The New Classical counterrevolution
from Lars Syll
In a post on his blog, Oxford macroeconomist Simon Wren-Lewis discusses if modern academic macroeconomics is eclectic or not. When it comes to methodology it seems as though his conclusion is that it is not:
The New Classical Counter Revolution of the 1970s and 1980s … was primarily a revolution about methodology, about arguing that all models should be microfounded, and in terms of mainstream macro it was completely successful … Mainstream academic macro is very eclectic in the range of policy questions it can address, and conclusions it can arrive at, but in terms of methodology it is quite the opposite.
In an earlier post he elaborated on why the New Classical Counterrevolution was so successful in replacing older theories, despite the fact that the New Classical models weren’t able to explain what happened to output and inflation in the 1970s and 1980s:
The new theoretical ideas New Classical economists brought to the table were impressive, particularly to those just schooled in graduate micro. Rational expectations is the clearest example …
If mainstream academic macroeconomists were seduced by anything, it was a methodology — a way of doing the subject which appeared closer to what at least some of their microeconomic colleagues were doing at the time, and which was very different to the methodology of macroeconomics before the New Classical Counterrevolution. The old methodology was eclectic and messy, juggling the competing claims of data and theory. The new methodology was rigorous!
Wren-Lewis seems to be impressed by the ‘rigour’ brought to macroeconomics by the New Classical counterrevolution and its rational expectations, microfoundations and ‘Lucas Critique’.
I fail to see why.
Wren-Lewis’ portrayal of rational expectations is not as innocent as it may look. Rational expectations in the mainstream economists’ world imply that relevant distributions have to be time independent. This amounts to assuming that an economy is a closed system with known stochastic probability distributions for all different events. In reality, it is straining one’s beliefs to try to represent economies as outcomes of stochastic processes. An existing economy is a single realization tout court, and hardly conceivable as one realization out of an ensemble of economy-worlds since an economy can hardly be conceived as being completely replicated over time. The similarity between these modelling assumptions and the expectations of real persons is vanishingly small. In the world of the rational expectations hypothesis, we are never disappointed in any other way than as when we lose at the roulette wheels. But real life is not an urn or a roulette wheel. And that’s also the reason why allowing for cases where agents ‘make predictable errors’ in the New Keynesian models doesn’t take us a closer to a relevant and realist depiction of actual economic decisions and behaviours. If we really want to have anything of interest to say on real economies, financial crisis and the decisions and choices real people make we have to replace the rational expectations hypothesis with more relevant and realistic assumptions concerning economic agents and their expectations than childish roulette and urn analogies.
‘Rigorous’ and ‘precise’ New Classical or ‘New Keynesian’ models cannot be considered anything else than unsubstantiated conjectures as long as they aren’t supported by evidence from outside the theory or model. To my knowledge no in any way decisive empirical evidence has been presented.
The failure in the attempt to anchor the analysis in the alleged stable deep parameters ‘tastes’ and ‘technology’ shows that if you neglect ontological considerations pertaining to real-world economies, ultimately reality gets its revenge when at last questions of bridging and exportation of model exercises are laid on the table.
Mainstream economists are proud of having an ever-growing smorgasbord of models to cherry-pick from (as long as, of course, the models do not question the standard modelling strategy) when performing their analyses. The ‘rigorous’ and ‘precise’ deductions made in these closed models, however, are not in any way matched by a similar stringency or precision when it comes to what ought to be the most important stage of any research — making statements and explaining things in real economies. Although almost every mainstream economist holds the view that thought-experimental modelling has to be followed by confronting the models with reality — which is what they indirectly want to predict/explain/understand using their models — they all of a sudden become exceedingly vague and imprecise. It is as if all the intellectual force has been invested in the modelling stage and nothing is left for what really matters — what exactly do these models teach us about real economies.
No matter how precise and rigorous the analysis, and no matter how hard one tries to cast the argument in modern mathematical form, they do not push economic science forwards one single millimetre if they do not stand the acid test of relevance to the target.
Proving things ‘rigorously’ in mathematical models is at most a starting point for doing an interesting and relevant economic analysis. Forgetting to supply export warrants to the real world makes the analysis an empty exercise in formalism without real scientific value.
Leave a reply to Asad Zaman 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 
As a business economist buried in current empirical data, what I note most from Lars is the important fact of (ir)rational expectations. We are learning animals, so in an age of nearly total global communication we adjust our expectations.to what we have just learned. This we combine with broadly unforeseen shocks like Mr Trump, Brexit, computing & social media. Asia rising has been long foreseen by specialists, but popular response in the West often seems surprised.
Kenneth Boulding: Mathematics brought rigor to Economics. Unfortunately, it also brought mortis.
Statistical frameworks; probability distributions in particular, do not lend themselves to concepts like contagion, emotion, counterparty risk, and their macro implications; bank runs, market crashes, panics, market manias, popular delusions, and the behavior of crowds. Even unemployment is significantly influenced by these.
Much better to note the correlation between credit and unemployment is -.93 and act accordingly per Steve Keen.
It is better to be roughly right than precisely wrong. (J. M. Keynes)
This dictum applies to the couple of mainstream economics and Post Keynesian economics. The former is precisely wrong, the latter is roughly right. However, is this a felicitous fact? Most of Post Keynesians are almost always arguing on their intuition, because they have no theories behind them. If they are not based on intuition, they cite this and that part from Keynes. They are most of the time roughly right but in some cases seriously wrong, because Keynes made often errors. Even though, they have no criterion to adjust their theories, because intuition, like instinct, is not a reflective knowledge. Most of Post Keynesians are even unaware that they lack their theoretical foundations.
The second sentence for the last should be read:
Even though, they have no criterion to adjust their arguments, because intuition, like instinct, is not a reflective knowledge
“Proving things ‘rigorously’ in mathematical models is at most a starting point for doing an interesting and relevant economic analysis. Forgetting to supply export warrants to the real world makes the analysis an empty exercise in formalism without real scientific value.”
Precisely. It’s mere abstraction and sophistry. It’s arguing over how many angels can dance on the head of a pin. What is required is a paradigm change which inverts and transforms realities in the body of knowledge under discussion. Such changes are the result of a new insight, and the new insight needed now is that if you take firm and rational control of the terminal expression point of all forms of inflation by implementing a relatively high percentage discount/rebate policy at retail sale you can inject virtually as much money into the economy as you please and not only eliminate inflation but beneficially integrate price deflation into profit making economic systems. Oh sure, you’ll need to do some further regulation humans not being entirely rational or ethical, but “It’s better to be roughly right than precisely wrong”, no?
While in no way ignoring the comments . I begin with Lars Syll’s; November 24,2018
“Rigorous and ‘precise’ New Classical or New Keynesian models cannot be considered anything else than unsubstantiated-conjecture’ as long as they aren’t supported by evidence from outside the theory or model. To my knowledge no in any way decisive empirical evidence has been presented.
The failure in the attempt to anchor the analysis in the alleged stable deep parameter’s’ tastes’ and techno9gy shows that if you neglect ontological considerations to real-world economics, ultimately reality gets its revenge when at least questions of bridging and exportation of model exercise’s are laid on the table.”
To me as a non academic but a thinking person with many years practical experience in the real world I think economic conversations need to start with practical experience in the real world before engaging in unsustainable hypothetical theory and models. In other words much of the economic conversation trends to be something like building castles in the air that have no connection to reality.
Furthermore there is an old but logical saying that you can not fix something unless you know why it broke. This brings us back to Lars Syll earlier blog on November 19′ 2018;
“In search of causality relativism is expanding, it is important to keep up the claim for not reducing science to a pure discursive level, we have to maintain the enlightenment tradition of thinking of reality as principally independent of our views of it and the main task of science as studying the structure of reality. Perhaps the most important contribution researcher can make is revealing what this reality is the object of science actually looks like . Science is made possible by the fact that there are structures that are durable and are independent of our knowledge or beliefs about them.”
Here although I am not an academic I think these paragraphs along with Asad Zaman on
Critical realism November 17,2018 deserve some serious thought because they define the importance of understanding the subject before postulating irrational solutions. At least this is the way most of the disadvantaged citizens in the real world see it. Finally I would like to make it clear that I fully support good open conversation where all those who listen to each other, without becoming paranoid about their own understanding of the subject, So that they are able to improve their understanding of the subject and realistically address such problems as the inequality between the rich and the poor. Ted
Your comments summarise the problems of failing analysis very well indeed.
The fundamental failing of economic analysis is that academic economists have never taken to heart the scientific method. I read chemical engineering and throughout my education the scientific method was implicit in everything. I had never heard of Popper and Lakatos until my attempts to understand the failure of there being no empirically validated theories in economics at all. I discovered that despite the many economics papers and books on methodology, no one practised invalidation of economic hypotheses at all. I can only imagine that, this is because if economists did this there would only be a handful of papers which had not been invalidated empirically.
Lars Syll’s remarks on new-classical macroeconomics are substantially correct. But most of the subsequent comments ignore the fact that constructing scientific theories in economics is objectively hard. The reason is that we are not observing a stable system, as most physical systems are, but a mutable and evolving one. Syll’s goes astray when he says “we have to maintain the enlightenment tradition of thinking of reality as principally independent of our views of it”. Economic systems are NOT independent of our views of them. Macroeconomic orthodoxy influences how the system is managed, influences how people expect it to be managed and therefore influences their behaviour. The system is self-referential as well as evolving. That is why when distinguished physicists and engineers have tried their hand at economics they make just as big a mess of it as most economists do.
Theory without measurement and testing is arid. But measurement and prognostication on the basis of past regularities, without theory, is also a trap in an evolving system. Investors in Long-term Capital Management eventually went bust! Theory is necessary but has to be pitched at the level appropriate to the questions being address. If you are trying to understand the behaviour of animals, like badgers, it is not a good idea to start with atomic theory on the grounds that badgers are made of atoms. New classicists would dismiss Keynes, Kalecki and Minsky for lack of “rigour” but their middlebrow theorising has produced insights which a useful macroeconomics.must encompass. Attempts to derive macroeconomics.from axioms of individual rational choice (the atoms) have led only to nonsense.
There are a number of generalisations about science which have many counter-examples and some of the claims about the difficulties of economics are overstated.
“[S]table system, as most physical systems are”: Newton produced his theory of gravitation from observations of planetary motion. Quantum mechanics from observations which could not be explained by classical physics — very “mutable and evolving”. Physical scientists and engineers deal routinely with transient and oscillation systems. The differential equations have to solved and this is generally by numerical analysis.
“Economic systems are NOT independent of our views of them”: This is only partially true. Human decision making may be described in this way. Human actions in the physical world must conform to physical reality. Neoclassical production theory does not meet that requirement and so it can only be a description of some alternate universe with different physical laws.
“[P]hysicists and engineers have tried their hand at economics they make just as big a mess of it as most economists do”: This has some truth, I would commend you to read my paper “Transient Development” (RWER-81). Its mathematical predictions from first principles align with the empirical evidence. It is NOT invalidated by empirical data and it explains paradoxes which have no explanation in conventional analysis.
These are all excellent points dear colleagues. But, my failure to see what is new about our discourse continues to haunt me even post Black Friday shopping frenzy!
All one has to do for real data, explanation, observation, and formulation of admittedly imperfect analyses is to visit a shopping mall, observe who is begging around the corner and compare to those on the verge of starvation.
True! However to demonstrate the validity of appropriate solutions, it is necessary to have quantitative predictions of what the application of proposed solutions will have. This is where the deficiency is. Conventional analysis fails to provide empirically justifiable predictions, which I feel is implied by what you have written above. If proper validation were to be applied to conventional analysis, the lack of justification provided by conventional analysis would be apparent to all. This will remove most of the obscuring fog of invalid theorising and leave proper examination of suggested solutions.
You make a good point. For example, current macro theory distinguishes between expenditures that are investments versus consumption. So individual incomes are thought to be saved for investment or consumed. Ignoring for now that a lot of investment comes from banks and not from savers, more fundamentally, look at we consider to be consumption rather than investment.
Having a child, feeding a child, educating a child, clothing a child, in brief, raising a child, is not considered an investment by economists. Think about that.
Correct. Abstraction too often results in objectification….of everything including the experience of oneself and everybody else.
This of course is why Wisdom, whose source of information IS direct existential experience OF oneself/one’s consciousness (no matter whether you believe it to be supernatural or integrally natural) ….is so important.
Systems were made for man, not man for systems. Economists need to make that truth their most fundamental ethic and policy effect….and we’ll undoubtedly find our way…because consciousness, again no matter how one wants to define it, is such a fundamental and powerful fact and experience.
Lars, I have the same difficulty with this posting as with many on this blog, and many who post on the blog. You say that models must be confronted with reality, otherwise the models are worthless. I need you to explain to me what this comparison entails. Most importantly, what is reality? All this leads me to the conclusion that for models to be useful they must be compared to data sets, perhaps dozens or more data sets. Looking for the combination of data that is consistent (in ways you leave undefined) with the predictions of the model. There is no reality in this work. Only a model or models and data. If this is what economists ought to do, and either are succeeding or failing in this effort, then why can’t you say so? All these references to reality are both time-wasting and futile. Plus, it’s confusing as hell. Not a desirable combination for teaching neophyte economists.
Now these questions raise others related to the comments here. First, what is the difference between data and theory, between the standard comparison of induction and deduction? The separation of the two is arbitrary. A result of cultural constructs that develop in the process of creating groups or entire societies. Economists tend to draw a hard and unbreakable line between the two. Economists also tend to make theory superior to observation or experience. For me the interesting questions are why economists make these assumptions? And how and where did they learn to reach such conclusions? But most important are the many questions about how actors (scientists and otherwise) connect observation/experience and theory. Obviously, this is a work of human judgement. But what is human judgement and how does it relate to the general understanding of science?
You ask important questions. I suggest that Caldwell (1984) might provide some answers
The following quotation from Caldwell (1986) pp. 675–676 is worthy of some thought:
“Much methodological work proceeds in the following way. A critic, who typically is a member of a particular group within economics (say, Group A), argues that Group B’s approach to economics is wrong. The critic accomplishes this by asserting that there is a proper and correct procedure for science. He then asserts that, while members of Group A regularly follow this procedure, members of Group B fail to follow it. Thus, Group B’s approach is unscientific and should not be further considered. The critic rests his case. Usually there is a counterattack in which a member of Group B asserts that Group A’s vision of science is flawed.
There would be nothing wrong with such an approach if there existed a widely agreed-upon definition of what constitutes legitimate scientific practice. For better or worse, most methodologists have looked to philosophers of science to provide such a definition. However, even a casual reading of the philosophy of science of the last three decades reveals that no widely agreed-upon definition of legitimate scientific practice exists. As a result, the kind of argumentation outlined above is bound to be inconclusive.
Even worse, the endless rounds of debate that are a natural consequence of this kind of approach to methodology make methodological work itself seem pointless. Because methodologists typically look to general models of scientific practice articulated within the philosophy of science for their arguments, they often neglect the richness and diversity of the ways economics is actually practised. Instead of shedding light on what economics is, methodological attacks and counterattacks involve stylized conceptions of how economics is done. For example, Austrians, Marxists, and other heterodox groups in economics are often dismissed by ‘mainstream’ critics for not following falsificationism. But it is certainly an open question whether any group in economics follows falsificationism.”
The final sentence explains much. I would go much farther than this. All schools of economics are invalidated by the empirical evidence. There are NO valid theories within their works and there are are no models which predict published empirical data.
References:
Caldwell, Bruce J. (1984). “Some problems with falsificationism in economics”. In: Philosophy of the Social Sciences 14.4, pp. 489–495.
– (Nov. 1986). “Towards a Broader Conception of Criticism”. In: History of Political Economy 18.4, pp. 675–681. url: http://dx.doi.org/10.1215/00182702-18-4-675.
Frank, I agree with you and Caldwell. No lesser authority than the American Physical Society makes your point. Twenty years back the APS appointed a committee to write a definition of science that all physicists could support and would help the APS counter the damage being done by pseudoscience. After a year of work APS membership rejected the proposed definition. APS never tried again. So, what constitutes legitimate scientific practice remains an open question. No argument.
My questions are much simpler. According to Lars, models must be confronted with reality, otherwise the models are worthless. I assume that reality is shorthand for something. Otherwise it is, like the models, worthless. So, what’s the answer? What’s the shorthand? Second, in terms of this shorthand, what are the answers for my second group of questions?
Ken reiterates the questions from his 2 Dec. 11.28 a.m,
I will not try to provide a general explanation of how corroboration and falsification should be carried out. I will only offer the example of my analysis, Transient Development RWER-81 pp. 135−167. There you can see that I provide extensive comparison with a wide range of published empirical data. In fact, there is no shortage of empirical data covering every imaginable area of economics.
I believe the reason my paper is the only example, demonstrating that the empirical data conforms with prediction, is that I solved the differential equations obtained from a first principles analysis. There are few examples of other papers which make similar attempts. Furthermore, it will only be through similar analysis that empirically validated analysis will be extended to cover a broader range of issues in the future.
Frank, you write in RWER 81, pp. 148-149, “Quite clearly, the empirical data, in both confirming stability and movement, present a reality which is more complex than simply testing whether the stability of the ratio provides a binary result. Transient analysis provides quantitative descriptions of the output rate, equation (14), and of capital, equation (12). The precise evaluation of the output:capital ratio is determined as follows.” And at p. 161, “Transient mathematical relationships, derived from first principles, provide a parsimonious quantitative description of the development histories of manufacturing projects and industries.” I can accept both statements as an effort to establishment a benchmark for examining empirical data. But I still don’t have the answer to my main question. Is this the reality Lars uses to “confront” models? I still can’t complete the circle of comparing models to reality, and then using the results.
Ken, I have asked something similar of Lars Syll in his response to my claim of my analysis NOT being invalidated by the empirical data, but similarly received no reply.
If I have understood your final question correctly, I think of this in a slightly different way. Valid theory must conform to the requirements of the quantity calculus. I am sure every physical scientist will agree with that assertion. Then valid models should predict, with reasonable accuracy, values which conform to what is found to be true in the real world they are modelling. If their model results map the real world situation, they are seeking to emulate, then their theory is validated but no theory can ever be proved to be true. Essentially the statement may be considered to be, there are no known counter-examples. In the physical sciences, there are examples of theories which are known to be incomplete in an absolute manner, but which are good enough to provide sufficiently precise results for practical application. For some reason economists seem to be unable to accept this situation, which is fully acceptable to physical scientists.
Frank, your description is not that different from Feynman’s description of science in the famous film of his lectures to introductory physics students. I tell you what I’ve said repeatedly about Feynman’s definition of science in this film – it’s bunk. First, figuring out what empirical results are and how to compare them with the results predicted by a model or hypothesis is always uncertain and subject to dispute. Humans are embedded in a culture which shapes how they experience everything. Also, the human senses are imperfect and ambiguous. What humans see, hear, feel, taste, and smell are always uncertain and incomplete. Plus, all empirical observations involve interpretation by the observer of what’s there. Sticky process. On the other side every empirical observation can be explained by an inestimable number of theories. Again, we bring in human judgement and interpretation to choose an explanation. Even deciding when we’ve reached that “good enough” point in our work rests on these same uncertainties. At every juncture uncertainty and disputability. This is the fundamental structure of all science. Of all human knowledge and action.
This was not one of Lars’ finest blogs. I agree with Ted he was clearer in some earlier ones, but here find Yoshinori, Frank, Craig and Ken not providing Ted with “good open conversation where all those who listen to each other, without becoming paranoid about their own understanding of the subject”. Gerald Holtham, however, seems worth responding to. Of course I am going to repeat my own views for his benefit, but let me explain that having been taught general science at school and challenged by realising I still had no idea what physics, chemistry and biology were about, nor of the relationship between mathematical and catholic [macro] economics, I have spent most of my long life trying to make sense of them, and still occasionally learn something new, as I did recently from chaos theory.
Gerald, then, says Syll “goes astray when he says ‘we have to maintain the enlightenment tradition of thinking of reality as principally independent of our views of it’. Economic systems are NOT independent of our views of them”. And later: “Theory is necessary, but has to be pitched at the level appropriate to the questions being addressed. If you are trying to understand the behaviour of animals, like badgers, it is not a good idea to start with atomic theory on the grounds that badgers are made of atoms”.
As, like Syll, a Critical Realist, I understand he is distancing the realist tradition from David Hume’s (and Ken’s) conflating knowledge with that which is perceivable. I differ from Syll in having concluded that reality includes energy, which can be known only by inference. I agree entirely with Gerald that economic systems are not independent of our views of them (i.e. at a perceptual level). However, only APPLIED theory has to be applied at the level appropriate to the questions being addressed. BASIC theory has to choose axioms able to account for the beginning rather than the present state of evolution, given atoms are derivable from energy but not vice versa.
It is then meaningful to say badgers are more massive than air (so terrestrial) and living (so programmed by genes) and intelligent (so adaptable to physical change) but not yet human (adaptable to conceptual change). That is not a lot, but it does get you looking in the right ball-park.
Similarly with economics as institutionally structured by our reaction to prior concepts. The road to the bank may not be paved with gold or even tarmac, but relationships between the different types of function in an economic system really exist as communication channels just so long as we use them. In the macro theory time evolves, generating a past as it moves from the present into the future. This is still true and should be reflected in Applied Economics, which it is when the economic system is considered as a PID control system. The Basic Theory shows you what economics is (and its having evolved into money making), and gets you looking at what is going on in the communication channels, wherein (like blood in us) you will find not only power being transferred but corrective information feedbacks performing the equivalent of antigens, antibodies and clotting agents. Too much of the clotting agent (money making) and the system seizes up. Limit credit (or tax away excess) and the monetary problem will largely disappear, letting us face the real problem of global warming.