However, transhistorical principles are not enough. Theoretical analysis must build a conceptual bridge with a real object. The theoretical analysis of a real object must begin by abstracting its basic features. Abstraction entails the identification of what is essential to, and enduring in, an entity, ignoring the accidental and superficial. More fundamentally, the identification of features, relations and structures depends upon acts of taxonomy and classification, involving the assignment of sameness and difference. Classification, by bringing together entities in discrete groups, must refer to enduring common qualities. However, as Alfred Whitehead (1926) argued, abstractions are essential but they always do some violence to the complex, changing reality. Also, as Nicholas Georgescu-Roegen (1971) has insisted, operational concepts have a contradictory or dialectical quality and cannot correspond precisely with real phenomena. All acts of categorisation and abstraction must, therefore, be provisional. All theoretical foundations must forever be under scrutiny.
The problem is to develop meaningful and operational principles of categorisation on which analysis can be founded. These principles must identify relatively durable entities, from which categorisation and analysis may proceed. The neoclassical and Austrian approaches thus start from the given individual. A crucial problem in this approach, as we have seen in preceding chapters, is that it does not account for the evolution of individuality itself. In contrast, the institutionalist tradition has a different answer to this problem, locating invariances in the (imperfect) self-reinforcing mechanisms of (partially) stable social institutions.
The institutionalist Walton Hamilton (1932, p. 84) defined an institution as a way of thought or action of some prevalence and permanence, which is embedded in the habits of a group or the customs of a people. ... Institutions fix the confines of and impose form upon the activities of human beings.
Because of their relatively stable and self-reinforcing qualities, institutions are chosen as relatively invariant units. Institutions have a stable and inert quality, and tend to sustain and thus 'pass on' their important characteristics through time. Institutions are both outgrowths and reinforcers of the routinised thought processes that are shared by a number of persons in a given society.
The power and durability of institutions and routines are manifest in a number of ways. In particular, with the benefit of modern developments in modern anthropology and psychology it can be seen that institutions play an essential role in providing a cognitive framework for interpreting sense data and in providing intellectual habits or routines for transforming information into useful knowledge (Hodgson, 1988). The cultural and cognitive functions of institutions have been investigated by anthropologists such as Mary Douglas (1987) and Barbara Lloyd (1972). Reference to the cognitive functions of institutions and routines is important in understanding their relative stability and capacity to replicate. Indeed, the strong, mutually reinforcing interaction between social institutions and individual cognition provides some significant stability in socio-economic systems, partly by buffering and constraining the diverse and variable actions of many agents. Institutions become cumulatively 'locked in' to relatively stable and constrained paths of development.
Hence the institution, as Philip Mirowski (1987, p. 1034n.) has suggested, may be regarded as 'a socially constructed invariant'. On this basis, institutions can be taken as key starting units in any concrete analysis. This contrasts with the idea of the individual as the irreducible unit of analysis in neoclassical economics, and applies to both microeconomics and macroeconomics. How is this principle manifest? It would suggest, for instance, that theories based on aggregates become plausible when based on corresponding social institutions. Money is a legitimate unit of account because money itself is an institutionally sanctioned medium; aggregate consumption functions should relate to a set of persons with strong institutional and cultural links; and so on. Again this contrasts with the approach based on reasoning from axioms based on the supposed universals of individual behaviour. An approach based largely on institutional specifics rather than ahistorical universals is characteristic of institutional economics, and has parallels in some of the economics of the Marxian and Post Keynesian schools.
Of course, this does not mean that institutions are regarded as immutable. Institutions themselves may change. We may consider 'institutional evolution' as a process, and even use biological metaphors in such a discourse. But when compared with biological evolution, institutions have nothing like the degree of permanence of the gene. What is important is to stress the relative invariance and self-reinforcing character of institutions: to see socioeconomic development as periods of institutional continuity punctuated by periods of crisis and more rapid development.
Notably, institutions fill the key conceptual gap that we have identified in neoclassical, Austrian and Marxian theories. Institutions simultaneously constitute and are constituted by human action. Institutions are sustained by 'subjective' ideas in the heads of agents and are also 'objective' structures faced by them. The concept of an institution connects the microeconomic world of individual action, of habit and choice, with the macroeconomic sphere of seemingly detached and impersonal structures. Actor and structure are thus connected in a spiral of mutual interaction and interdependence:
The economic life history of the individual is a cumulative process of adaptation of means to ends that cumulatively change as the process goes on, both the agent and his environment being at any point the outcome of the last process.
The above remarks on the analytical significance of institutions are general and ahistorical. Yet we have moved from the general principles of evolutionary and systems theory towards a type of theoretical framework which can be used to categorise and dissect specific economic structures. There is much more methodological and theoretical work to be done here, but ostensibly the gap can be filled by institutional economics. This theoretical work is far from complete but institutionalism does seem to offer a favourable basis for further theoretical development, with its core concept of an institution and its deployment of the evolutionary metaphor. The grounds for this judgement include considerations such as the following.
Notably, the very concept of an institution points from the sphere of general principles to the study of the specific. Although some general principles regarding institutions can and have to be established, these tell us very little about the nature and dynamics of specific institutions. However, the approach being advocated here is not the 'add-social-context-and-stir' method found in some 'economic sociology' and rightly criticised by Viviana Zelizer (1993, p. 194). Institutions are not merely constraints, bearing upon a pre-existing and 'non-institutional' economy or market. Institutions and culture are not the context of a pre-given economy because the economy is not pre-given without institutions and culture. Economies and markets are themselves constituted as collections of institutions, and are not merely constrained by them. The market, for instance, as Harrison White (1988, p. 232) rightly stressed, is 'intensely social - as social as kinship networks or feudal armies'. The market, in short, is itself an institution (Hodgson, 1988).
Consider the theory of market price as an illustration. The concept of relative price in institutionalism is quite different from that in both Marx and the neoclassicals. Marx relied on the labour theory of value. Neoclassical economics proceeds from the concepts of supply, demand and marginal utility. By contrast, in institutionalism prices are social conventions. Such conventions are various and reflect the varied types of institution, mode of calculation, pricing process and commodity under capitalism. As Richard McIntyre (1992, p. 47) noted: 'There is no "general" theory of price because there is no general process of price formation.' Nevertheless, some common features of the price formation process in modern capitalism, such as markets, consumers and capitalist firms, can be used to develop some general outline principles, just as an even more general theory of conventions can be elaborated.12
Institutional economists have rightly argued that it is essential to focus on specific institutions and to understand their nature and dynamics. However, there is a danger that the need for general theories and concepts is lost in the search for specificity. In the past, the concentration on the specific and concrete, to the detriment of abstract and systematic theory, became a problem in institutional economics in the 1930s when according Gunnar Myrdal (1958, p. 254) - himself a strong devotee of institutionalism - it degenerated into datagathering and 'naive empiricism'. Having failed to develop its theoretical foundations sufficiently, institutionalism cracked and subsided.
There are clearly two temptations to be avoided here. One is to erect an ahistorical theory: 'theory without data'. The other is to eschew theory and a system-building for data-gathering: 'data without theory'. But it must be emphasised that this is not a matter of finding a golden mean between such extremes. The very ideas of 'theory without data' and 'data without theory' are misconceptions. They are both false navigational poles. It cannot be a question of the appropriate mixture of the two basic ingredients of theory and empirics because data cannot be considered or appraised independently of a theory. All statements of fact are theory-laden. All attempts to gather data are informed unavoidably by a set of classificatory concepts and implicit or explicit theories. As well as the importance of concrete data, the primacy of theory has to be emphasised.13
Clearly, institutional economics needs to be further developed to deal with the important issues raised here. This requires methodological work and conceptual analysis to supplement the foundational work of Veblen and other early institutionalists. An important supplementary idea discussed here is the impurity principle.
Was this article helpful?