Keynes on microfoundations
from Lars Syll
The atomic hypothesis which has worked so splendidly in Physics breaks down in Psychics. We are faced at every turn with the problems of Organic Unity, of Discreteness, of Discontinuity – the whole is not equal to the sum of the parts, comparisons of quantity fails us, small changes produce large effects, the assumptions of a uniform and homogeneous continuum are not satisfied. Thus the results of Mathematical Psychics turn out to be derivative, not fundamental, indexes, not measurements, first approximations at the best; and fallible indexes, dubious approximations at that, with much doubt added as to what, if anything, they are indexes or approximations of.
Where ‘New Keynesian’ and New Classical economists think that they can rigorously deduce the aggregate effects of (representative) actors with their reductionist microfoundational methodology, they have to put a blind eye on the emergent properties that characterize all open social and economic systems. The interaction between animal spirits, trust, confidence, institutions, etc., cannot be deduced or reduced to a question answerable on the individual level. Macroeconomic structures and phenomena have to be analyzed also on their own terms.
Leave a reply to davetaylor1 Cancel reply
—– look inside —– $5.94 / $20.00
—– look inside —– $4.90 / $8.00
—– look inside —– $15.99
—– look inside —– $5.99 / 12.99
—– look inside —– $5.93 / $12.99
—– look inside —– $4.97 / $9.90
—— Ugarteche, Puyana and Madi ——
Gerson Lima / Maria Alejandra Madi
Edward Fullbrook and Jamie Morgan
————— Michael Hudson ————–
Maria Alejandra Madi / Jack Reardon
————- Edward Fullbrook ————-
—————— Steve Keen —————–
————— Richard Smith —————
————– Gustavo Marques————
– Victor Beker and Beniamino Moro –
————– Lars Pålsson Syll ————-
—————– Stuart Birks —————-
Edward Fullbrook and Jamie Morgan
———— Armando Ochangco ———-
Shimshon Bichler / Jonathan Nitzan
————— Mauro Gallegati ————–
————— Herman Daly —————-
————— Asad Zaman —————
—————– C. T. Kurien —————
————— Robert Locke —————-
Real-World Economics Review Blog 
A comment on Edgeworth’s book (1881) written in 1926 does not apply to the state of the art of heterodox economics.
Please read
Fredric S. Lee 2018 Microeconomic Theory / A Heterodox Approach, Routledge.
The late prime promoter of the term heterodox economics pleads the need of microeconomic theory for heterodox economics including Post Keynesian economics.
See also
Marc Lavoie’s book review on our book that appeared in the volume one of a new heterodox journal Review of Evolutionary Political Economy
https://link.springer.com/article/10.1007/s43253-020-00004-5
I think we all agree that the orthodox conception of micro-foundations is misplaced, to put it politely. But this doesn’t answer the question of micro-foundations more generally. One response is to say, as Lars Syll appears to be suggesting, that only a macro-level analysis is valid – although his use of the word “also” at the end suggests a welcome lack of dogmatism on this question.
This is a widespread problem with heterodox economists, even the ones I most admire: the insistence on the macro level being *the* primary causal impulse. This is the wrong dichotomy! To deal with emergent properties and causal interactions, the lower-level causal processes must be considered, along with any causation that may also occur at the higher level.
Sometimes, a great deal can be learnt with quite simple tools – notably, with just the concept of feedback (reinforcing/positive as well as balancing/negative), one can explain a large proportion of the observed regularities in the economy, including stability in most product markets, booms and busts, capitalist growth, etc. An implication is that the behavioural input to economics needs to focus on what keeps these feedback systems going, rather than exogenous preferences etc which function as inputs to an automatic (because optimizing) economic system. I have a chapter coming out soon on this, in a book on Feedback Economics published by Springer and coordinated by the System Dynamics Society.
I could enlarge on this to consider more “complex” systems than feedback, e.g. Mandelbrot’s work on fractals; and with the way that this systems perspective connects with the evolving nature of the economy. But I will leave those aside for now.
Finally, I was happy to see Marc Lavoie’s review of the book by Yoshinori Shiozawa and colleagues. I think that their work demonstrates how the concept of micro-foundations is essential to analyzing the economy, and that it can be done well! Which doesn’t mean that there is necessarily no causation at macro level – the idea that multiple types of causation occur in the economy (as in biology and much else!) is fundamental.
I should also have said that with a feedback analysis, the emergent properties at the level of the whole system are explained in terms of the interaction of the lower-level causal links. It’s obvious, but there is so much confusion around words like “holistic” and “complexity”, that it may need to be spelt our here as the relationship between levels.
Thank you for your kind words, evidencebas. I stand in and for heterodox economics, but it requires self-criticism. It must become science more based on evidence and logic (including mathematics) and less based on methodology. What is needed for it is to stop methodological bavardage and set to work in sincere empirical research and theory building. Methodology does not replace theory.
So ‘evidencebased’ and Yoshinara are both so evidence-based that they are not prepared to consider either the possibility microeconomics should be based on macroeconomics or the evidence that microeconomics has failed the majority of us and is in any case based on unstable evidence like power relationships and auction prices that can be talked up or down. (Cf. Fullbrook’s proposal that unit totals can be reliably sub-divided between us).
So the evidence from Australia this morning is that the Queen’s representative didn’t after all have the power to dismiss the elected government and install neo-liberal American stooges. On 13th July Norbert Haring’s observations on the Rockefeller Foundation’s “lockstep” ambitions left Iconoclast being “simple and brutally blunt” about the likely human outcome: chaos resulting in at best a drastic reduction in the human economy.
I’m being deliberately provocative, because that is what I feel about Yoshinori and ‘evidencebased’ ignoring what I have repeatedly shared with them about the difference between our reacting (or not) to information, and “the interaction of the lower level causal links”. We humans are consciously able to sense differences as well as forces. Far from “Mandelbrot’s work on fractals” being more “complex” than feedback systems, as I have again repeatedly explained, the chaos is entirely restateable and evident when one considers the problem-avoidance/ trying-to-take-advantage D logic of the PID feedback method of equilibration (staying on course) used in navigating ships. Only advantage-seeking Rockefellers and their naive admirers believe otherwise. Fractals are beautiful. Understanding how to avoid chaos caused by taking short-cuts and not correcting deviations is useful, for most of us necessary, and in effect what Keynes intuited.
Thank you, Dave, for being deliberately provocative. Have you read Chapter 1 (§1.5.3 Micro-Macro Loops and a New Methodology in particular) of our book
Microfounations of Evolutionray Economics?
https://www.researchgate.net/publication/334508762_Microfoundations_of_Evolutionary_Economics
(You can read the draft of Chapter 1 that has the same title as the book itself:
https://www.researchgate.net/publication/301766363_Microfoundations_of_Evolutionary_Economics )
Please remark the title of the subsection. The existence of micro-macro loops denies that either microeconomics or macroeconomics is itself complete.
Researchgate having made its conditions of casual entry unacceptable, I can only point out the disagreement: Yoshinori feeling the need to deny a completeness which I never claimed, believing as I do that all a theory can do is point us more or less in the direction we will need to look to see what is going on. Mass producers for urban economies – hoping to even out and maximise production – prefer hard facts; but seasons change so better to go with the flow.
I infer Yoshinori’s empirical method posits a theory, then looks for evidence relevant to it, whereas I as an intuitive look at the evidence to see what is relevant to the problem in hand, then look back through the evidence for an explanation of what I am seeing. The key point here is that traditional verbal/symbolic logic works by the terms referring to causes and effects, not allowing for the freedom we humans have not to act on suggestive information, and indeed not escaping from the fallacy of composition when combining interacting causes.
When I argue from geometric first principles for information science and dynamic PID macro logic, I am seeing cybernetics in terms neither of Ashby’s “requisite variety” nor Weiner’s biological “interacting causes” nor even our gazing at a distant “fixed” star, but as what I’m now finding is being referred to as “Second Order Cybernetics”. I once characterised this in the three-stage process of dialling up another person so one can (fourthly) have a telephone conversation with him. The theory is about the telephone, not what they are going to talk about (though decoding what foreigners are saying may improve from information science).
Reading Karl Muller’s “De Profundis” on the late Ranulf Glanville and his second order cybernetics, it looks as if the “requisite variety” comes into the picture via intuitive minds. Most of us – in practice fortunately – are creatures of habit.
Dave Taylor thought that I am “not prepared to consider … the possibility microeconomics should be based on macroeoconomics.” This is the position often expressed by many Post Keynesians. So sincere economists who argue for microfoundations surely have reflected on this objection and have one or other replies.
My answer is the existence of micro-macro loop. This is not a claim that there are bi-directional links or causal relations between micro and macro. The important point is that micro to macro link is everyday causation. A state of macroeconomic is generated by various interactions between micro agents (including firms). I argued how and why the agent’s actions take the form of routine behaviors in Chap. 1 of our book (Microfoundations of Evolutionary Economics). Chap. 2 (prices) and 4 (quantities) explain how a large network as big as global economy normally works as interactions of those routine behaviors. Many routine behaviors are possible for a given situation. They evolve by selections. This is an evolutionary process. Macroeconomic characteristics provide environment for this evolution. In this sense, micro behavior is conditioned by macroeconomic properties. This evolutionary process normally takes long span of time to take an effect. Thus, even though there are bi-directional causal relations, and therefore we call micro-macro loop, everyday economy can be analyzed by assuming a fixed set of behaviors, whereas each component evolves along time. Majority of Post Keynesians reject this first part of analysis. (I even doubt how many Post Keynesians even reflect on the second part of analysis.)
I do not know whether Dave Taylor knows that most of Post Keynesians refuse to think how macroeconomic phenomena emerge from interactions of many individual agents. I am not only talking to you. I am also talking to my eventual audience who are Post Keynesians. For them, I contend it is necessary to build mcirofoundations of Post Keynesian economics.
Fallacy of composition is another popular argument to refuse microfoundations for Post Keynesian macroeconomics. I will post my response on this subject in the different reply after your last post on July 17, 2020 at 5:02 pm.
I can’t see any conflict here. We all agree that you cannot understand the aggregate by knowing all about individuals. Nonetheless the behaviour of the aggregate and its “emergent properties” depend entirely on the interactions of individuals and groups of individuals. And yes individuals act in awareness of what is happening at the aggregate level.
The key point is Yoshinori’s : “What is needed .. is to stop methodological bavardage and set to work in sincere empirical research and theory building. Methodology does not replace theory.”
Quite so. Methodology is a displacement activity when real theory construction is proving too difficult.
I almost agree with this. It might be better to think of methodology as theorising about theorising, i.e second order theorising.
Sorry, I didn’t have time to comment until now.
I very much welcome davetaylor1’s intention to be provocative. But I wish he’d read what I wrote before saying I am “not prepared to consider either the possibility microeconomics should be based on macroeconomics or the evidence that microeconomics has failed the majority of us”.
On the first point, I said “To deal with emergent properties and causal interactions, the lower-level causal processes must be considered, *along with any causation that may also occur at the higher level*”. So explicitly leaving room for interactions at micro – or meso (e.g. firm) – level to influence macro, for macro to influence micro, and for causation within each level. Multiple causation is the rule, and typically it is of different types.
On the second point, I started off by saying “I think we all agree that the orthodox conception of micro-foundations is misplaced, to put it politely”. The point here is that a valid account of causation at the micro and/or meso level is possible. I’m not sure whether davetaylor1 is ruling this out a priori, or whether he has evidence that the micro level is irrelevant (it’s hard to envisage what such evidence would look like), or perhaps he does agree that the micro/meso level is – or may be – relevant, as well as the macro.
The pity is, I think davetaylor1 is saying some interesting things that I, and others, could learn from, but I’m finding it hard to work out what he is saying. Partly because I am supposed to already know what he is saying (“ignoring what I have repeatedly shared with them”). The various ideas on Second Order Cybernetics look really interesting, but I am unable to piece together what is actually being said here (my inadequacy I’m sure). Is there something you could recommend that I read? – by yourself and/or by others?
By the way, I continue to think that methodological discussion is important if done in the right way. For example, learning more about systems and their properties can contribute a great deal to thinking about economics. And to replacing the orthodox view based on methodological individualism/atomism, optimization, etc.
Sorry I’ve not replied more quickly, both here and to Yoshinori’s post, ill-health having left me unable to concentrate.
My basic answer to Yoshinori has to be that he still hasn’t avoided the fallacy of composition. Looking back through Waldrop on “Complexity” and Gleick on “Chaos” in light of a book I have on the difference between “Logic” and “Semantics” (fun, since it reads logic one way up and semantics the other), I find Waldrop doesn’t even mention “Complication” and Gleick, having failed to understand Shannon’s information science, dismisses it as irrelevant. I suggest Yoshinori is taking the semantic (“complicated”) view of Complexity whereas I am taking the logical (the word literally meaning “with [logically distinct] parts”) . The question the paper I sent him tries to explain what it is that can be logically distinct in an interactive whole? The answer is, such situations our language (words, logos, logic) enables us to distinguish: as longitude, latitude and a good clock enable us to locate the position of events in space-time. Having located a position we need to look there if we are to see what is going on.
In other words, an exact science needs to take account of the adequacy of the language we use. Which is what the vocabulary of the science of electromagnetism and Shannon’s information science (including choice of bits as the unit of information) is all about.
Where Yoshinori’s evolution traces the “complicated” Humean (all words, no reality) evolution of what has been taken to be economics since Adam Smith’s time, my “complex” (two dimensional words and energetic motion) explanation is framed by the two sexes and two sided brains of post-animal humanity: manifest in left and right handedness, left and right in politics, sensory (word-driven) and intuitive (memory-driven) personality, with our brains evolving through time, adding memories from childhood via adult production and distribution to an old age filled with memories not only of words but of different ways of looking at things. Now the majority of us work with words, the ability to try out different ways of looking at things is escaping us just when we most need it. I have also tried to point out that sensory thinkers tend to assume context and seek the meaning of words, where intuitive thinkers tend to take account of historical contexts and ways of looking which make sense of processes in them.
I hope ‘evidencebasedeconomics’ understands that my unfair comment was partly a pun on his pseudonym. Even if he is only pseodo-seriously blaming himself for our personality problem in communicating “(my inadequacy I’m sure)”, I am taking it seriously because that is not what I intended, which was to consider the reason for the inadequacy. Let me take the blame for his taking literally the sentence he objected to (about his not being prepared to ..). Honestly, I had intended to end that with a question mark, and only realised its omission when it was too late, the comment having already been accepted.
I hoped I had dealt adequately with the causation issue in my response to Yoshinori (15 July), beginning “The key point here …”. I don’t think Newtonian causality is irrelevant, nor that the figures useful to mass-producers need to be exact, but I’m trying, like C E Shannon, to establish what CAN be said exactly. In terms of the physical Law of Least Action one has to have a position before one can have motion relative to it, and likewise with acceleration, and forces generating that acceleration. At the stage when life has formed, one had to have microbes before one can evolve linear plant structures, independent (cycling) food resources before physical (animal) mobility, with independent memory (language) to achieve mental mobility in time. Computer languages have to provide for values before they can have variables, modes of interpretation and hence programs which can disentangle the complicated mix of multi-function, multi-user, online applications now found on computers and other information systems like us. Complex number notation had to enable the four winds to be distinguished before navigators could aim for a position and use PID feedbacks to correct their aim. Though this having happened represents technical progress, unfortunately people and their memories can be forgotten and ignored, while those encoded in language can be misleading, as in economics. Moral progress involves deciding NOT to do what we have become technically able to: as in lie, enslave, or kill each other with guns and bombs.
The macro-mesa-micro issue goes back to whether one is defining the terms empirically (so that the Fallacy of Composition occurs in theories of them), or logically (requiring all four prior stages simultaneously, as in the children, men and women and oldies providing or directing the energy physically active in the economy). I am not in a position to impose this as a definition of what a macro economy is. Whether or not I am the only person trying to interpret it this way (as I seem to be), I can and do insist that starting up theorising on a logical basis is necessary if the semantics (the meaning of the word ‘economics’) hung on it is not to be arbitrary. My basis opens up a very much more human, eco-rooted and reliable interpretation of ‘economic’ that ties in with the original meaning of it, i.e. household management. Call what we have ‘capitalism’ if you like, but it is a mistake if not a lie to call it ‘economics’.
Let us remember we are talking about Keynes on microfoundations, postulating on p.6 of his “General Theory” that “The classical postulates do not admit the possibility of … involuntary unemployment”; on p.13 that “there has been a fundamental misunderstanding of how in this respect [wage determination] the economy in which we live actually works”; on p.16 the classical theorists famously resembling “Euclidean geometers in a non-Euclidian world … for which there is no remedy except to throw over the axiom of parallels and to work out a non-Euclidean geometry”. On p.159, also, we find an explanation of the semantic definitions I’m objecting to: “Americans are apt to be unduly interested in discovering what average opinion believes average opinion to be”. The psychological issue is that sensory academics read Keynes looking for ‘economic’ facts and find unfamiliar ways of reasoning (saving face by ignoring them or changing the subject); whereas I as an intuitive reader (already familiar with Chestertonian and biblical scholarship) wondered why Keynes said what he said and tried to follow it up, adding Christians like Ruskin and GKC to his “brave army of heretics” on p.371.
Just in: link from Hazel Henderson to Creating Alternative Futures, with very relevant foreword from Elise Boulding.
Here’s Hazel in Chapter 1:
“We will see how this narrow logic has caused us to over-reward competition while ignoring cooperation and all
the cohesive activities that bind the society together, which we have relegated to the status of unremunerated
activities, to be performed by women. It has caused us to overvalue property rights while undervaluing amenity
rights and to overvalue individual freedom over community needs. Psychologist Robert Ornstein notes that this type
of linear, sequential, quantitative, reductionist cognition is a function of the left hemisphere of the human brain. The right
brain hemisphere processes information in spatial, simultaneous modes, and is the source of intuitive, imaginative
modes of cognition, such as the great hypotheses of science. Both modes of cognition are equally important: the
brilliant intuitive leap of a good hypothesis and the careful processes of its validation or rejection.
“Nowhere is our culture’s overdose of left-brain, Cartesian cognition breaking down faster than in the Tower of
Babel it has created in academia, where reality has for so long been carved up into neat little disciplinary boxes”.
I have explained in one of my post above on July 17, 2020 at 6:13 pm about my concept of micro-macro loop. It is a (rather methodological) concept different from “micro-meso-macro” issue. I do not repeat my explanation here.
As for fallacy of composition, my answer is simple. Why do some phenomena seem paradoxical? Most provably you have a wrong theory (be it spontaneous “scientific” theory of common people or scholars). Why don’t you try to correct that wrong theory by finding a true mechanism under the seemingly contradictory or fallacious understanding? This is not possible without building a theory how interactions of individuals generate macro or mass level phenomenon. So, the existence of “”fallacy or composition” means in reality the necessity of microfoundations.
No, Yoshinori, it means in reality the whole becoming greater than the parts, because it becomes able to do things mere combinations of the parts are not able to do.
But as I’ve said, I’ve been unwell, so I haven’t been able to complete reading and digesting the chapter you sent me. For now let me say your thinking on what made Japanese industry successful for a while (satisfaction with their own situation combined with the challenge of improving quality) is very close to the practical conclusion I reached. The differences concern my getting over the income problem with sufficient credit in lieu of insufficient wages, and largely honorary credit awards for actual achievements (rather than compounded profit shares).
“So, the existence of “”fallacy or composition” means in reality the necessity of microfoundations.”
Yes, but what is also necessary is to find an economically significant aggregative point in the micro-economy where policy can have an equally significant effect in resolving the problems manifested in the economy’s aggregates.
Like the points of retail sale and of note signing for instance.
Craig, I know that Keynes said that economists should be like dentists who relieve the pain of patients. If an economic policy is right one, we can boast we are dentists of national or regional economy. But, it is possible that a policy is wrong or misplaced. Then, we suffer side-effects. Economic policies should be tested by (1) experiences (consequently by history and evidence) and (2) theory (deeper understanding of the mechanism). I am afraid so many people like to talk about policies and contend that they are right to claim it, but history tells they were often wrong.
One of recent history is the failure of “reflationists” who claimed reflation policy (claiming to bring up inflation rate to 2% per an by working on the expectation of the people). Seven years of Abenomics and quantitative easing of “another dimension” (Kuroda) proved that all expectations of reflationists had no basis.
I do not say that Craig claims reflationist policy. I guess you are against it. What I want to say is policy without theory has a high risk to embrace serious side-effects out of imagination. Resolving problems in economy is much difficult than curing some kind of diseases. Are you sure that you can be a good dentist?
Thanks for the reply Yoshinori. The policies I’ve advocated here for years will:
1) beneficially deflate prices and increase purchasing power for all economic agents,
2) end poverty,
3) enable the integration and accomplishment of the dearest aspects of traditionally opposed economic perspectives (left-economic democracy and right-greatly reduce the sting of high taxation and the theft that re-distributive taxation actually is) and
4) enable us to direct consumer, industrial and fiscal policy toward green ecologically sane goals and thus have a chance to survive the existential threat of climate change.
That’s the removal of four infected wisdom teeth and the fitting of two brand new full denture implants.
Theorizing can go on forever, even after conclusions that the economy is monetary in nature (Keen and Graeber), that money is most basically accounting (Keen belatedly) and that the economy is correctly and inevitably embedded within a cost accounting infrastructure (no one since C. H. Douglas) which will enable us to directly implement solutions to the rigged austerity everyone here agrees is part and parcel of the problem.
Policy, most importantly paradigm changing policy implemented at strategic points in the economic/productive process, is what is needed. Policy is the action of systems. I suggest we focus on policies that will benefit all economic agents individual and commercial as soon as possible and also enable us to individually and fiscally confront the existential threat of climate change.
“Theorizing can go on forever, even after conclusions that the economy is monetary in nature”.
It can, if those whose interests are solely monetary, or whose thought (like Craig’s) is based on a monetary tradition, are not refuted when they proclaim the “conclusion” that “the economy is monetary in nature”.
Working from first principles I’ve concluded the opposite. I’ve accepted Fullbrook’s alternative to cost accounting and gone along with E F Schumacher’s requirement of technical adequacy, Sen’s requirement of sufficiency, Boulding’s reminder that we live on and have to sustain ourselves on Spaceship Earth, and here, Yoshinori’s warning that “policy without theory has a high risk to embrace serious side effects out of imagination”. (Relying on politics has indeed proven to be bad policy).
I’ve tried to square Yoshinori’s circle by imagining the effects of a possible theory and like a detective, testing it against the known facts. So imagine I’m right about macro theory outlining the bones of the economy and Jungian psychology (first found by G K Chesterton, now justified physiologically and repeated above by Hazel Henderson) explaining the brains behind the theorising. The fact is money has neither bones nor brains. Money merely makes more money, not something better. It is we who can become capable of – and need to do – something better, both socially and environmentally.
Having put my foot in it with ‘evidencebased’, apologies for my joking to shorten the argument. And good for Keynes (GT p.60): man enough to be able to change his mind about money and thus anticipate PID control as Chesterton anticipated Jungian physiological psychology.
Really ‘Craig’ ? You & your old time ‘funny-money’ chevaliers, still embody the ancient , wisdom of the great Japanese {Zen Buddhist ?} haiku: “When the glaciers melt & recede, old weeds bloom afresh”. But ! ‘FISCALLY confront climate change’ ? Egad ! Is that ‘rigged’ too ? i.e. By whom ? heh’ heh’ heh’ . As if we can’t guess. But CAN THEY actually ‘rig’ the climate? If they’re that omnipotent, wouldn’t it be ‘Pareto-opitimal’ to join THEM, rather than always be pointing fingers at ’em ? But, Take comfort from your righteous agony, Craig old sport. Because Major Douglas, Anton Drexler, Dietrich Eckhart, Nesta Webster et al, are still pining away for you ‘in the place where they are at, where it’s always double drill and no canteen’ to paraphrase Kipling. Please GOOGLE: Norman L. Roth
Sorry to offend so badly Norman. It’s just that such policies seamlessly implemented into the current system are too simple for the intellectual vanities of the erudite and too generally temporal reality altering not to be a paradigm change.