The Competitiveness of Nations

in a Global Knowledge-Based Economy

October  2003

AAP Homepage

Brian J. Loasby

Cognition and Markets

CRIC Workshop on Market Relations and the Competitive Process

Centre for Research on Innovation and Competition

Economic & Social Research Council

Manchester, UK

4-5 May 2000



Cognition and the Growth of Knowledge







Whether as an explanation of decision making or as a guide to making decisions, rational choice theory is not very interesting.  What is called a decision is merely the logical precipitate of the premises; everything that might be regarded as a determinant of choice is already in place, and assumed to be known (if only as a probability distribution) to the chooser, and so it is the explanation of these determinants – and why they, and only they, are determinants – that deserves attention.  Now this should not be a source of complaint, for, paradoxical as it may seem, choice theory is not about decisions.  Its purpose is to provide an essential element in constructing theories of equilibrium, and equilibrium in economics is routinely defined as a state of affairs in which there are no choices to be made.  In models of multi-period equilibrium, such as two-stage games, the conditional choices in later periods have to be fixed in the process of proving the existence of an equilibrium set; and if there are multiple equilibria, choice theory provides no criteria by which agents may choose between them.  Within choice theory agents make no choices.

When choices are deduced, only premises matter; and it is standard practice to explain differences in behaviour, including changes in behaviour over time, by differences in the premises from which behaviour is deduced – usually differences in opportunity sets, and often with specific emphasis on incentive structures.  One might therefore have expected economists to have shared Herbert Simon’s view that the premise should be the unit of analysis.  The reason that they do not is that decision premises are assumed to reflect precisely the fundamentals of economic analysis; they are therefore innocuous.  (There are a few exceptions, such as the theory of speculative bubbles.)  It is therefore not accidental that few economists pay specific attention to organisational forms, because in equilibrium models any organisational structure must be transparent to the basic data; and it is notable that orthodox and quasi-orthodox economists who have tried to explain why firms, as organisations, exist have taken care to isolate their explanation from the theory of production, which continues to be directly based on the supposedly fundamental data of the economy.  As Coase (1991, p. 65) has pointed out, this has led economists “to neglect the main activity of a firm, running a business”.  In a rational choice equilibrium, running a business is a trivial activity.

What prompted Simon’s proposal to focus on decision premises was his belief that they are problematic and yet capable of investigation by observation and experiment, and in particular his belief that organisational design is a significant influence on the premises that are used in various parts of an organisation.  To understand organisational behaviour, therefore, it is not sufficient to postulate rationality in the peculiarly restricted sense that is used in much economic theory (though it is normally appropriate to postulate intelligent and purposeful behaviour); one needs to investigate the procedures by which occasions for decision are identified, options are sought and examined, and choices made; and this investigation should pay particular attention to the premises which guide these procedures.

In this paper my primary concern is not the firm but the individual, though we shall need to consider to the activities of firms.  But our brief excursion into organisational design

2 Index

was not irrelevant, because in considering the individual as decision maker we shall pay particular attention to the ways in which individual knowledge is organised, and to the decision premises which shape both the development and use of that individual knowledge.  In accordance with the title, we shall give some emphasis to the influence of markets on the growth and use of individual knowledge and also note how knowledge within firms may benefit from market relationships.  We begin by identifying some basic elements of human cognition – not in order to produce a psychological theory, but to provide a credible psychological basis for our economic reasoning.


Cognition and the Growth of Knowledge

The principle of biological evolution is that of genetic selection for features which contribute to the fitness of each organism.  These features include both physical and behavioural characteristics, in a sorting process which necessarily extends over very many generations.  The standard biological model therefore assumes a stable selection environment, and implies the possibility of extinction if there is an unfavourable environmental change; we should not therefore be surprised that only a very small proportion of the species which have appeared on earth have survived, or that species are disappearing at a high rate in the present era of environmental change, much of it the result, sometimes intended and sometimes not, of human activity.  Our own species, of course, is not exempt from the possibility of extinction; but it does have a biological advantage in our evolved, and quite exceptional, capability to recognise significant change and to develop new behaviours in response.  This potential for learning may itself be interpreted as a biological adaptation to environments in which change is too rapid or too transitory to allow adequate genetic response through mutation and selection (Schlicht 2000); it provides real options to cope with uncertainty.

Though this capability might seem to be represented by the economic concept of rationality, it is much more than that, and much of the rationality involved is procedural rather than substantive (as Simon has pointed out), relying not on system-wide optimisation or strategies in the game-theoretic sense of conditional actions, but on the deployment of cognitive potential in the creation of localised structures.  The possibility of individual adaptation does not arise from the evolution of a general purpose programme within the brain which can be applied to all situations – which seems to be the conception implicitly underlying rational choice theory, especially when extended to rational expectations – but by the evolution of an architecture which is appropriate for distributed programming; this allows sensory inputs to be assembled into distinctive patterns, each of which triggers effective actions, in the process of interaction with specific environments.

There are two linked evolutionary reasons for this architecture.  The first reason is that genetic evolution proceeds by successive adaptations, each step of which has to be compatible with (even if it does not positively favour) survival, and the step by step construction of a general problem-solving programme would entail many stages in which the use of the incomplete system would be likely to lead to disastrous mistakes.  The second reason is that brain tissue has particularly high energy requirements.  The


emergence of homo sapiens has been marked by a very substantial increase in brain size, and therefore by an increasing need to economise on energy demands.  We should not therefore be surprised that multiple cognitive systems, each adapted to a limited function, and switched on only when required, proved to be the successful pathway to improved cognition, rather than a central processing system which would need to be permanently engaged.  The division of labour is a fundamental principle of development, in the brain as well as in society; and the network architecture of the brain made such a division possible.

These capabilities for adaptation were well advanced before the development of consciousness; and consciousness appears to have emerged, as one might expect, from this architecture and the functions that it supported.  If the unconscious brain collects impressions into categories, and associates each category with a particular action, it is natural that consciousness should begin with awareness of some of these categories, and in particular with a perception that one of the categories presently employed is not appropriately defined, or is linked to actions which are not particularly effective.

(The resemblance to the Carnegie language of aspirations and achievements is not accidental.)  The response to dissatisfaction would then be a search for better ways of ordering sensory impressions or for better connections between impressions and actions.  It has been quite frequently observed that we consciously seek to make sense of our environment by imposing patterns upon it, thus aiming deliberately to improve upon tacit skills by codification, or sometimes by giving a conscious direction to a process which remains beyond any detailed conscious control; and it is one of the most remarkable features of modern formal economics that its models of human behaviour make no reference to this characteristic activity of pattern-making, both conscious and unconscious – even though modern formal economics is itself an impressive example of the drive to impose patterns.  Consequently what is called ‘rationality’ within modern economics is a rather small, if important, part of human cognition, and makes a limited, if important, contribution to the explanation of human behaviour.

The two stages of development of human cognition which have just been outlined were presented as an evolutionary hypothesis in a paper written by Alfred Marshall (1994) as a young Fellow of St John’s College for a Cambridge discussion group.  The significance of this paper for both the content and method of Marshall’s subsequent work as an economist was identified by Tiziano Raffaelli; and it has been suggested (Butler 1991) that the motivation for writing it was supplied by a double crisis of religious and mathematical faith, the common feature of which was the perceived failure of axiomatic reasoning to guarantee empirical truth.  The mathematical crisis was precipitated by the invention of non-Euclidean geometry; the point was not that Euclidean reasoning was wrong, or even inapplicable, but that its applicability was a distinct issue, which could not be settled by appeal to the quality of reasoning; as Ziman (1978, p. 99) warns us, ‘scientific knowledge cannot be justified or validated by logic alone’.  (This was also the lesson of Marshall’s later encounter with Cournot’s argument that falling costs entailed monopoly – a lesson that was entirely lost amid the reinvention of Cournot’s logic in the 1930s, although Dennis Robertson seemed to have a sense that there was something wrong about the assumed relationship between logical and empirical truth.)  Marshall’s response,

4 Index

like Hume’s response to his own argument that it is impossible to prove any general empirical proposition, was to turn to the question of how people acquired what we call knowledge, and Marshall did so by drawing on Darwin’s principles of evolution to sketch an evolutionary psychology.  Caution in applying the patterns of axiomatic reasoning which had proved inadequate, and use of the evolutionary principles of the organisation of knowledge are persistent themes of his work as an economist.

Marshall appears to have been unaware of Hume’s work, or of Adam Smith’s (1980) remarkable development of it in his psychological theory of the growth of science.  As we have observed, the basic, and very powerful, human cognitive skill is the creation and application of patterns (of which logical reasoning is a special case); and Smith characterises scientific activity as the invention of ‘connecting principles’ which provide a credible account of relationships between phenomena.  For this cognitive skill to be put to effective use it is important that people should be both sensitive to patterns and motivated to seek out or manufacture patterns.  Not the least of the merits of Smith’s analysis is the attention that he gives to motivation, and what that motivation is.  The driving force is the link between cognition and emotion.

Smith begins by identifying three human emotions: surprise, wonder (what we might now call perplexity) and admiration.  An event or impression which does not conform to our expectations is surprising, and surprise causes discomfort; if we can find no way of accommodating it to our cognitive categories we are conscious of a failure to understand, which is at least disconcerting and may be dangerous; we therefore feel an urgent need to find a new way of making sense of our environment.  “Wonder, therefore, and not any expectation of advantage from its discoveries, is the first principle which prompts mankind to the study of Philosophy, of that science which pretends to lay open the concealed connections that unite the various appearances of nature” (Smith 1980, p. 51).  From an evolutionary perspective, we can easily see that the emotive power of ‘wonder’, in Smith’s sense, might sometimes be much more effective than ‘an expectation of advantage’ in promoting fitness-enhancing cognitive activity and thus improving adaptation to novel circumstances.

As we have already noted, sense-making is achieved through the invention of patterns, or new cognitive structures, which Smith carefully distinguishes from the perception of empirical truth; and if we are successful, either through our own efforts or by discovering a means which is already being used by someone else, then our sense of relief produces admiration in proportion to the discomfort, thus, in psychologists’ terms, providing reinforcement to our new cognitive skill.  Thus “the repose and tranquillity of the imagination is the ultimate end of Philosophy” (Smith 1980, p. 61); and because this repose and tranquillity rests on human inventions, rather than revelations of the principles that are actually in operation, it is always liable to further disturbance, which stimulates further invention.  Hayek (1952) was later to note that the categories that we use to order sensory impressions may differ substantially from the categories which have been developed to accommodate the development of corresponding scientific knowledge, and suggested an evolutionary explanation which is compatible with the argument of this section.


In a very recent paper, Ekkehard Schlicht (2000) has associated both the motivation to resolve difficulties by inventing new patterns, and the criteria by which possible new patterns are assessed, with an aesthetic sense which, he suggests, often seems to take priority over more strictly instrumental concerns; as in Smith’s analysis, an emotive drive may be more powerful than the prospect of practical benefits.  Indeed, when people are consciously searching for a theory which will account for some puzzling phenomenon, or for an effective product design, they often work on the principle that ‘if it looks right, it is right’.  Schlicht provides some examples from modern science; and Smith’s detailed account of the succession of cosmological theories incorporates repeated attention to the significance of aesthetic criteria, for example in the motivation of Copernicus to incorporate the heavenly bodies within “a new system, that these, the noblest works of nature, might no longer appear devoid of that harmony and proportion which discover themselves in her meanest productions” (Smith 1980, p. 71).  In explaining the triumph of Newtonian physics, Smith (1980, p. 105) recognises the rhetorical power of its supreme aesthetic appeal; and in his Lectures on Rhetoric he uses this appeal in recommending that, in giving an account of some system, one should “lay down certain principles… from which we account for the several phenomena, connecting all together by the same chain” because “the Newtonian method is undoubtedly the more philosophical, and in every science… is vastly more ingenious and engaging” (Smith 1983, p. 146).  The incentives to which both Smith and Schlicht appeal do not coincide with the incentives to which economists typically give priority, though the latter may complement them; but they still appear to be important incentives to scientific investigation.  In this, as in other aspects of its practice, science is still an art.

Smith went on to observe that as science progresses, it tends to divide into specialisms, and these specialisms encourage attention to details which would otherwise not be noticed.  This closer observation increases the chances of perceiving anomalies which then provoke efforts to modify or replace patterns that would be acceptable to the non-expert; thus the division of labour within science leads to increases in scientific knowledge.  As we know, Smith (1976b) was later to found his theory of economic development on this principle, in an exemplary demonstration of “the Newtonian method”.  We can now see that what underlies the unique potential of the division of labour within human society, in comparison with specialisation within any other species, is the architecture of the evolved human brain, which is capable in principle of forming any of an enormous variety of networks, far more than any single brain can actually sustain.  By appropriate specialisation, therefore, a human community can generate far more knowledge than any single person is capable of.  Marshall added to Smith’s principle of the division of labour Darwin’s principle of variation in response to environmental differences, which had already been incorporated in Marshall’s evolutionary hypothesis (noted earlier) and underlay his insistence on the importance of variety among the firms in each industry.  Economic development is therefore the result of the division of labour, supplemented by variation within each field of knowledge, both acting through the capacity of the human brain to develop new cognitive networks and motivated by the human desire to make sense (and not just money) by imposing patterns.

6 Index


A necessary consequence of this process is the increasing interdependence of human society.  Most of the knowledge and the skills that each of us need are held by other people; and although knowledge and skills are non-rival in a narrow sense, because of the basic facts of human cognition they are rarely reusable without cost by those who do not already possess them.  The lowest costs are typically incurred in acquiring knowledge within familiar fields from those who seem to be better informed, or in improving performance skills, mental as well as physical, by adopting the practices of those who seem to perform better, because we can usually assume that we are already organising our knowledge in similar ways to those whom we are seeking to emulate.  Willingness to borrow ideas and methods from others who appear to be relevant models is a familiar human characteristic, explicable by its survival value (Schlicht 2000).  Nor is it difficult to understand why we should so often converge on a few models, as Smith realised; his psychological theory of science gives due attention to the factors affecting the diffusion of each cosmological system.  Our cognitive limitations provide the incentives to borrow; the cognitive limitations of others, by impelling them to make and use patterns rather than attempting to optimise using information that we cannot observe, makes it easier to understand their rules of behaviour, as Heiner (1983) argued; and our shared characteristics make it easier to imagine ourselves in someone else’s situation, and to tailor actions and communications to their perceptions, as Smith showed in both his Theory of Moral Sentiments (1976a) and Lectures on Rhetoric (1983).

When we do so converge on rules for deciding how to behave, the shared conventions and procedures may be called institutions; and our understanding of institutions is enhanced if we begin with an appreciation of human cognition, and the consequent advantages of adopting other people’s procedures for good reason – that they seem to work for them – even when our actions have no effect on anyone else (Choi 1993), rather than by postulating games between people without cognitive limitations – and often with no acknowledged limits on their information either, although the specification of the game often imposes artificial limits on the players’ possibility sets.  Schlicht’s focus is on the conventions that facilitate interaction, and in particular on the criteria for a good rule, and this leads him to emphasise the importance of “evaluation of an aesthetic kind, relating to formal features like symmetry, analogy, or good continuity” (Schlicht 2000, p. 3).  Such evaluations are also important in choosing good rules for our private procedural rationality.  As Schlicht points out, these features are not relevant to anyone who is deemed capable of optimising everything except the use of scarce cognitive resources, but in a changing environment optimisation on each separate occasion is not a sensible objective, even by the criterion of overall optimisation.  We may recall Schumpeter’s (1943, p. 83) proposition that “[a] system… that at every given point of time fully utilizes its possibilities to the best advantage may yet in the long run be inferior to a system that does so at no given point of time, because the latter’s failure to do so may be a condition for the level or speed of long-run performance”.  Schumpeter applied this proposition to “any system, economic or other”; it is a proposition about uncertainty and


discovery, which is as applicable to the individual as to the economy; indeed it is its applicability to the individual that underlies its application to the economy.

Schlicht identifies imitation (which is rarely a precise copy) as an effective mechanism for coping with changes which do not last long enough to permit adaptation at the genetic level, but long enough for knowledge which has been acquired by individual pattern-making to be reusable by others at relatively moderate cost.  Sometimes this reuse may be formally organised.  Frank Knight (1921), having defined uncertainty as a situation in which no existing pattern or procedure is sufficient, and the entrepreneur as someone who is willing to impose an interpretative framework on that situation, then suggests that this entrepreneur may find it quite easy to persuade people to accept his offer of employment rather than attempt to cope with uncertainty on their own.  (Their motivation may well correspond to that on which Smith based his theory of scientific development.)  The entrepreneur therefore exercises authority within his firm precisely in Barnard’s (1938, p. 163) sense, which locates the decision about what is authoritative with the recipient, not the source.

However, our analysis should indicate that, as Barnard’s definition permits, authority is by no means restricted to hierarchical relationships, or even to relationships within formal organisations.  Anyone whose instructions, advice, or example we accept in any particular context is a figure of authority within that context; and we would find it very hard to manage our lives without many such figures.  It is important to recognise the source of both institutions and authority in individual efforts to take advantage of the range of human potential in coping with individual cognitive problems and the urge to make sense of the complexity of our environment, because it provides a basis for understanding why the emergence of institutions which serve to co-ordinate interactions is often much easier than would be suggested by the multiplicity of equilibria in many game-theoretic models, and why people should be much readier to accept subordinate roles in organisations than might be expected from a model populated by cognitively self-sufficient agents.  It also explains the effectiveness of marketing institutions which match cognitive needs.



A great deal of effort has been devoted by economists to examining the efficiency of exchange.  However, it is not often that this examination begins by posing the question in an appropriately cognitive form: to what extent can the problems of differentiated knowledge be resolved through the exchange of knowledge which is embodied in goods and services?  To do so immediately raises questions about the design of goods and services and also about the arrangements for exchange – or, in other words, about the working of markets.  These are not necessarily distinct questions, for as Marshall (1919, p. 181) observed, “[p]roduction and marketing are parts of the single process of adjustment of supply to demand.  The division between them is on lines which are seldom sharply defined: the lines vary from one class of business to another, and each is liable to modification by any large change in the resources of production, transport or the communication of intelligence.”  New products may require the creation of new market institutions, and the possibility of creating new market institutions may suggest product

8 Index

redesign.  We are about to consider the working of markets, each of which will be treated as “a specific institutional arrangement consisting of rules or conventions that make possible a large number of voluntary transfers of property rights on a regular basis” (Ménard 1995, p. 170); and we shall focus on the ways that these rules or conventions reduce the costs of transactions by simplifying cognitive tasks.  But before doing so we should recognise the importance of transfers of knowledge which is not embodied in goods or services, and of arrangements for the development of knowledge, for these aspects of co-ordination are also often dependent on the working of markets.  The markets to be considered are not, of course, the unanalysed markets of so much microeconomic – and of almost all macroeconomic – theory.

The failure to analyse markets as institutional arrangements which structure processes is a natural, and almost inevitable, consequence of choosing to start the analysis of exchange with perfect competition, for competition cannot be perfect unless every potential buyer has equal and costless access to every seller, and vice versa.  The moment that one provides any structure to these market relationships competition is no longer perfect.  The methodological necessity of avoiding structure accounts for the extraordinary status of the ‘auctioneer’ as the organiser of the perfect market (for whom Walras has received undeserved blame): the auctioneer has neither preferences nor motivation, consumes no resources, is able to communicate with everyone without ever establishing any kind of personal contact and without cost to either party, and never makes a mistake; most extraordinary of all, the auctioneer is a monopolist who is trusted by everyone never to exploit his position, and always justifies this trust.  Exactly the same characteristics are attributed to the ‘social planner’, who is subject to the same methodological necessity.

Since these perfect markets deliver efficient allocations without cost, it is not surprising that economists found some difficulty in explaining the existence of firms, which used resources in doing what the auctioneer did for nothing; nor is it surprising that Coase found it necessary to assume that economic agents, if not the auctioneer, incurred some costs in using markets.  If Coase had been better trained in economic theory, he might have realised that costs of using markets would be difficult to reconcile with perfect competition, and responded by following Joan Robinson (1933) into her world of monopolies; if, however, he had been less well trained, or alternatively had been more familiar with the work of earlier economists, he might have addressed to markets the question that he posed about firms: if they cost something to operate, why are people willing to incur these costs?  Markets should be explained, not assumed.

The route to an explanation was provided by Carl Menger ([1871] 1976), who set out to analyse the structure of an economy as an increasingly complex system for meeting human needs.  Beginning with goods which can be directly applied to their owners’ needs, he introduced what was to become a characteristically Austrian concern with indirect ways of meeting these needs, such as making machines to make tools for growing crops which can then be turned into food.  In this analytical system, the exchange of what one already possesses for something which can be used in the production chain is itself an indirect way of meeting needs.  It is therefore not surprising, though it is significant, that Menger introduces exchange at the opposite end of the range from what has become the


standard practice, with simple bilateral deals, and that he also recognised that making a deal incurs costs.  We now have isolated exchanges but no market, and so have the possibility of explaining markets as a means for reducing the costs of exchange, if there is a demand for multiple exchanges.

It is the next step that is crucial.  Rather than exchanging what they already have, some people may begin to use what they already have in the production of goods which are intended to be exchanged with others; and if this plan is successful, it may be worth while acquiring additional goods, not for use, but as inputs into regular production.  Exchange now becomes an essential part of the business of improving one’s condition, and the efficiency of the exchange process itself becomes a constraint on this improvement, and therefore a matter for attention.  If the creation of a market can improve the efficiency of exchange, it is, in Menger’s terms, a good of higher order, and its value is derived, in the first instance, from the value of the exchanges that it makes possible through this improvement in efficiency – and ultimately by the value of the indirect contribution that it makes to satisfying human needs. 

The most obvious means of improving the efficiency of exchange is through the development of a generally-accepted medium of exchange, which removes the need to match the goods that are included in each transaction.  Menger concentrates on money, because it exemplifies the emergence of a pervasive institution as a result of individual initiative and the readiness (not necessarily immediate) of people to copy behaviour that seems to work effectively for others.  In the course of developing this argument, he does, however, point out that goods differ substantially in their marketability – that is, in the costs which must be incurred in the process of exchange.  Unless the producer has some special advantage, commodity production will be concentrated on goods with relatively high marketability, though, in accordance with Marshall’s observation, a change in product design or production methods may be undertaken in order to improve marketability.  (Canning and refrigeration provide two important examples).  However, there are always some costs of exchange.  Here the importance of indirect methods again comes to the fore, because the costs of achieving a single sale can often be substantially reduced by appropriate investment – by creating a set of institutions which constitute a market.  The customer might pay for this investment, but would naturally need to be compensated by a lower price; and it is usually the producer who expects to be most dependent on particular categories of exchange and therefore has the greater incentive to develop procedures which will facilitate exchange within those particular categories – in other words, to create a special market.  Casson (1982), who has given particular attention to the firm as market-maker, provides a convenient summary of the obstacles to trade, and of the ways in which the producer may seek to overcome them: all these ways reduce the customer’s cost of making a single transaction (and often the producer’s cost of a single transaction) by a substantial initial commitment of resources.

The obstacles to trade discussed by Casson are primarily the result of deficiencies of knowledge; and so are the costs invoked by Coase (1937) to explain the creation of firms – though Coase did not see that firms make markets as well as goods.  Both require knowledge to be organised; this is a cognitive issue, to be handled by procedural, not

10 Index

substantive, rationality.  Costs per transaction are reduced by the development of a set of conventions about the way in which it is to be carried out, including, for example, the location, the provision of information, opportunities to inspect or try out the product, standards for product characteristics, methods of payment, provision for servicing, and the commitments that are implicit in the deal.  Because of the standard use of preference functions in consumer theory, it is important to mention that the use of money not only makes payment straightforward; the use of money prices also simplifies the assessment of each potential purchase.  The ‘measuring rod of money’ is among the most important of conventions.  The effect of these institutions is to ease the cognitive task of making a satisfactory purchase (and also the cognitive task of making a satisfactory sale).  This cannot be achieved without a cognitive investment, and though the cost of this may fall mainly on the producer, customers must develop their own new routines.  Everyone has to learn how to shop in a supermarket, how to buy through mail order, how to buy PEPs or ISAs, and know how to use the internet effectively.  This learning produces potentially reusable knowledge; consequently everyone who makes this cognitive investment experiences the equivalent of what Penrose (1959) called “the receding managerial limit” as their new capabilities become embedded in new patterns of behaviour and therefore release cognitive resources for other tasks, and especially tasks which can incorporate the new routines.  On the other hand, significant changes in institutional arrangements, such as the replacement of PEPs by ISAs and, on a much larger scale, the development of internet markets, impose new cognitive burdens.

The importance of simplifying the customer’s cognitive task is illustrated by considering those markets in which many customers are willing to go to a good deal of trouble to get what they want, and where the common practice of supplier-fixed price does not apply.  Three outstanding examples are the markets for houses, cars and collectables.  Here there are two powerful incentives for the customer to incur substantial transaction costs: purchases typically involve a substantial outlay, and they have substantial implications for the customer’s style of living, and even for self-image.  Such cognitive structures as life-style and self-image are a natural consequence of the pattern-making characteristics of the human brain which have been considered earlier, and the maintenance of these structures will normally be a major objective.  It is therefore worth taking a good deal of trouble to ensure that any major purchase is compatible with the relevant patterns – or, occasionally, that an available purchase has sufficient appeal to justify some modification of those patterns.  Rules for making such decisions have a substantial aesthetic component, even when the opinions of family, friends, colleagues, or neighbours are not of direct concern.

When there is substantial diversity among customers, the greatest reduction in transaction costs for a particular group may be achieved by the creation of sub-markets in which the rules and conventions, as well as the products, are precisely adapted to that group.  This is yet another instance of the effects of the division of labour, which as always is limited by the extent of the market; but the extent of the market is influenced by the incremental cost of transacting in it.  Firms, like individuals, differ in their capabilities, including their capabilities in devising institutional arrangements for reducing particular kinds of transaction costs for particular groups of potential customers, and these differences are


reflected in what they attempt and (even more) in what they achieve.  But thinking of the ways in which firms can develop arrangements that enable them to sell more easily should not distract our attention from the contribution of these arrangements to the firm’s knowledge about its business.  They are an important part of what Marshall called its external organisation; and we have already drawn attention to the pervasive importance of externally-organised knowledge to augment our own cognitive skills.

However, the emphasis in this paper is on consumers.  Market institutions are an important part of consumers’ external organisation; the rules and conventions form part of their interpretative system and their framework for decision making.  How we decide is not without influence on what we decide; and the power of market-making firms to shape the way that we decide seems to be well appreciated in many businesses, as is manifest in a brief observation of marketing strategy.  This power is often attenuated by the efforts of rivals; as Hayek (1948, p. 106) pointed out, competition is a “process of formation of opinion”, or in language used earlier, a means of deciding whose communications will be provisionally accepted as authoritative.  There is no choice without a framework, and as in Penrose’s theory, the routinisation of procedures makes possible the acquisition or creation of new knowledge and new skills.  If the process of transacting is easy, we can concentrate on what to transact, and how to develop new skills as consumers – or indeed as humans.  Our cognitive resources, like the productive and managerial resources of the Penrosian firm, are capable of development, and when developed may deliver new productive services; we have therefore an incentive to learn about the possibilities and the opportunities (Penrose 1959, p. 77) or what we might call their option value.  We may be encouraged to do so by the recognition that the array of markets which we can use without much effort also have an option value for us; they allow us relatively easy access to a wide range of other people’s capabilities.

We should not forget that the greater part of the value of transactions within a modern economy has no direct reference to customers at all; and the same incentive to reduce the costs of individual transactions by substantial investment in developing appropriate institutions applies here.  Business customers may have stronger incentives than their suppliers to simplify the process of transacting, and this may not merely be a matter for the purchasing department; for production is closely linked with purchasing as well as marketing.  Arrangements with suppliers may reduce the costs associated with production, and modifications to product and process may simplify relationships with suppliers.  Amid the flurry of excitement about e-commerce, there have been several predictions that its impact may be greater on relations between businesses than on retailing; if this is so, it will be because of changes not only in the handling of transactions but because of associated changes in the internal operations of firms.



We make sense of our situation by constructing order; the quality and applicability of our knowledge capital, as of physical capital, depends on its structure.  But even in our private lives we are not dependent on our own resources; our common mental architecture makes it possible to connect to external knowledge capital, though by no means to all of it.  

12 Index

Market relationships provide links which make possible the construction of additional knowledge, and by allowing many activities to be guided by rules and conventions, provide scope for imagination.  Co-ordination by institutional arrangements define boundaries, which should lightly be assumed to match the fundamental data – which is not known anyway; but from a cognitive perspective it also defines a space to think within those boundaries – and even to make selective forays across them.  The history of human development (which is only part of human history) is one of discovering ways of circumventing the limits apparently imposed by the cognitive capacity of each individual, by an increasing division of cognitive labour and by developing procedures which enable us to make new combinations from the results of that division.



Barnard, Chester A. (1938), The Functions of the Executive, Cambridge MA, Harvard University Press.

Butler, Robert W. (1991), ‘The historical context of the early Marshallian work’, Quaderni di Storia dell’Economia Politica, 9, 2-3, 269-86.

Casson, Mark (1982), The Entrepreneur: an Economic Theory, Oxford, Martin Robertson.

Choi, Young B. (1993), Paradigms and Conventions: Uncertainty, Decision Making, and Entrepreneurship, Ann Arbor, University of Michigan Press.

Coase, Ronald H., (1937), ‘The nature of the firm’, Economica, N.S. 4, 386-405.

Coase, Ronald H. (1991), ‘The nature of the firm: influence’, in Oliver E. Williamson and Sidney G. Winter (eds), The Nature of the Firm: Origins, Evolution, and Development, Oxford, Oxford University Press, 61-74.

Hayek, Friedrich A. (1948), ‘The meaning of competition’, in Individualism and Economic Order, Chicago, University of Chicago Press, 92-106.

Hayek, Friedrich A. (1952), The Sensory Order, Chicago, University of Chicago Press.

Heiner, Ronald A. (1983), ‘The origin of predictable behavior’, American Economic Review, 73, 560-95.

Knight, Frank H. (1921), Risk, Uncertainty and Profit, Boston, Houghton Mifflin; reprinted Chicago, University of Chicago Press, 1971.

Marshall, Alfred (1919), Industry and Trade, London, Macmillan.

Marshall, Alfred (1994), ‘Ye machine’, Research in the History of Economic Thought and Methodology, Archival Supplement 4, Greenwich CT, JAI Press.

Ménard, Claude (1995), ‘markets as institutions versus organizations as markets? Disentangling some fundamental concepts’, Journal of Economic Behavior and Organization, 28, 161-82.

Menger, Carl ([1871] 1976), Principles of Economics, translated by J. Dingwall and B. F. Hoselitz, New York, New York University Press.

Penrose, Edith T. (1959), The Theory of the Growth of the Firm, Oxford, Basil Blackwell; 3rd edn Oxford, Oxford University Press, 1995.

Robinson, Joan V. (1933), The Economics of Imperfect Competition, London, Macmillan.


Schlicht, Ekkehart (2000), ‘Aestheticism in the theory of custom’, Journal des Economistes et des Etudes Humaines, 10, 1.

Schumpeter, Joseph A. (1943), Capitalism, Socialism and Democracy, Allen and Unwin.

Smith, Adam (1976a), The Theory of Moral Sentiments, ed. David D. Raphael and Alec L. Macfie, Oxford, Oxford University Press.

Smith, Adam (1976b), An Inquiry into the Nature and Causes of the Wealth of Nations, ed. Roy H. Campbell, Andrew S. Skinner and W. B. Todd, 2 volumes, Oxford, Oxford University Press.

Smith, Adam (1980), ‘The principles which lead and direct philosophical enquiries; illustrated by the history of astronomy’, in Essays on Philosophical Subjects, ed. William P. D. Wightman and J. C. Bryce, Oxford, Oxford University Press, 33-105.

Smith, Adam (1983), Lectures on Rhetoric and Belles Lettres, ed. J. C. Bryce, Oxford, Oxford University Press.

Ziman, John M. (1978), Reliable Knowledge, Cambridge, Cambridge University Press.



The Competitiveness of Nations

in a Global Knowledge-Based Economy

October  2003

AAP Homepage