1.1 The question of institutions

Over the years many voices of dissent have been raised against economics and the earlier political economy for seemingly failing to pay due attention to the institutions of specific societies and hence to the actualities of economic life. When critics, from Richard Jones to J.K. Galbraith, have produced works of their own which in their estimation have been free of this defect, they have never made more than a very modest impression on the mainstream of economic theory. Sometimes, as in the case of the German historical economists and to a lesser extent in that of the American institutionalists, these critics gathered for a time a following of their own and, among other things, contributed to the use of economic statistics and advanced the cause of social security legislation. But a single discipline which does justice both to the logic of economic relations and to empirical accounts of the institutional framework remained, except to convinced Marxists, an elusive ideal. The whole issue was debated impassionedly in the lengthy ‘Methodenstreit’ begun by Menger and Schmoller, but no consensus was reached. There has remained to the present day what Eucken called the great antinomy between the individual-historical and general-theoretical approaches.

It may not always be easy to see what is at issue. Theory, it may be said, is surely not an end in itself, but the means for gaining a better understanding of a country’s economy and its problems and, where applicable, for formulating appropriate policies. The institutional peculiarities of a country do enter this analysis as factors that qualify the conclusion. Theory may even be used to study the evolution of an institution, so that theory is the tool and the understanding of an institution the end product. In either case there is no conflict between the use of theory and the recognition of institutions. This is, very broadly, the way Menger presented his case in the Methodenstreit. Yet it does not really get at the heart of the matter.

One could perhaps interpret the institutionalist position as follows. Economic theories in themselves presuppose certain institutions (though Menger possibly thought that the ‘exact’ discipline of economics did not). Ricardian political economy, for instance, required a certain system of land tenure and a society whose members fell naturally into three categories. Modern general equilibrium theories require markets, a law of contract, private property, and so on, and perpetrate, according to some views, a subtle sin of omission in failing to recognize that tastes or preferences arise out of the evolving institutional structure. Now, the institutionalist critique has pertained to these presuppositions or implicit institutions and has wanted to see them replaced by the explicitly recognized institutions – those that merely qualify the conclusions of analysis or are the end products of analysis. It is the integration of theory and institutional fact that has not been accomplished, though some would claim that it was precisely Marx’s achievement to have done so. Most theorists, it seems, are not opposed in principle to such an integration. They simply do not know how to bring it about. Some attempts in that direction have of course been made. The theories of monopoly and monopolistic competition are examples, as are the many variants in which equilibrium theory may be had with or without forward markets, information costs, and so on. But this falls very far short of what institutionalists have considered necessary.

However, the interpretation of institutionalism given here is not really in the spirit in which most of the critics in question wrote. In fact, they showed so little uniformity that it is difficult to generalize. The only traceable movement of this critical thought started with the historical school in Germany and eventually spread rather weakly to Britain and much more strongly, in the form of institutionalism, to the United States. Common to all in this movement, as it is to modern institutionalists, is a predisposition to see economic questions, in the way of 19th-century thought, in terms of cultural evolution and development. But beyond this, and even in this, there were great variations. Roscher and Hildebrand played with the idea of laws or stages of economic development, but Knies played down the expectation of finding laws in social evolution. Under Schmoller’s guidance the movement in Germany turned towards a purer economic history, though the idea of finding common threads was not given up. It also influenced Max Weber’s sophisticated historical analysis of institutions, but Weber was also part of a broader movement which included, among others, Dilthey and Croce. In England (more strictly, Ireland), Leslie made the emphasis on history and evolution into the injunction: ‘Back to Adam Smith’. He and Ingram also brought the ideas of Comte into the subject. In the United States, the movement took the quite disparate forms of Veblen’s bitter social critique, Commons’s study of the legal foundations of capitalism and Mitchell’s painstaking statistical analysis of business cycles.

In many cases, writers did not go far beyond simply justifying their preoccupation with historical fact and statistics by criticizing the classical school for relying excessively on abstraction, deduction and the assumption of self-interest. They then described their own work as realistic, inductive and cognisant of a variety of ethical standards. Early on many had earned themselves the designation ‘Kathedersozialisten’ (socialists of the professorial chair) because of their advocacy of state intervention in the economy and these leanings were later evident also among the American institutionalists. But by no means all spoke from the direction of the political left. Often they merely voiced objections to what they saw as an excessive academic preoccupation with markets and price theory. One spokesman for American institutionalism, referring to the meaning of the term ‘the economy’, said: ‘Institutionalism proposes to find that meaning in the interplay of institutions and technology … just as classical theory has sought the meaning of the economy in the interplay of wants and scarcity.’1 It may therefore be a fair comment that the institutionalist critique of economics has been fairly amorphous and that in effect its many expressions have had in common only the wish to see what was described earlier as an integration of theory and institutional fact. Since this sentiment had its heyday in the second half of the 19th century, it is perhaps natural that it should have been channelled into the evolutionary, developmental approach which many then considered to be the most respectable scientifically, in contrast to the more analytical bias of the 18th and early 19th centuries, to which there was a return in the 20th century.

This study will have as its raison d’être the sentiment which inspired institutionalists in the past, but it will not be based on the institutionalists’ most usual approach, that is, the close attention to cultural evolution. Instead, the approach will be wholly analytical. The objective will be to develop a conceptual framework in terms of which it can be shown, first, how a knowledge of institutions fits into our understanding of economic questions and, second, why the presuppositions of micro-economic equilibrium theory make the accommodation of received theory to the institutional peculiarities of specific eras and areas so very difficult. For these purposes it will be necessary to delve into some epistemology. The relation between theory and the particular events of our everyday experience will be considered in some detail. It will then be seen that the question of how institutions may be brought to the closer attention of theorists is a part of the wider question of what we are to understand by a fact in economics. This question is susceptible of two answers and it will be argued that the distinction between these two types of fact is vital to economics. The outcome of all this will differ greatly from the most usual institutionalist analysis. In fact, it will have a greater affinity with the views of the intellectual descendants of the other side in the Methodenstreit, who seem also to raise their voices in dissent, but within the community of theorists itself.

1.2 Three aspects of institutions

When economists speak of the institutional framework of an economy they seem to be referring to a set of very diverse entities. An institutional framework may include customs, usages, norms, attitudes and even fashions, a monetary system, a political constitution, a tax system, laws of contract, of inheritance and of land tenure, established means of collective bargaining and many more. It may seem difficult to find common elements in such diversity. However, for the purpose of this study three aspects of institutions are important. It must be stressed that they are important in the present context2 and not necessarily the most important in any context. Furthermore, since it may not be possible to consider under these three aspects anything anyone has ever called an institution, they may also be taken to delimit the concept of an institution for present purposes.

1.2.1 Consistent conduct

An institution may be described in terms of a possibly unlimited number of statements about consistencies in the conduct of individuals.

Consistent conduct must here be understood as a regular (or fairly regular) conjunction of a type of conduct with a type of situation, that is, an individual’s conduct is consistent when he regularly does something, or refrains from doing something, when a certain situation arises. However, this should not be taken to imply that people sometimes behave like automatons, or that institutions are based on conditioned reflexes. Customary or conventional conduct, such as serving refreshments to visitors or making payment by cheque rather than by some other means, is quite compatible with premeditated and purposeful action. Consistent conduct could be rephrased as conduct appropriate to certain occasions or circumstances. Using the means-ends terminology, one could then say that certain ends are appropriate to certain occasions in the sense that the occasions present opportunities for doing something agreeable, or call for conduct considered correct, kind, polite, in good taste and so on. When ends are unrelated to occasions, it may nevertheless be appropriate to use conventional means for achieving them, because such means may seem to be the most efficient, most convenient or least risky under the circumstances. In fact, it has been conjectured, as in the conjectural history of money, that the means for achieving some types of ends become institutionalized because they are the most efficient and that thereafter they continue to be the most efficient because they are institutionalized.3

Here, the word appropriate has of course been used in different senses, namely, appropriate according to sentiment and logically appropriate. In action, however, the two senses may often become intertwined. For instance, a man may be on his best behaviour not only because he feels that certain occasions call for certain conduct, but also because he wants to make a good impression in order to obtain a more remunerative position in order to give his family a better life and so on. He may therefore consider his conduct appropriate in both senses even though he may know of quicker ways of giving his family a better life. More generally, in calculating whether x or y is the more efficient means of achieving z, one may have to take into account that x is, say, the more socially acceptable or simply the more decent of the two means and that one values being accepted socially or being decent. In other words, the appropriateness according to sentiment may be a factor in calculating costs and logical appropriateness, so that x may be appropriate in both senses.

However consistent conduct may also include behaviour which one would perhaps hesitate to call premeditated and to which the concepts of means and ends would therefore be inapplicable. Customs, like habits, may be followed without thought. For instance, an office worker may have a cup of tea at every break. If this has continued for many years one may not be inclined to say that the worker plans to have it or even wishes to have it. The worker simply does what is appropriate to the occasion. However, uncontrollable behaviour, like flinching one’s eye during an eye examination, would clearly not be regarded as institutionalized.

It is not difficult to see that customs, usages and norms may be described in terms of what people are likely or are unlikely to do in certain circumstances. But even, for example, the institution of banking may be described in terms of what, under certain circumstances, is likely to be done by a client with funds coming in and with debts to be settled, by a teller doing his daily work, by a businessman wanting to expand his stocks, by a bank manager considering an application for a loan and so on.

Of course, this would not describe how the appropriate conduct of the various parties fits together, that is, how the banking system functions. In this case the functioning of the whole system may be deduced from the institutionalized behaviour (or appropriate conduct) of the various parties and it is no doubt part of the business of economics to make such deductions. However, it may be argued that the functioning or overall order of a system must be distinguished from what is actually institutionalized, since there are some cases in which the functioning is clearly not part of what is meant by an institution. It may be said, for instance, that the telephone is an institution in developed countries. To understand what is meant by this, one does not have to know how the electronic equipment works, how repairmen locate faults or how some of the funds paid by subscribers find their way into the repairmen’s pockets at the end of each month. All that is meant is that a large number of people find it appropriate to reach for a telephone when they want to communicate with someone some distance away.

It could be argued in an analogous way that the institutional part of market forms also comes down ultimately to ideas on appropriate conduct. In the case of a produce market which in some respects comes close to the model of perfect competition, the institutional basis is surely not the mere fact that there are many farmers, places where buyers and sellers can meet and so on. Rather, it is that farmers want, or think it fit and proper, or simply regard it the natural state of affairs, to farm individually rather than to combine into corporations or communes, and to compete against each other rather than to club together in selling. Of course, technical agricultural factors may be involved as well, in which case the market does not have a purely institutional basis.

Finally, institutions may also be described in terms of what is inappropriate conduct or what people are unlikely to do. In describing the political systems of Britain and the United States, for instance, it may be worth noting that a British or American general is unlikely to attempt to seize the government or, if he did contemplate such an action, that he is unlikely to find enough people to support him. In an analogous way one can say that the institution of private property entails that a buyer not only expects to be able to take physical possession of what he buys, but also expects others to (consider it appropriate to) refrain from using his purchase, even though they may desire it and it may be physically feasible for them to use it. If his expectations should be disappointed, then (on the positive side again) he would expect the police to regard it appropriate to act on his complaint.4

1.2.2 Consistent conduct as a fragment of the whole

The set of institutionalized or consistent conduct does not embrace all conduct. While an institution always implies consistent conduct, it also has the connotation that there is conduct which is otherwise.

It would make no sense to speak of institutionalized conduct in a world in which all conduct is institutionalized. If there are conventional ways of doing things, there must be also other ways of doing things. If there are customary greetings, gifts and fringe benefits, there must also be other greetings, gifts and fringe benefits. While it is possible to imagine a primitive society in which all conduct is determined by biological needs and ritual, the ritual would be called institutional only in contrast to the situation in more familiar societies. In other words, an institutional framework is not a comprehensive system. Institutionalized or consistent conduct must always be seen as conjoined with action that can be described as unusual, novel, creative, unique and so on. All this may be quite obvious, but the apparently partial consistency of conduct, as will be shown later, is something which economic theory has found very hard to handle.

1.2.3 Consistent conduct as empirical orientation

A knowledge of institutionalized or consistent conduct may be used as a means of empirical orientation, that is, institutions may be ‘points of orientation’.5

In the ordinary business of life, people use a knowledge of consistent conduct as an aid in interpreting the actions of others, that is, in making sense of the doings of their fellow men, or as a guide to action in planning their own action. When a knowledge of consistent conduct is acquired by induction for these purposes, it consists of empirical facts which are quite independent of theories about the function of institutions in a social order and of the question of why institutions are, or how they came to be, what they are.

1.3 On the empirical content of economics

It is a fair question whether economics can have an empirical content, or if it can, whether it has, or if it has whether the logical form of its empirical content is fully understood. Since institutions, viewed as empirical facts, are obvious candidates for filling any possible vacancies in this empirical content, the issues that may be raised in seeking answers to this question will form much of the subject matter of the following chapters. The preliminary remarks that will be made here will try to show, first, that the answers are at the very least not immediately apparent, second, that economics has to contend with particular difficulties in regard to knowledge and, third, that micro-economic equilibrium theory more or less ignores these difficulties.

1.3.1 Theory without empirical content

For much of its history the main body of economic theory was associated with the advocacy of free trade and of an unrestricted market mechanism. It is not immediately apparent whether there is an empirical aspect to an argument urging that certain measures be taken for the greater happiness of mankind. Certainly, in the hands of Adam Smith the analysis of a natural (and by implication a desirable) order was based on a keen study of institutions. We have the opinion of one of his students that the fourth and last part of his lectures on Moral Philosophy, which ‘contained the substance of the work he afterwards published under the title of … the Wealth of Nations’, considered ‘the political institutions relating to commerce, to finances, to ecclesiastical and military establishments’.6 In subsequent and more formal developments of theory the notion of natural order usually had some role to play, but its institutional basis receded and was almost gone by the time of the neo-classical equilibrium model. Enthusiasm for promoting the market economy seems to have waned as well; the notion of equilibrium is not necessarily associated with a desirable order and its debt to the reformist zeal of former times is largely overlooked. The result is an equilibrium concept which seems peculiarly unrelated to the particulars of actual situations or at least to any that anyone actually claims to know.

One may wish to see economics as the pure logic of choice. In that case, economics cannot have an empirical content. ‘Like logic and mathematics’, said von Mises, economics ‘is in us; it does not come from without’; ‘no experience, however rich’ could disclose it; it is derived from a ‘logical analysis of our inherent knowledge of the category of action’. Furthermore, in von Mises’s opinion, there ‘are no such things as a historical method of economics or a discipline of institutional economics’.7 Irrespective of whether one agrees with von Mises, one must admit that the logic of choice is valuable in handling factual material. Even an applied economist has said that in so far as economic theory is useful ‘in helping us to take decisions on policy, it is the simple, most elementary and in some ways most obvious propositions that matter’ and he mentioned, among other things, opportunity cost and the maxim that bygones are bygones.8 However, such usefulness presupposes that we know something to which the logic can be applied. It is said that the logic of choice needs special or subsidiary assumptions, some empirical input on which it can work and in this regard we are apparently left to our own devices. The only guidance that von Mises offered was that the ‘question whether or not the real conditions of the external world correspond to these assumptions is to be answered by experience’.9

One may of course suppose that statistical time series are the empirical raw materials of economics and provide economics with an empirical content. How the logic of choice may be applied to statistics is again not immediately apparent but in the case of equilibrium models there are at least some ideas, for the work of some econometricians seems to be based on the belief that statistics can be incorporated into equilibrium models. However, this raises another question. Are statistics themselves the empirical facts in which economists are interested, or does interest really lie in certain relations that are meant to be distilled from statistics? The methods of econometrics seem to indicate the latter interest. However, the distillation process has not proved to be an easy matter. Professor Hutchison has recently assembled a number of quotations in which eminent economists express their dissatisfaction with the position of empirical fact in their subject. Among them is the comment by Professor Leontief that: ‘In no other field of empirical enquiry has so massive and sophisticated a statistical machinery been used with such indifferent results.’10

Also in the factual field are studies, often with merely a superficial relation to theory and almost disparagingly called descriptive, which give an account of, say, the marketing of an agricultural product, the protection of an industry or the progress of an anti-inflationary policy and so on. Here also one may ask whether the various episodes of such reports constitute the facts in which we are interested or whether there are lessons to be learnt (or distilled) from history. The point at issue is the same as that raised in the case of statistics, because statistical time series also record aspects of the events of the past.

1.3.2 Empirical constraints versus empirical past

There is something in the nature of a social science which makes the question of an empirical content far more complicated than it is in the natural sciences. In the latter there is a straightforward relation between knower and known, whereas in economics, as in all the social sciences, the knower’s known includes other knowers whose known includes other knowers and so on ad infinitum. The matter may be stated differently. ‘Economics is a study of mankind in the ordinary business of life’, as Marshall said, and the ordinary business of life consists largely of a conscious effort to achieve certain ends. It therefore involves decisions on appropriate means and these must be based on knowledge or what is believed to be knowledge. Knowledge of economic action is therefore knowledge partly of practical knowledge, and this practical knowledge within knowledge carries with it peculiar difficulties which must be considered more closely.

The individual who is the subject of economic enquiry has to decide which are the most efficient means for attaining a desired end under the conditions he expects to prevail between the present and the time of his expected success. However, by most conceptions of what is economic, his actions will necessarily involve other people, so that both the expected conditions and the efficiency of the means depend on what others will do. In order to plan with precision, the economic subject has to be able (at the least) to keep his options open until he knows what others are doing. But the others are in the same position. They also cannot calculate their best course of action until everyone else has committed themself to their course of action. Because of the interrelation of knowers and known, attempts at fully informed action would require a grand pre-reconciliation of all action or lead to a waiting game in which no one would ever get started. But there is no computerized tâtonnement and no deadlock; people do act and we believe they act purposefully.

The problem of how purposeful action is possible in circumstances such as these has been part of the underlying theme of Professor Shackle’s many writings. He has referred to it as ‘the epistemic problem, the problem of how the necessary knowledge on which reason can base itself is to be gained, the problem of what to suppose that men will do when time’s sudden mockery reveals their supposed knowledge to be hollow’.11 The solution he has put forward is that there is something between a completely calculable world and a ‘cosmos in which no act places any constraint whatever upon the character of the sequel’, that there is ‘bounded uncertainty’, that ‘there are constraints as to what range of diverse things can happen’.12 His vision is of a ‘kaleidic society’ in which now one and then another rival orientation and rival interpretation gains the ascendancy and the arrival of ‘the news’ can change the whole picture.13 Nevertheless, a bounded uncertainty needs constraints, even if they are not universally agreed upon. Though Shackle does not say so himself and though it is not really in the spirit of his work, institutions viewed as consistent conduct in the midst of novel action can augment just such constraints on uncertainty.

Shackle acknowledges the influence of Hayek’s study of the role of knowledge in economic affairs. However, Hayek’s solution to the epistemic problem, characteristically directed towards an understanding of the overall order of a market economy, is somewhat different. He focuses attention upon knowledge of a different kind of fact. For the businessman planning a new venture there is ‘hardly anything that happens anywhere in the world that might not have an effect on the decision he ought to make’. But he does not have to know what everybody else is doing, for the great significance of the price system is ‘how little the individual participants need to know in order to be able to take the right action’. He merely has to watch various prices because all that is relevant to him in the doings of others is reflected in them. The market is therefore ‘a mechanism for communicating information’, but the information it conveys is not about constraints on the range of things that can happen, but rather about ‘particular circumstances of time and place’ or ‘of the fleeting moment’, that is, about events that have happened or are happening.14 The distinction refers once more to the two sides of empirical knowledge which were alluded to earlier with regard to statistics and descriptive studies and which will be considered in some detail in Chapter 2.

1.3.3 Observer perspective versus operating subject perspective

Because of the circumstance that the knower’s known includes other knowers, there is a tendency in the social sciences to treat of two kinds of knowledge, namely, that of the observing social scientist and that of the people he studies. Often the difference between them is that the observer, uninterested in the details of his subjects’ affairs, deals only with the broad categories of his subjects’ knowledge, such as (in economics) tastes, production coefficients, factors of production and so on. In the standard neo-classical equilibrium model this tendency is very marked. The equilibrium theorist is not concerned with the details of how his subjects go about their daily lives, but rather with the general features of the order to which their actions give rise. Unlike the two writers considered in section 1.3.2 (Shackle and Hayek), the equilibrium theorist is unconcerned with the details of his subjects’ affairs to the extent that he does not pay much attention to their knowledge problem, but concentrates on the order that would prevail if his subjects did not have such a problem.

Hayek pointed out long ago that there is often a confusion in economics about the concept of a datum. ‘Datum means, of course, something given, but the question which is left open, and which in the social sciences is capable of two different answers, is to whom the facts are supposed to be given … to the observing economist or to the persons whose actions he wants to explain’. If it is to the latter, the market can have an information function only if the facts are not given equally to all. This was Hayek’s interest in the matter.15 However, with regard to the knowledge of the observer and the observed, the important question is not to whom but what is given. When a problem handled by equilibrium theory is qualified by the expression ‘given tastes’, this expression signifies much less in the economist’s knowledge than it supposedly does in the knowledge of his subjects. Hayek was himself to write later (quoting Pareto) that it would be absurd to think that in the case of such entities as tastes the economist is able to ‘fill in all the blanks’, that is, to ascertain and specify tastes in detail.16 In other words, the only significance of ‘given tastes’ for the economist’s knowledge is the fact that they are given, that is, that they do not change during the course of the analysis, and that they have the properties of all tastes, namely, that they are logically consistent and obey the rule expressed as the diminishing marginal rate of substitution.

Since these elements of the economist’s knowledge, and others such as diminishing returns, are comparatively simple, they lend themselves to treatment as axioms in purely logical or axiomatic constructs and equilibrium is the non-specific solution to a mathematical optimization problem. However, the axiomatic constructs are meant to represent something empirically real, albeit in a very idealized form. Equilibrium is therefore visualized as a balance of interests, a harmonious reconciliation of the purposeful actions of many individuals. In this way the economist has been able to use the logic of choice to arrive at a rigorous analysis of a natural order. In view of the importance of the notion of natural order in the history of the subject, this is a considerable achievement.

But it can be attained only at the expense of putting a great burden on the supposed knowledge of the subjects of the enquiry. Even though the topics of false trading and of information costs have been paid some attention and such theoretical devices as price criers, tâtonnement and re-contracting have been considered, the question of how the economic subjects actually manage to bring about a harmonious reconciliation is left largely unanswered. The most usual procedure is to side-step the issue by assuming that the subjects’ knowledge is complete and that there is therefore perfect foresight. As Shackle put it, the epistemic problem has been ignored, and in Hayek’s view, the function of the price system has been pre-empted.17

The empirical part of the economist’s knowledge is therefore far more modest than that which he attributes to his subjects’ knowledge. To the economist the expression ‘given tastes’ signifies a few relatively simple propositions. To the subjects he studies it means knowing each other’s orders of preference for a vast and diverse number of goods and services, the ratios at which these goods and services are substituted for each other when a person already has a combination of them at his command and so on. In other words, the economist can only achieve his neat analysis by assuming that the subjects of his study have full information on something about which he knows next to nothing, and of which the little he does know is of questionable empirical validity.

1.4 Purview of the following chapters

The discussion in sections 1.2 and 1.3 may now be used to give more definition to the tasks set for this study in section 1.1. In the turn of phrase adopted in this chapter, one of the problems to be investigated is why equilibrium theorists appear to have little use for a specific knowledge of conduct appropriate to occasions and circumstances.

It may occur to one, in considering this question, that there seems to be very little in an equilibrium model that can properly be called conduct of any kind. The range of interest of value-theoretic, neo-classical equilibrium models is such that people are not required to execute shrewd or imaginative plans; all they have to do is produce and buy. However, this in itself implies some institutionalized conduct. People in equilibrium models are accustomed to trade and have an impeccable respect for property rights and the law of contract. There never is anybody in an equilibrium model who decides that the best way of obtaining a desired object is to hit its present owner over the head. Some institutionalized conduct is therefore presupposed when individuals are cast in the roles of producers and buyers. Furthermore, each individual calculates very precisely what he should do in his capacity as a producer and as a buyer. If one can refer to this as his appropriate conduct, one finds that such conduct is based ultimately on his own and every other individual’s preference field, the state of technical knowledge and the availability and ownership of resources, about which he as everyone else (except the observing economist) is fully informed.

It seems therefore that equilibrium theorists feel no need for any information on what people are likely to do on some occasions and in some circumstances because they assume that all of their subjects’ conduct is already fully knowable from information on certain apparently very basic factors. In the context of an equilibrium model, occasions and circumstances can come about, in any case, only by the confluence of the conduct of various individuals, since exogenous factors such as the weather, earthquakes, epidemics and the like are not unnaturally left out of consideration. There is then quite obviously no point in seeking a knowledge of the conjunction of types of conduct with types of circumstances, because all of that is already taken care of by those basic factors which the equilibrium theorist indeed does not claim to be able to ascertain specifically, but which he has identified non-specifically.

In view of the dominance of the notion of equilibrium in economic theory, the all-embracing nature or comprehensiveness of an equilibrium system built on preference fields, technical knowledge and the availability and ownership of resources will therefore stand in the way of any attempt to let institutions play a more substantial part in economic analysis. It will be argued in a later chapter that, of these four factors, it is the preference field idea to which the comprehensiveness of equilibrium theory can really be attributed. If this is so and one feels, as many have in the past, that economic theory can be criticized for neglecting institutions, one must try to show that there is something wrong with the preference field idea. Such an attempt will be made in this study.

However, its primary purpose will not be a critique of neo-classical economics, but rather the development of an alternative conceptual framework which could ultimately enable the economic theorist to use institutions as a means of empirical orientation. (The present study can of course go only a short way along this route.) The new conceptual framework will be evolved, in the course of an epistemological investigation of micro-economic theory, from the question of the acquisition of knowledge or from what Shackle calls the epistemic problem. In the process, economic concepts will be judged not only by their capacity for creating a formal logical order but also by their capacity for making known the empirically real. From this point of view, the idea of preference fields could be criticized in many ways. One could say, for instance, that it is not practically feasible that they should be known, or that they do not exist and therefore cannot be known, or simply that as theoretical devices they are heuristically unjustifiable. In the approach that will be adopted here it will be seen that these apparently incompatible objections do not in effect differ very much.

The approach is to ignore the distinction between the knowledge of the economist and the knowledge of the people studied, to deny that the economist’s capability for acquiring knowledge is in any way different from that of the person who executes economic plans. The observer and the observed will be on the same footing. The only distance between them is to view the observer as merely one potential knower among many who try to learn and to make sense of the world around them.

Then, if it is absurd to suppose that the economist can ascertain people’s preferences in detail, it is also absurd to suppose that anybody else can; and if it is possible for the person in the street to use a knowledge of institutionalized conduct as a point of orientation, it is also possible for the economist to know such institutionalized conduct and to use it as a point of orientation. In other words, it is then possible for institutions to play a part in the empirical content of economics.

1.5 Plan of the following chapters

Chapter 2 outlines and explains the essentials of the conceptual framework which is being proposed. Chapter 3 deals with the deterministic presupposition which, as will be argued later, has partly guided equilibrium theorists. It also considers certain objections to and qualifications of determinism in economics and then compares these and determinism to the conceptual framework outlined in the previous chapter. The chapter puts forward the proposition that micro-economic theory is largely a blend of axiomatic constructs and deterministic models. It then tries to isolate the deterministic elements. Chapter 4 considers the role which the notion of rational action plays in what was dealt with in the previous chapters. Chapter 5 briefly traces the development of common-sense notions of needs, tastes and preferences into the concept of an ordinal preference field, and tries to show that it is extremely unlikely that an ordinal preference field, or whatever it represents, can be both consistent over time and comprehensive (in the sense that it is behind all the choices of an individual) and that the concept, as it has evolved in the 20th century, is likely to be based on a confusion between different logical forms of empirical fact. Chapter 6 discusses the nature of an economics which does not rely on the concept of a preference field, but which provides for empirical orientation by means of institutions.

Notes

1

C.E. Ayres, ‘The Co-ordinates of Institutionalism’, Papers and Proceedings of the American Economic Association, in American Economic Review, May 1951, p 52.

2

A more usual approach is to look for common principles in the evolution of institutions. See, for example, F.A. Hayek, ‘Notes on the Evolution of Systems of Rules of Conduct’ in Studies in Philosophy, Politics and Economics, (London: Routledge, 1967) pp 66–81, in which he investigates the interplay between rules of conduct and the ‘overall order’ of an economy or a society. Some sociological theories stress the complementarity of the functions of institutions, which sometimes leads to an analogy between societies and biological organisms. This is not the approach adopted here.

3

Carl Menger elaborated this view considerably in his exposition of money. It first appeared in the chapter on money in his Grundsätze der Volkswirthschaftslehre of 1871. In his Untersuchungen über die Methode der Socialwissenschaften of 1883 he extended the argument to customs, common law, languages, towns, and so on, and argued that the logical form of the analysis is the same as that of price theory. (This idea will be explained in later chapters.) When he repeated the argument about the origin of money in Geld of 1909, he likened the difference between money and commodities to the difference between roads and other pieces of ground. See The Collected Works of Carl Menger (London: London School of Economics Reprint, 1933–6) vol 1, pp 250–60, vol 2, pp 172–83 and vol 4, pp 3–27.

4

The distinction between physical possession, which even a receiver of stolen goods gets, and legal ownership is a dominant theme in John R. Commons, Institutional Economics (Madison: University of Wisconsin Press, 1961; first published New York: Macmillan, 1934). Commons, generally recognized as having been one of the three leading figures in American Institutionalism, maintained that orthodox economics did not make the distinction and that this went a long way towards explaining why institutions played such a faint role in economic theory. He distinguished between exchange and transactions. ‘Transactions … are not the “exchange of commodities” in the physical sense of “delivery”, they are the alienation and acquisition, between individuals, of the rights of future ownership of physical things as determined by the collective working rules of society’ (p 58). In orthodox theory, he maintained, exchange and transactions were identical and this gave a double meaning to wealth, namely, ‘the physical meaning of holding the materials of nature for one’s own use in production and consumption, and the proprietary meaning … namely the right to exclude others and to withhold from them what they want but do not own’ (p 302). By neglecting the proprietary meaning, orthodox economists ‘concealed the field of institutional economics’ and it was ‘this concealed ownership side of the double meaning of Wealth that angered the heterodox economists’ among whom was Marx’ (p 55). By assuming physical possession and legal ownership to be identical, orthodox economists could ignore statute law, ethics, customs and judicial decisions when constructing ‘a theory of pure economics based solely on the physical exchange of materials and services’ (p 56).

5

This is an expression used by L.M. Lachmann in The Legacy of Max Weber (London: Heinemann, 1970) p 38. In the place quoted he does not restrict the term to institutionalized conduct. However, later in the book in an essay entitled ‘On Institutions’, he says that institutions provide ‘means of orientation’ (p 49), that the rules of a game ‘constitute a set of orientation points’ (p 61) and, quoting and translating Weber, that institutional norms are used by a certain group as ‘a means of orientation of their (legal or illegal) acts because certain expectations concerning the conduct of others attach to them’ (p 62).

6

The student was a Mr Millar who later became Professor of Law at Glasgow University and apparently a close friend of Adam Smith. The extracts appear in a long passage quoted in and probably solicited for Dugald Stewart’s Account of the Life and Writings of Adam Smith, LLD. The source here is a reprint of the short biography included in Smith’s Theory of Moral Sentiments (London: Bell, 1892) p xvii. See also Nathan Rosenberg, ‘Some Institutional Aspects of the Wealth of Nations’, Journal of Political Economy, 68, 1960, pp 557–70.

7

L. von Mises, Human Action (London: Hodge, 1949) pp 64 and 66. The term ‘pure logic of choice’ seems to have been coined by Hayek and von Mises does not use it, having himself made up the term ‘praxeology’. The two terms have more or less the same meaning except that Hayek’s has a slightly pejorative connotation, since he does not agree with von Mises’s views on the purely a priori nature of economics.

8

Ely Devons, ‘Applied Economics: The Application of What?’ in The Logic of Personal Knowledge, Essays Presented to Michael Polanyi on his Seventieth Birthday (London: Routledge, 1961) pp 155–69.

9

L. von Mises, The Ultimate Foundations of Economic Science (Princeton: van Nostrand, 1962) p 44.

10

T.W. Hutchison, ‘“Crisis” in the Seventies: The Crisis of Abstraction’ in Knowledge and Ignorance in Economics (Oxford: Blackwell, 1977) pp 62–97. The quote in the text is on p 71. The original source is the Presidential Address to the American Economic Association, 1970, in American Economic Review. Others quoted are Professors Ragnar Frisch and Harry Johnson, Lord Kaldor, Sir Henry Phelps Brown and Mr G.D.N. Worswick, Director of the National Institute for Social and Economic Research. The last-named has been especially outspoken: Those responsible for ‘some econometric theory’ are not ‘engaged in forging tools to arrange and measure actual facts so much as making a marvelous array of pretend-tools which would perform wonders if ever a set of facts should turn up in the right form’. ‘There now exist whole branches of abstract economic theory which have no links with concrete facts and are almost indistinguishable from pure mathematics.’ The only ‘distinguishing feature is that some of the axioms and some of the terminology show traces of the ancestry of this particular branch of mathematics, which originated in the distant past in some real economic question’. Then there is Sir Henry’s complaint that ‘the human propensities and reactions’ which economics ‘purports to abstract are not in fact abstracted … but are simply assumed’.

11

G.L.S. Shackle, Epistemics and Economics (Cambridge: Cambridge University Press, 1972) p 447.

12

Shackle analyses the idea of bounded uncertainty in great detail in Decision Order and Time in Human Affairs (Cambridge: Cambridge University Press, 1961). The quotations are from pp 4 and 271.

13

For something of the flavour of the kaleidic society, see Shackle, op. cit. (note 11) pp 76–9.

14

F.A. Hayek, ‘The Use of Knowledge in Society’ in Individualism and Economic Order (London: Routledge, 1949) pp 80, 81, 84 and 86. The same idea could also have been gleaned from the articles ‘Economics and Knowledge’ and ‘The Meaning of Competition’ in the same book. In all three articles Hayek of course says a great deal more than is reported in the text.

15

Hayek, ‘Economics and Knowledge’, op. cit. (note 14) p 39. His italics.

16

Hayek, ‘The Theory of Complex Phenomena’ in Studies in Philosophy, Politics and Economics, op. cit. (note 2) p 35.

17

Shackle, op. cit. (note 11) pp 221 and 447. Hayek, ‘The Meaning of Competition’, op. cit. (note 14) pp 94f.