man, who can subsist only in society, was fitted by nature to that situation for which he was made.

(Smith 1759:85)

The argument of this chapter is that much modern economics is drastically undersocialised because it lacks an understanding of the distinctive characteristics of the evolved human mind, despite the significant insights provided by three of our most famous economists, Adam Smith, Alfred Marshall and Friedrich Hayek. This deficiency results from a failure to apply what may be considered the defining principle of economics, that of analysing the implications of scarcity. These implications challenge the adequacy of a theoretical structure based on the confrontation of preference functions and opportunity sets, even when extended to include formal interdependence, as in game theory; they require both a more modest view of human cognitive abilities and a more extensive view of human motivation and potential.

The fundamental scarcity

Economists study the consequences of boundary conditions. Most of these studies deal with actions within a particular set of boundary conditions; in models of constrained optimisation analytical outcomes are determined by the constraints. Some economists have tried to analyse behaviour which is intended to change these conditions, or which has the effect of changing them. I shall argue that the characteristics of the human mind have major influences on the ways in which we attempt both to operate within boundaries and to change boundaries, that models of self-interested optimisation provide an insufficient basis for understanding either of these attempts, and that social relationships—in which I include both firms and markets—are essential. It is not my intention to discuss the social influences on individual behaviour which have been emphasised by sociologists; my aim in this chapter is to explain the basis for the social organisation of activities, and in particular the significance of this organisation for the growth and use of knowledge. As Andrew Skinner (1996) has demonstrated, Adam Smith developed his economics within a system of social science, and my argument draws on some of the connections supplied by Smith, linking them to the work of others.

Standard economic analysis is focused on boundaries that are external to the individual agent: these, which include initial endowments and the potential actions of other agents, define the opportunity sets that individuals face. Structure—conduct—performance was once the standard label of industrial organisation, but it is actually the underlying principle of theory construction throughout economics: both individual actions and overall effects are deduced from a specification of the structures that define the boundaries of possibility. This principle is not uncommon in other sciences, both social and natural: the most prominent recent example is the attempt to explain the most significant features of human behaviour by the structure of our DNA.

Nevertheless, there is a significant limitation to the application of this principle within economics: few economists consider the boundaries that are internal to the human mind. The most fundamental of all scarcities which we face is the scarcity of our cognitive powers: the possibility of constrained optimisation is itself severely constrained by the limitations of human cognition. This is what Herbert Simon meant by ‘bounded rationality’. Because Simon’s phrase is often assumed to refer to certain imperfections, such as incomplete information, which can be incorporated in an adjusted model of optimisation, I prefer the term ‘bounded cognition’ (which was, I believe, first suggested by Richard Langlois): the proposition that meaning depends on interpretations is a major theme of this chapter.

We can actually perform only a small proportion of the cognitive operations that would be necessary for the universal rationality that is routinely assumed in so much economics. Typically we must either abandon optimising procedures or perform them on what is, at best, a caricature of the problem that we face; Simon has explored both possibilities. Attention is a scarce resource, so decisions must be severely rationed; ‘the decision-maker’s information about the environment is much less than an approximation to the real environment’ (Simon 1959:272); alternatives must be sought, and their consequences imagined. ‘Rationality’, in the sense of deductive logic, is therefore not the central issue in most decision-making: Chester Barnard, an experienced and thoughtful manager whose work was a major inspiration for Simon, pointed out that most of the action takes place before there is any possibility of applying logic (Barnard 1938:305). Logic may thereafter be useful in deducing implications from data, which is possibly false and never complete, and from conjectures, which may be inspired but are necessarily fallible; but it is not sufficient for making decisions. The connection between bounded cognition and business organisation in Simon’s work is not accidental; both the structure and the limited scope of all organisations, formal and informal, are devices to limit the demands on individual cognition by combining boundary-setting for individuals with mobilising the cognitive resources of a group.

This social organisation of economic activity, guided by the fundamental principle of the efficient allocation of scarce resources, has received relatively little attention in economics—though as I will indicate it is not a novel interest for economists. Ronald Coase (1972:60–61) pointed out that what was called industrial organisation was actually applied price theory; this displays the implications of various market structures but tells us nothing about industrial organisation, when that is defined as the distribution of activities between firms. Nowadays industrial organisation is applied game theory, which is no more informative about this distribution. Coase (ibid.: 8) also notes that ‘when economists … speak of market structure, it has nothing to do with the market as an institution’. I shall argue that the inability of standard economics to explain how activities are organised within an economic system is an inevitable consequence of its failure to recognise the cognitive limitations, and the distinctive cognitive potential, of the human mind.

Economists who use choice-theoretic models appear willing to consider any kind of scarcity except the pervasive scarcity of the cognitive powers of every individual, which not only restricts the application of logical reasoning to perceived problems but also influences what is perceived as a problem and how it is perceived: the external boundaries to which an individual reacts are themselves images within the mind, and should not be assumed to correspond to the actual boundaries. (This warning naturally applies to economists’ theories.) How and why these images differ between people is an important topic for research, not least for explaining what is identified as a problem and what may be considered as a solution. Because of our cognitive limitations, and the logical impossibility, demonstrated by David Hume, of establishing general empirical truths, we are surrounded by uncertainties; and somehow we have to deal with them.

Frank Knight (1921) defined uncertainty as a property of situations in which there are no demonstrably correct procedures for assigning probabilities to a set of possibilities. George Shackle (1972) focused on the absence of correct procedures for identifying the set of possibilities over which probabilities might be distributed; and this, I suggest, is more fundamental, for what prevents us from determining probabilities is precisely our inability to list all the possible ways in which particular events may come about, or all the possible ways in which they may be prevented.

Uncertainty is a formidable threat to standard techniques of analysis; when boundaries cannot be defined then the principle of deducing outcomes from a complete specification cannot be applied. The standard response of conflating uncertainty and risk (often with the aid of subjective probability), so that these techniques can be applied to a fictitious problem—a practice to which Knight (1933: xiv) forcefully objected—is therefore understandable as an apparent means of protection from a threatening unknown; but it may lead to significant and sometimes dangerous errors. If uncertainty is banished from the model, it is merely transferred to the relationship between the model and the situation to which that model may be applied. Having conflated uncertainty and risk in the analysis that won them the Nobel Prize in Economics (Merton 1998; Scholes 1998) Merton and Scholes also conflated them in formulating the business strategy of their company Long Term Capital Management, and quickly brought us, in 1998, very close to a world financial crisis.

It is, I believe, no accident that the standard conflation of risk and uncertainty is paralleled by the conflation of information and knowledge, because uncertainty, as Knight and Shackle so clearly saw, is a problem of knowledge, not information. The calculation of risk, especially by Bayesian analysis, utilises information; but this can only be done by presupposing certain knowledge of the possibility set to which it relates. Principal—agent and game-theoretic models can make effective use of asymmetric information, but only because every agent within these models is presumed to know precisely what this information, or its absence, signifies. Arrow (1974:40) correctly points out that ‘[a] signal hitherto unheard is useless by itself’; but in asserting that this is because ‘it does not modify any probability distribution’ he is identifying a second-order effect. The primary reason is that a novel signal cannot be incorporated into any existing structure of knowledge. There is, of course, no reason why a structure of knowledge should support any probability distribution. That novel signals may nevertheless be of extraordinary importance will be argued later.

Knowledge is an organised system of selected connections—and there is no demonstrably correct procedure for making the selection. The information content of any signal depends on the particular set of connections by which it is interpreted, and so the information that is derived from a signal very often differs according to the knowledge systems of the receivers, and may not be at all what the originator intended to convey. This is a common cause of communication failure; but as we shall see it can also contribute to innovation. Difficulties and differences of interpretation can be a serious problem in understanding what past economists were trying to say—and that is itself a sufficient reason for introducing prospective professional economists to the history of the subject. If I may mention an example that was decisive for my own attitude to economics, the core information content of imperfect competition theory as expounded at Cambridge was conceived by the expositors to be the ubiquity and inevitability of market failure; but to me the core information content was the inadequacy of this theory as an account of decision-making in economic systems, which is what I thought—wrongly, of course—economics was about. As I eventually realised, we had developed different knowledge systems. This disjunction between choice theory and the practice of decision-making eventually became the theme of a book (Loasby 1976).

Knight and Shackle both recognised that uncertainty was not simply a threat; it has a very different aspect. Without uncertainty ‘it seems likely that all organic readjustments would become mechanical, all organisms automata’, and ‘it is doubtful whether intelligence itself would exist in such a situation’ (Knight 1921:268). A world of ubiquitous rational choice would be a world without intelligence. If the future could be definitively predicted, it must logically be impossible for anyone to take any action that could change it; and few of us would be happy with that. Consequently ‘a life with uncertainty eliminated or perhaps even very greatly reduced would not appeal to us’ (ibid.: 348). As Shackle (1972, 1979) emphasised, it is uncertainty that allows scope for novelty, creativity, and the growth of knowledge, both theoretical and practical, because all of these entail exploration of some region within the space of ideas where particular boundaries may be abolished or redefined. The creation of knowledge is not a deductive operation: as Hume (1739–40:164) observed, ‘no kind of reasoning can give rise to a new idea’. For both Shackle and Smith, imagination is crucial to the growth of knowledge. Yet Shackle (1969:224) declared that ‘the boundedness of uncertainty is essential to the possibility of decision’; the limitations of our cognitive resources compel us to choose (consciously or unconsciously) where to direct our efforts, and to accept the unknown opportunity costs of doing so. Imagination requires some bounds if it is to be productive, and these bounds are to be found both in the organisation of the human mind and in all forms of external organisation. Boundaries differ between people, and for each person they differ between situations.

Acts of intelligence and imagination are themselves forms of organisation for which there are no demonstrably correct procedures. In Knight’s (1921:206) words, ‘to live intelligently … we must use the principle that things similar in some respects will behave similarly in certain other respects even when they are very different in still other respects’. Such organisational principles of relevant similarity must be related to ‘the purpose or problem in view’ and will have a limited range of applicability, and any individual’s skill in constructing and modifying such classifications must vary between domains. Barnard (1938:301) noted how long it took, even after thorough preparation, to adapt to a new position in the same organisation. This variation across domains poses a major allocation problem of differentiation and integration for all human communities, and provides a major role for the sociality of economics, in which, as Cattaneo (1861) argued, intelligence should be a basic principle. Intelligence cannot be reduced to deductive reasoning or information processing: ‘optimisation’ is simply not good enough. As I asked once before, ‘why should we be satisfied with the analysis of rational choice when we have the opportunity to study intelligent action?’ (Loasby 2001:410).

Theories of the mind

Knowledge is inseparable from uncertainty; but for that very reason knowledge can be improved; and this improvement can be organised. That is a theme which I believe can be fairly attributed to Adam Smith, though as with all major themes there is no single originator. The idea that improvements in human well-being depend substantially on the development of knowledge about production, and also consumption, and that such knowledge can be fostered by economic organisation, underpinned Marshall’s economic studies. It is a theme that has recently attracted substantial attention, including major institutional support in several universities and international associations such as DRUID and the International Schumpeter Society. The history of this theme over 250 years would provide a splendid topic for some ambitious and talented historian of economic thought. As a minimal indication of the potential I will briefly review some significant features of the theories of the mind that were constructed by Adam Smith, Alfred Marshall, and Friedrich Hayek. (For a more extensive account, which includes some evidence from neuropsychology, see Loasby 2004.) Camerer et al. (2005) present a systematic exploration of the implications of neuroscience for economics, though without making the connection to these economists—an understandable but, in my view, unfortunate instance of boundary definition.

All, as it happened (and was it mere happenstance?), encountered problems of knowledge before they took up economics, and all responded by developing what we may call evolutionary theories, based on the scarcity of cognitive powers, and envisaged the development of knowledge and capabilities (in Gilbert Ryle’s 1949 terms ‘knowing that’ and ‘knowing how’) through the formation and modification of selected connections, according to Knight’s principle of domain-relevant similarity. I shall not examine the relationships between their theories, but simply use some simi...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• List of illustrations
• List of contributors
• Economics in relation to other disciplines: an introduction
• PART I Economics in relation to the humanities and social sciences
• PART II Economics in relation to the life and natural sciences
• PART III Economics and mathematics
• PART IV Economics and architecture
• PART V Economics and geography
• PART VI Economics and sociology
• References