The Foundation of Economic Method
   by Larry Boland
Geo Allen and Unwin, 1982, 2003
*  ESSAY *
The kind of criticism being pursued in this book is "imminent criticism" which involves working within the assumptions of the system to explore their consistency and the capacity of the system to succeed on its own terms. "External" or "transcendent" criticism takes issue with the assumptions of the system and it tends to have little impact on people who for one reason or another feel constrained to work within the system.

Boland describes how the received views on methodology in economics are Conventionalism and Instrumentalism. He addresses the way that these assumptions and related philosophical issues permeate the way that economists actually go about their tasks (mostly model building these days) even though they are only addressed on rare occasions in the most perfunctory manner. If the economists are taken to task for this cavalier treatment of philosophical issues they have a devastating rejoinder: "show me something in Hands, Blaug or any of the other thousands of references that they cite which will actually help me or make a useful difference to what I am doing".  The most useful answer of course is the work of Boland.

He is concerned with the neoclassical research program (NRP) and he is especially concerned to bring out in the open some stated and unstated items on the research agenda. The core of his argument is that the NRP, in all its forms and variants, is designed (1) to be consistent with acceptable ways of dealing with the problem of induction, and (2) to provide a methodological individualist explanation of economic behavior of the economy. He notes that this program is impossible (due to the insolubility of the problem of induction), also it tends to lapse into psychologism.

The Introduction is titled "Understanding the Methodology of Economics". LB draws an uncommon and somewhat subtle distinction, between two perspectives, one being the views on "scientific" explanations held by economists and the other being the economists' perceptions of the methodology of the economic "actors" and decision-makers as they go about their business in the marketplace. It seems that both perspectives tend to be similar despite the general lack of discussion of these things, and they are both based on an inadequate theory of knowledge.

Because there is so little discussion of methodology in the main dwelling of economics, detective work is required locate the methodological ideas that are at work. LB was determined to identify the actual practice in the profession, as distinct from the incantations that are offered in the early part of textbooks. 

Methodololgy as Agenda

"The analysis of neoclassical methodology presented here will focus on the research agenda of every neoclassical theory, analysis, article, etc...Specifically, a research agenda is an ordered list of theoretical or philosophical problems that either are to be solved by the research conducted or are problems whose solutions play a necessary role in the solutions of other problems to be considered" (6-7).

He notes that the paradigms and research programs of Kuhn and Lakatos are mostly concerned with explicit agenda items but he wants to get at the implicit items on the agenda.

He instances modern demand theory as an example of a typical neoclassical agenda. The aim is to identify a set of assumptions which will yield a traditional set of results. "Stating the purpose this way immediately begs two questions. What is the traditional set of results? And what assumptions are admissable into the set?" (8)

Samuelson set the requirement that any given theory of demand should yield his 'fundamental assumption of consumption theory', namely the demand curve for any normal good should slope downwards. Various problems flow from there. "These problems form the visible agenda of neoclassical demand theory; and its specification is the task of a broader methodological agenda, which is usually hidden because it is taken for granted."  The broader agenda is concerned with questions which probe the very rationale of the assumptions that underpin the stated agenda, for example "What problem is solved by establishing that demand  curves are usually downward sloping?"

This book argues that the hidden agenda of the field consists of two related but autonomous problems; first, the Problem of Induction; second, the Explanatory Problem of  Individualism. These are explored in chapters 1 and 2. Their influence on research programs in neoclassical economics is pursued in chapters 3 to 6. Their influence on the practice of methodology in the field is presented in chapters 7 to 9. Alternative approaches are examined
in chapters 10 to 12.

Chapter 1: The Problem of Induction vs the Problem with Induction

"Even if methodologists today avoid promoting the hierarchical distinctions of Inductivism [hypotheses, theories, laws] the dominant methodological perspective is that the fundamental problem facing all economists is one of choosing the one 'best' theory or model. It is this choice problem which is the primary remnant of Inductivism" (26)

Regarding the insoluble problem of induction, the author notes "Nevertheless, what it is and how it is either 'solved' or circumvented is fundamental to understanding all contemporary methodological discussion". (13)

He sets out to provide a clear statement of the problem, starting with its elementary parts, one of  which is justificationism: "the methodological doctrine that asserts that nobody can claim to possess knowledge unless he or she can also demonstrate (with a proof) that his or her knowledge is true; this is, everyone must justify his or her knowledge claims".

Much more could be said about the implications and ramifications of justificationism, and the barriers that it creates for understanding Popper's philosophy, but this is not the place for it. The author proceeded to The Problem of Induction in Economics where the main aim of methodology has been to skirt the problem, usually by shifting to The Problem of Conventionalism: "the problem of finding generally accepted criteria upon which to base any contingent, deductive proof of any claim to empirical 'knowledge'" (18).

In practice, the generally accepted criteria have evolved into a form of normal science where the puzzles are concerned either with econometric models or mathematical models. In the first instance, the requirements of  science are met by using data (with some talk about falsification, confirmation and the like). In the second instance the criteria are along the lines of simplicity, economy, elegance and other considerations of mathematics.

Chapter 2 The Explanatory Problem of Individualism

"Methodological individualism is the view that allows only individuals to be the decision-makers in any explanation of social phenomena" (28).

This is a methodological view and it is not to be confused with various other forms of individualism such as the ideology that asserts the rights of the individual against the social collective, or the ontological view that individuals are in some way more real than collectives, or the epistemological view that the truth is warranted by individual (subjective) perceptions (either of  sense data or intuitions).

The usual form that MI takes in the neoclassical research program is psychologism, "the methodological prescription that psychological states are the only exogenous [unexplained] variables permitted beyond natural givens (weather, contents of the universe, etc)" (30)

Or more precisely, psychologistic individualism, "the version of psychologism which identifies the individual with his or her psychological state" (30)

Institutions (and social aggregates of all kinds) pose a challenge to individualism, especially to psychologistic individualism which is a form of reductive individualism. Can institutions be treated as "givens" or "exogenous" factors which can be left out (unexplained) in our models?

"In a fundamental way, specification of the exogenous ["rock-bottom" or unexplained] variables is probably the most informative theoretical assertion in any theoretical model. The various competing schools of economics might easily be characterised on [this] basis...Marxian models take 'class interest' and 'rates of accumulation' as exogenous givens." (31-32).

The bottom line of LB's argument here is "that psychologism is retained because it is a part of the Conventionalist program to deal with the problem of induction" (34).

Moving on to the topic of Individualism as an Explanatory Problem, LB notes

"The reason why psychologism survives is that it is supported by the common presumption that rationality is a psychological process...based on a belief that Hume was able to overcome the Problem of Induction. It is also supported by the view that rational decision-making must in some way involve inductive rationality" (37.

These views are underpinned by the deeper assumption of justificationism, which LB mentioned early in the book.  He notes that relatively sophisticated commentators such as Shackle and Simon persist with the psychologistic view of rationality. Against this, LB offers Popperian critical rationalism, where rationality applies to arguments rather than people or thought processes. In the light of these considerations, LB writes "Thus, any discussion of rational decision-making need not involve psychology. So we ask again, why is psychologism still commonly accepted?" (39).

My suggestion is that this is accepted, because it is underpinned by "deep" assumptions (unstated agenda items) like justificationism and subjectivism, and it makes people's heads hurt when they are invited to consider the alternatives to them. This is why LB suggests that no single person could hope to change the hidden agenda in a single lifetime (41).

Under the sub-heading "Individualism as an Agenda Item" the author provides an intriguing chain of arguments to demonstrate that individualism remains on the agenda because it fits into a "long-run inductive research program...and without Inductivism the individualist view may seem rather weak." (39-40)

LB hints that there are other reasons apart from inductivism to sustain some forms of individualism but until these are spelled out and taken on board by the profession "neoclassical economics is based on reductive methodological individualism by default" (41).  He  also points out that psychologism also fits nicely with Conventionalism (which is also based on Inductivism) so the various elements on the agenda support each other.

Chapter 3 "Psychologism vs Disequilibrium Models"

This chapter begins with a summary of the argument this far: the neoclassical research programs (NRPs) are based on a hidden agenda with two items: first the problem of induction which is usually deflected to the problem of conventions; second, a reductive version of methodological individualism which identifies individuals with their psychological states, especially their given utility functions.

"Our purpose is to establish a clear understanding of what neoclassical economics is rather than to determine what some philosophers think it should be" (48).

This can be done without accepting the hidden agenda items in the way that justificationists expect us to choose or commit to theories.

"When we say that we take the hidden agenda items as givens it does not mean that we are thereby accepting them. On the contrary, we shall be arguing that, paradoxically, it is the acceptance of the two items which gives rise both to the many theoretical problems that avant-garde economists find fascinating and to the obstacles which block their solution" (49).

The Price System and Psychological Individualism

A serious limitation is imposed by the requirement that "no social institution that appears in our explanations must be allowed to play the role of an exogenous given". This is called the problem of social institutions: "How do we assure that every institution which is introduced as a given in the short-run (or partial equilibrium) version of a model can be explained in terms that include only the exogenous variables permitted in the long-run (or general equilibrium) version of the model?" (49-50)

The price system is considered as a social institution. It seems that Arrow raised some technical considerations about the difficulty  of creating a situation where all individuals simultaneously set the price but no single one can set a price and deny the influence of all the others. Two schools of thought have emerged to treat the hypothetical situation where the market is not in equilibrium despite the theoretical requirement that all the bidders in the market are supposed to be offering the same bids and everyone is maximising their utility. (1) we need a theory of disequilibrium trading. (2) we need a neoclassical theory of the individual bidder.  The author reports that the problem still stands.

Chapter 4 "Rational Expectations and Theories of Knowledge"

"What is really surprising about rational expectations models is that they employ a 500 year-old theory of knowledge and at the same time ignore the 200 year-old refutation of that theory!" (67).

I will completely skip this chapter because Critical Cafe dwellers (the original audience for this material) should not need to be reminded about the defects of inductivism.

Chapter 5 "On the 'Necessity' of Microfoundations for Macroeconomics"

"The demonstration of the existence of microfoundations for macroeconomics is considered essential by many leading economists... However - and this is not widely pointed out - this 'necessity' presumes that microeconomic theory, in the form of general equilibrium theory, is a successful individualist program. In some quarters, as we explained in the previous chapters, this is still an open question" (80).

An external critic might say, why does the foundation have to be general equilibrium theory? But that would mean quitting the game of mainstream economics and joining some other league, like the Marxians or the Austrians.

General Equilibrium vs Macrotheory

The author makes the point (elaborated in previous chapters) that a market in equilibrium is a market where all participants are maximising.

General equilibrium

Under this heading the author argues that the Hicks strategy of identifying the individual's maximisation with the individual's equilibrium "has in effect built in a presumption of general equilibrium in order to explain the behavior of any individual. We argue that this is a major source of the difficulties that have led to confusions concerning the differences between macroeconomics and general equilibrium analysis" (82).

Macroeconomics: Keynes' 'departure'

In a 1937 article Keynes claimed that his general theory was a departure from tradition in at least two ways. One was his assumption of radical uncertainty concerning the future, not just the kind of "betting uncertainty" where meaningful odds can be assigned to a finite "field" of alternatives. The other departure was to address the long-neglected matter of macrotheory, as he claimed  "the theory of effective demand, that is the demand for output as a whole, having been entirely neglected for more than a hundred years" (quoted on page 83).

People who think that Keynes rendered neoclassical economics irrelevant will presumably be playing in the post-Keynesians league and will not read this book. Those who are in the neoclassical game will assume that the Keynesian system still needs microfoundations, whether or not Keynes accepted the neoclassical hidden agenda (which he may well have done, after all he wanted to solve the problem of induction by way of probability before he turned to economics).

Aggregative economics and microfoundations

From an individualist point of view, macro factors are simply the sum of a lot of individual things, in the way that birth, death and suicide rates are the unintended result of many of individual events. Macroeconomists tend to holism, as though the macro factors have a kind of existence of their own. This is my gloss on the situation. Under this subhead the author noted some of the problems of keeping the vision of macro factors in focus.

The Problem of Microfoundations

"The key question underlying the dispute over microfoundations is: Are there any limitations to the success of the neoclassical microtheory in terms of methodological individualism? For example, does the individual decision-maker require perfect knowledge?" (85). Following this line of thought leads to two problems; one is the nature of general equilibrium (examined in this chapter), the other is the need for stable expectations or stable functional relationships (see the next chapter).

Chapter 6 is "Time and Economic Theory"

"What we should be asking is not whether neoclassical economics is timeless but whether its treatment of time is adequate...[it may be argued that] an adequate 'dynamic' model must include at least one dynamic process. But we will eventually have to ask: can such a model ever be consistent with the hidden agenda?" (96)

The outcome of some fairly heavy arguments is that "although neoclassical models are not strictly timeless, they are still incapable of rendering explanations of dynamic processes" (98).

At the end of the chapter there is some discussion of a book by Hicks on the methodology of macroeconomics. He was most interested in the use of macro for the analysis of facts. He also insisted on the need for causal explanations. "This puts the methodological questions of the adequacy of macroeconomic theory at centre stage, in the spot-light. He does not go far enough" (110).

This brings the author back to consider the two "levels" of methodology - (1) the views that economists have about their own methods and (2) the views that they have about the methods that are used by actors in the marketplace. So he turns to consider the economists' views of methodology.

Part III Conventionalist Methodology in Economics

Chapter 7 Positive Economics as Optimistic Conventionalism

Here the author reverts to the neoclassical mainstream, referring to the previous chapters as a survey of various 'avant-garde' research programs. The two main currents in the mainstream are "positive economics" (recently called 'applied economics), and "economic theory" (which was a couple of decades ago called 'mathematical economics'). He suggests that they correspond to optimistic (naive) Conventionalism and pessimistic (sophisticated) Conventionalism (with a reference to Agassi).

Positive Evidence about Positive Economics

This section notes the standard format for papers in positive economics. After the introduction there is a section on 'the model', then 'empirical results' and then 'conclusion'. "...this format satisfies the dictates of Conventionalism" (116).

The author conducted a survey of the literature to verify that the standard format was indeed standard and he then turned to the theoretical implications of this discovery.

"To understand the relationship between the standard format and the research program to verify neoclassical theory, we need to consider the following questions. What constitutes a successful empirical analysis? What would be a failure? What is presumed in the use of 'testing conventions'?" (119).

It seems that the primary requirement of papers is the provision of some new facts.

The Logic of Model-building in Positive Economics

The author suggests that each applied model is a implicit test of neoclassical theory by applying it to the real world. To abbreviate a somewhat complicated story, LB concludes that normal model building and testing does not really constitute a test of neoclassical theory at all.  "The point of formalizing our view of models is to show that building models of a theory in effect insulates the theory from empirical testing" (121).

Chapter 8 Analytical Theory as Defeatist Conventionalism

This chapter is about the mathematical strand of modern microeconomics. Where positive economists want a foundation of true empirical facts, these theorists want a foundation of mathematical (logical) truths. 

"The paraphernalia of the pursuit of logical truths include the following 'buzz-words': proposition, theorem, lemma, proof, corollary, hypothesis, condition and definition. These words play a prominent role in the format of recent theory articles. Usually they are printed in capital letters to highlight the format" (131).

The result of most of the effort in this direction is trivial mathematical puzzle-solving in the tradition of Kuhn's normal science.  Paul Samuelson was a major influence in this branch of the subject (he is the man who up to the fall of the wall thought that the Soviet economy was growing fast and might catch up with the US in a decade or two).  Interestingly, he was a major influence in a pseudo-Popperian direction (at least in his rhetoric) with his aim of deriving "operationally meaningful theorems" from economic theory, meaning hypotheses which could conceivably be refuted.  In a separate paper LB noted that the major US journal in economics in 1950 contained one equation. Nowadays the overwhelming majority of papers are mathematical.

Chapter 9 Instrumentalism as a Rejection of Conventionalism

This chapter examines the rather different line pursued by Milton Friedman in his 1953 article expounding the unimportance of the truth or falsity of assumptions and the all-important (instrumental) function of theory for making correct predictions. This is sometimes called Pragmatism but Friedman is an Instrumentalist, not a Pragmatist or a Conventionalist.

The author notes that both Conventionalism and Pragmatism arise from the failure to solve the Problem of Induction. "The former deals with the problem by denying its original objective, which was to establish the truth of scientific theories. The latter deals with the problem by accepting a weak criterion of truth, namely 'usefulness'" (143). The potential for depicting Friedman as a pragmatist is obvious but this is not the correct label, as the author explains.

As an aside, regarding the truth, there has been a lot of critical commentary on the pragmatic criterion of usefulness, and the "dismissal" of correspondence. Bertrand Russell was particularly unhappy with the pragmatists. However the matter of truth as correspondence was not really evaded by the pragmatists, merely shifted from the immediate theory under consideration to the question of whether or not it is actually true to say that it works (or works better than alternatives).

Getting back to Friedman, Boland was involved in a somewhat farcical situation when he wrote a paper to explain that Friedman had adopted the Instrumentalist position and his many critics had yet to lay a glove on him. (Part of the farce was the travails that were required to get the paper accepted by a journal, but that is a story that LB has narrated elsewhere. On this topic let me issue a reminder that some of his books and numerous papers (including the essay on Friedman) can be obtained in PDF format from his site.

LB is especially scathing of Samuelson's very popular critique of Friedman.

The author ends this chapter (and Part III of the book) with the conclusion that neither Instrumentalism nor Conventionalism will do if we are searching for "a more universal, lasting understanding of the workings of the economy". Instrumentalism ignores the truth of theories and Conventionalism denies any truth status to theories. "If a true theory of the economy is our objective, then we will just have to look beyond the dispute over methodology between Friedman's Instrumentalism and the Conventionalism of Samuelson or Lucas". (152).


Chapter 10 Contemporary Methodology vs Popper's Philosophy   of Science

"So far we have examined the effects of the hidden agenda on neoclassical theoretical problems and research programs; now we wish to examine its effect on the neoclassical view of methodology. Generally speaking, methodology is rarely discussed in the leading journals" (155).

Methodology and the Hidden Agenda

Apparently methodology has traditionally been concerned with the items on the hidden agenda (suggesting that it is not that well hidden) and for this reason it "is paradoxically considered either a waste of time or too dangerous to handle". Students are warned off the subject and it is mostly relegated to the side-alley of the history of ideas (an alley that became a dead end under the influence of  people who followed Lakatos).

The author asks "Does this orthodox attitude towards methodology merely represent a deep-seated insecurity about the adequacy of the hidden agenda?" (156). One is tempted to answer "yes" but how many economists think enough about methodology to even get as far as insecurity regarding the hidden agenda?

Skipping along though Methodology and the History of Economic Thought, we encounter the achievement of Samuelson and Wong (maybe the wrong guy ended up with the Nobel Prize, anyway Stanley is clearly smarter than the author of this book because he got out of academia in time to make a useful career in law).

We find a heroic statement by Blaug on progress in economic theory. We move through Conventionalism and the Sequence of Models (Kuhn's normal science in action) and we come to ...
Misappropriation of Popper's View of Science

Popper's views came to economists in the unhelpful form of "falsificationism" and various forms of his ideas are attributed to both Friedman and Samuelson, who are probably the most influential "methodologists" in the profession, each catering for different followings. However any celebrations about the Popperian triumph would be premature, as the author points out "So far, Popper's only real accomplishment in economics is the suppression of any open advocacy of Inductivism" (165) but as LB has demonstrated in this book, Conventionalism and Instrumentalism reign supreme in economics and Popper rejected both. LB notes that the main reason for misinterpretation of Popper among economists has been the influence of Lakatos, in his attempts to build a bridge between Kuhn and Popper " a great extent he has succeeded. But the cost of the reconciliation has been the abandonment of most of the more important aspects of Popper's philosophy of science" (165). 

The Foundations of Popper's Methodology

"There are two essential and related considerations without which no clear appreciation of Popper's views can be reached. One is Popper's view of Plato's 'Socrates', the other is the observation that Popper has strong ties to what is usually called the Austrian School of economics" (166).  It looks like I should get out more often, specifically, I should have read this book a couple of decades ago, or my economist friend Colin Simkin should have told me about it. It is very helpful to grasap the affinity between Popper and the Austrian school (at the deep structural level I mean, most commentators emphasise the differences between Popper and Hayek, not the similarities). Before Hacohen's book, people would have thought that Popper got the Austrian line on the social sciences from Hayek or (earlier, in Vienna) from Weber (in the original German), Weber being probably the most effective expositor of "Austrian" methodology outside the hard core Austrians who had little influence on the development of Western social science. Hacohen discovered that Popper only found out where Weber was coming from when he read him much later in English translation.

Under the sub-heading "Popper's anti-Justificationism", LB noted that Popper rejected the problem of induction, while unfortunately claiming that he had solved the problem. Sometimes, to clarify the value of Popper's rejection (or solution) I suggest that the problem of induction is really the problem of the empirical base, like how do we use evidence to seek for the truth, develop better theories, promote the growth of knowledge etc. For that reason I am not overly interested in going back to Hume to find how the problem of induction is or should be formulated. It seems to me that getting a grip on the notions of non-justificationism and objective knowledge is more important than jousting with philosophers about confirmation theory or inductive probabilities.

Larry Boland has got hold of both of these notions and they are effectively his own "hidden agenda" though he does not elaborate upon them at length, bearing in mind the readership that he would have been seeking among his peers in economics.

False Problems Raised by Popper

These are the demarcation problem, degrees of corroboration and the growth of knowledge.

The demarcation problem. 'If we do not get involved with the older Logical Positivist views of methodology, then the so-called Demarcation Problem is at best uninteresting" (170). Tell that to Popper!  Bartley tried, without success. That is why it is sometimes helpful to explain Popper's views on falsification in terms the most (actually the only) logically decisive way of using evidence, and also the most cost-effective way (an argument to appeal to laboratory managers and grants administrators, also journal editors). See Crick's comment 'It's getting rid of false ideas which is the most important thing in developing the good ones...You should not get bogged down with experimental details. You should make some sort of bold assumptions, and try them out' (interviewed by Wolpert for "A Passion for Science", OUP 1988.

Degrees of corroboration. This turned out to be a dead end and it is described by LB as a source of intellectual fog. Like the ill-fated formal measure of verisimilitude, this was an attempt by Popper to play the positivists at their own game instead of articulating the non-justificationism which is so important as an element of the hidden agenda.

Friedman and Popper

Interesting brief discussion of some confusions about Instrumentalism and Conventionalism.

Conventionalist Pseudo-Popper

"Blaug identifies Popper's philosophy of science as the 'watershed between old and new views of the philosophy of science' [1980, p 2]. The new view, according to Blaug, is the Conventionalism of Kuhn's or Lakatos' compromised version of Popper's view" (173).

Hence the debacle of 'post-watershed' methodology in the history of ideas in economics.

Chapter 11 Putting Popper on the Agenda

This chapter begins with the comment that the attempt to get around the problem of induction results in the adoption of Conventionalism but "there is no reason for our having to choose one theory rather than another other than short-term practical considerations" (174).  This view has very subversive implications for most of the dialogue that one encounters in the philosophy of science. It probably needs more discussion. It is fairly obviously correct when you do a situational analysis of the working scientist. He or she is concerned with theoretical and experimental problems (of course the two are linked). He or she might hope to come up with better theories (in pure research) or better solutions to engineering problems (in applied science and technology) and more effective tests or experiments (in both). These activities do not call for theory choice in any ultimate sense, beyond the practical or contingent sense of  deciding what we are going to work on this week or before lunch.

"The only methodological perspective, other than the impossible Inductivism, which is directly concerned with the realism or unrealism of economic models is Popper's so-called critical rationalism...Let us explore what it would mean for an inclusion of Popper's methodology in the hidden agenda of neoclassical economics and its possible role in neoclassical theory" (174-5).

Adjusting the Neoclassical Hidden Agenda

First eliminate the problem of induction. No need to choose between competing theories. Understand the problem and the alternative solutions that have been proposed "...look for more and more alternative solutions". (175)

Moving along to the second item on the hidden agenda, methodological individualism, the move here is from psychologism to consider the theories that are held by decision-makers. The essential requirement here is a theory of objective knowledge which enables us to discuss the "subjective" theories that are held by decision-makers in an "objective" manner. For example it makes a difference whether an economic actor is using a maximising strategy, a minimum risk strategy or a satisficing strategy, and that difference can be explored without any immediate need to explain why the particular strategy has been adopted, unless of course the focus of investigation shifts to that particular question, for example if the actor is invited to reconsider their strategy.

"What we will argue below is that a major ingredient in every decision is the theories held to be true by the decision-maker and that in the absence of an inductive logic such theories cannot be reduced to the given nature of the physical world. Why any individual may consider a particular theory to be true may or may not be at issue. It all depends on the problems that the individual is trying to solve." (176)

Real-time Individualism in the Short Run

"Discussing arbitrary changes in the research agenda of neoclassical economics is really not very interesting unless we can see how the new agenda affects the nature of any neoclassical theory...We wish to show here that by dropping Inductivism and Conventionalism and instead relying on Popper's views of knowledge and learning, the way is open to the development of real-time explanations in neoclassical theory" (176).

The author suggests four agenda items for an individualistic explanation of dynamic processes. [bear in mind that this was published in 1982, which means that it was  written long before volume three of Popper's Postscript appeared with the theory of metaphysical research programs described in it]. Even without that hint, LB labels this four-point agenda "the Popper-Hayek program". 178

Anti-psychologism [note that Popper not only expelled psychologism from epistemology, he also expelled it from human psychology].
Rational decision-making (according to the logic of the situation, with reference to Hayek 1937/48 and Hicks)
Situational dynamics. Behavior can change as a result of learning as well as from changes in the situation.

This approach modifies Hayek's account by replacing his inductivism with Popper's concept of objective knowledge. [Jack Birner has a paper on the evolutionary epistemology of Hayek and Popper, in which he comes down on the side of Popper and suggests there is actually a Marxist strain in Hayek's thinking, with a strong inductivist element in his major book on psychology].

LB notes that Hayek's point about the importance of learning has been taken but has been blunted by persisting elements of inductivism, as though we just need more information (of a different kind) to get the kind of rationality that the traditional agenda demands. In contrast:

"We shall argue that the status of an actor's knowledge may give a reason for change, but it does not tell us what the change will be. [should non-determinism be a fifth item on the agenda?] However, knowing the actor's learning methodology may provide a clue to what change he or she may attempt to effect" (181). This is a major point which probably should be printed in bold italics. The author has elaborated on it with a couple of pages of argument which I will not attempt to summarise.

A Neoclassical Program for a Real-time Short Run

The aim is to explain decision-making by economic actors in real time (as opposed to the instantaneous processes that achieve equilibrium in a large set of simultaneous equations). The program is based on a dynamic concept of knowledge, which is endogenous (part of the system that has to be explained). It involves learning, and learning takes time (and is irreversible) so the flux of time has real meaning in this system. So is the notion of error, and learning from mistakes. The end of this chapter refers back to previous chapters which described the problems with existing neoclassical models, in order to indicate how this approach is better.

This program has to cater for disequilibrium in a way that the mainstream models cannot. "However, this state can be completely explained if we explain how the actors respond to knowledge incompatibility" (and to unintended consequences arising from error). (187).

Chapter 12 Problem-dependent Methodology

"Our profession's reliance on Conventionalism to deal with the Problem of Induction has always put a high value on agreement, that is, on having our views accepted by our colleagues" (188) This of course reinforces all the other tendencies that promote Kuhnian normal science.

Hence, to make progress we are supposed to use the "one correct method" for analysing the economy. The author's task in this chapter is to challenge the traditional "one true method" school of thought. It is misleading on two counts:
1. it presumes the "one correct method for all of science.
2. it reflects the justificationist assumption that "authoritative support" is required for scientific explanations.

The Traditional View of Methods

"If you want to play an immediate role in the development of modern economics, you must learn how to use the currently accepted method. Paradoxically, even attempts to change the accepted method must proceed according to the currently accepted method". (190).  In other words, you have to pull yourself up by your bootstraps even though the remainder of the profession is standing on them.

Authoritarianism and the Hidden Agenda of Science

In addition to the problem of credentialism, the author flags the danger of inadvertently advocating authoritarianism. He reminds us that his aim is merely to call attention to the role of the hidden agenda.

Methodological Agreement

A certain amount of agreement is required to enable the profession (and the universities) to function. However there is a danger of forced or coercive agreement, which blocks the non-hostile and cooperative critique of major issues including first principles and the items on the hidden agenda.

The False Choice Problem

"The primary source of disputes over criteria such as simplicity, generality, or falsifiability is the Conventionalist's choice problem itself. It is a false problem". (192)

The field has polarised between Instrumentalists and Conventionalists but they argue at cross-purposes because they tend to have different problems and aims.

The Fundamental Choice Problem

In light of the above, it helps to put the objectives first (put more work into defining the problem, as Tyrell  Burgess suggested). Then use the method that is appropriate to the problem.