Modern economics — confusing models with reality
from Lars Syll
What does concern me about my discipline … is that its current core — by which I mainly mean the so-called dynamic stochastic general equilibrium approach — has become so mesmerized with its own internal logic that it has begun to confuse the precision it has achieved about its own world with the precision that it has about the real one …
While it often makes sense to assume rational expectations for a limited application to isolate a particular mechanism that is distinct from the role of expectations formation, this assumption no longer makes sense once we assemble the whole model. Agents could be fully rational with respect to their local environments and everyday activities, but they are most probably nearly clueless with respect to the statistics about which current macroeconomic models expect them to have full information and rational information.
This issue is not one that can be addressed by adding a parameter capturing a little bit more risk aversion about macro-economic, rather than local, phenomena. The reaction of human beings to the truly unknown is fundamentally different from the way they deal with the risks associated with a known situation and environment … In realistic, real-time settings, both economic agents and researchers have a very limited understanding of the mechanisms at work. This is an order-of-magnitude less knowledge than our core macroeconomic models currently assume, and hence it is highly likely that the optimal approximation paradigm is quite different from current workhorses, both for academic and policy work. In trying to add a degree of complexity to the current core models, by bringing in aspects of the periphery, we are simultaneously making the rationality assumptions behind that core approach less plausible …
The challenges are big, but macroeconomists can no longer continue playing internal games. The alternative of leaving all the important stuff to the “policy” type and informal commentators cannot be the right approach. I do not have the answer. But I suspect that whatever the solution ultimately is, we will accelerate our convergence to it, and reduce the damage we do along the transition, if we focus on reducing the extent of our pretense-of-knowledge syndrome.
A great article that also underlines — especially when it comes to forecasting and implementing economic policies — that the future is inherently unknowable, and using statistics, econometrics, decision theory or game theory, does not in the least overcome this ontological fact.
It also further underlines how important it is in social sciences — and economics in particular — to incorporate Keynes’s far-reaching and incisive analysis of induction and evidential weight in his seminal A Treatise on Probability (1921).
According to Keynes we live in a world permeated by unmeasurable uncertainty – not quantifiable stochastic risk – which often forces us to make decisions based on anything but “rational expectations.” Keynes rather thinks that we base our expectations on the confidence or “weight” we put on different events and alternatives. To Keynes, expectations are a question of weighing probabilities by “degrees of belief,” beliefs that often have preciously little to do with the kind of stochastic probabilistic calculations made by the rational agents as modelled by “modern” social sciences. And often we “simply do not know.”
How strange that social scientists and mainstream economists, as a rule, do not even touch upon these aspects of scientific methodology that seems to be so fundamental and important for anyone trying to understand how we learn and orient ourselves in an uncertain world. An educated guess on why this is a fact would be that Keynes concepts are not possible to squeeze into a single calculable numerical “probability.” In the quest for measurable quantities one puts a blind eye to qualities and looks the other way.
So why do economists, companies and governments continue with the expensive, but obviously worthless, activity of trying to forecast/predict the future?
Some time ago yours truly was interviewed by a public radio journalist working on a series on Great Economic Thinkers. We were discussing the monumental failures of the predictions-and-forecasts-business. But — the journalist asked — if these cocksure economists with their “rigorous” and “precise” mathematical-statistical-econometric models are so wrong again and again — why do they persist wasting time on it?
In a discussion on uncertainty and the hopelessness of accurately modelling what will happen in the real world — in M. Szenberg’s Eminent Economists: Their Life Philosophies — Nobel laureate Kenneth Arrow comes up with what is probably the most plausible reason:
It is my view that most individuals underestimate the uncertainty of the world. This is almost as true of economists and other specialists as it is of the lay public. To me our knowledge of the way things work, in society or in nature, comes trailing clouds of vagueness … Experience during World War II as a weather forecaster added the news that the natural world as also unpredictable.
An incident illustrates both uncer-tainty and the unwilling-ness to entertain it. Some of my colleagues had the responsi-bility of preparing long-range weather forecasts, i.e., for the following month. The statisticians among us subjected these forecasts to verification and found they differed in no way from chance. The forecasters themselves were convinced and requested that the forecasts be discontinued. The reply read approximately like this: ‘The Commanding General is well aware that the forecasts are no good. However, he needs them for planning purposes.’
Leave a comment
—– 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 
I may be an old largely uneducated member of the public however I find your above blog resonates with my actual experience in the actual real world because from observation it appears to me that statistics and forecasts continually bear no relationship to the real world Ted
A battle of methodologies regarding the foundations of probability theory took place in the 1930s. One one side were Keynes and Knight who argued that there was a fundamental difference between risk – where probabilities could be quantified, as in roulette, or dice – and uncertainty — where the probabilities, were unknown, as in a horse race. On the opposite side were Ramsey, De-Finetti and others who argued that rational behavior in face of uncertainty involved ascribing subjective probabilities to the uncertain events, which reduced uncertainty to risk, eliminating this difference. This fallacious argument became widely accepted, and led to a complete disregard for the existence of uncertainty, and the pretence-of-knowledge represented by the assumption that we know all the probabilities of all the uncertain events. This arguments continues to command widespread support and acceptance. In order to understand the difference between risk and uncertainty, it is necessary to refute the argument for existence of subjective probabilities made by Ramsey, De-Finetti and later, Savage. This is what I have done in my recent paper “Subjective Probability Does Not Exist” I have shown that the argument is fallacious because it equates preference with choice, even though one can make choices without having preferences. This easy equation of two different things was at the heart of logical positivism, which was wildly popular at that time. The observable choice was taken to be the sole scientific way to capture the unobservable preference, and positivists argues that science involves replacing unobservables in theories by their observable manifestations. This positivist mistake led to the confusion between risk and uncertainty which persists to this day
Quote from above: “the future is inherently unknowable”
While there is some truth in that assertion, there are many nuances. I would always separate prophesy from prediction. Prophesy is essentially guesswork. Prediction is not. Build a road bridge across an important river crossing point — predicting that the bridge will be heavily used might be seen as a certainty — but it remains a prediction until the work is complete and empirical facts emerge. Most decisions are based on: if this action is taken, this will be the result — a perfectly valid procedure.
The significant word lies in the title — model. The range of possibilities encompassed by this word enormous. Discussion should not be at the meta-level of all models, but of specific models. Such as DSGE mentioned in the inset by Caballero. Whilst the comments are true, limiting them to this level of specificity is less useful. All equilibrium theory is nonsensical. Markets do not clear! However wrong that assumption, it is made in order to simplify the mathematics because the valid mathematics is hard.
Models lie on a continuum from valid to invalid. Models will, therefore, produce results consistent with their position on that continuum. Comment must be based on the quality of the results in aligning with reality — not an inchoate cry against all models.
The above leads to examining the best alternative to models. It is first principles analysis. First principles analysis is not based on models. Appropriate first principles mathematical analysis produces solutions which are valid in the real world. Only when this is fully understood and accepted will economics be able to progress and become a science.
Frank, you frequently speak of first principles analysis never define it in any great detail. It would be helpful if you would define this term and/or point to some sources you use for your understanding of what this means.
The Wikipedia page “First principle” is a convenient source.
It states: “A first principle is one that cannot be deduced from any other”; “In mathematics, first principles are referred to as axioms or postulates” and “In physics and other sciences, theoretical work is said to be from first principles, or ab initio, if it starts directly at the level of established science and does not make assumptions such as empirical model and parameter fitting”.
In “Transient Development” (RWER-81), the first principles I invoke are the definitions of productivity and of technical progress. With the assumptions of returns to scale and of maintenance being proportional to the use made of the production tools, a theory of production is developed. It does not rely on any form of curve fitting. It is ab initio.
In neoclassical production hypotheses, equations are introduced which are intended as models of reality. Their parameters are fitted to the empirical evidence. As convenient mathematical descriptions of the data, this is a perfectly valid procedure. However, the analyses generally consider how closely the equations fit the data. The closeness of fit is then interpreted as supporting the model being a proper theory of production. The authors make the logical error of asserting that correlation implies causation.
In my Figure 4, you can see how Solow (1957) fits five different curves to his data. Then by examining the closeness of fit, he discards the only one which might have had any theoretical validity. How ridiculous the curves (other than the discarded straight line) are (when plotted over the range of capital values necessary for the real theoretical relationship) is immediately apparent.
Reference
Solow, Robert M, “Technical change and the aggregate production function”, The Review of Economics and Statistics 39, 3 (1957), pp. 312–320.
I have been failing to send the above for three days. I succeeded using Internet Explorer. Three other browsers on a Linux machine reported errors associated with the php being used. I did however use one of those browsers to post another comment on a later blog posting.
Thanks Frank, sorry to be so much trouble ;-) Will look at closely, still reading paper.
I wonder if this argument applies to predictions by climatologists of anthropogenic global warming.
“Anthropogenic global warming” isn’t a prediction, it is an observed fact. Models differ on just how this will effect us and are less certain the longer the prediction is. These models are based on physical facts which are well known, repeatable, and reliable. If you pump CO2 into the atmosphere the climate will warm, period. That’s just a consequence of well known physical laws, akin to a “prediction” that if you push the gas pedal down further your car will go faster.
All economists, without exception, simply do not understand how to use models. No forecasting, no final results. Math is just a tool to generate large sets of simulated strategies.
Human decision-makers look at the sets and select the most acceptable – sufficiently good by a number of criteria and sufficiently protected from multiple risks.
I am inclined to believe that All Models are false (none are valid) but some are useful. Useful is always relative to some GOAL for modeling. Friedman and his gang have got ONE part of the defence of ridiculous neoclassical models right — the model need not (and indeed should not) resemble reality — a 1-1 scale map would be useless. However, the second part is to show that the model is USEFUL for some purpose — when we examine this carefully, we find that neoclassical models are INDEED useful for the purpose of justifying the inequality of wealth and for justify tax cuts for the rich — in general for ET1%; see my post
In reply to Ishi Crew February 3, AT 2:OOPM and
Ed Seedhouse February 3 at4:53PM
In reply to global warming is that the populist concept of global warming is seriously distorted when only one part of the story is told in order for misguided zealots and money grubbers to pursue their aims. Note I am not I am not denying that we need to be concerned about the increase in global warming in order to do something constructive about it. My concern here is that due to intention and or lack of knowledge essential information is ignored. For example that historical and scientific knowledge has accumulated vast amounts of actual physical and scientific information. That actually prove that our world has experienced climate changes ice ages over millions of years.
This claim is backed up by Environmental Archaeologists who have done a great deal of work on this subject. On source of this information can be found in Archaeology Theories, Methods and Practice (1991) where they explain deep see cores have yielded vast amounts of information. For example Microscopic fossils of the foraminiferan species Globorotalia truncatolinoides whic coils to the left during cold periods and the right during warm ones—Second, they analyse, by mass spectrometer, fluctuations, in the ratio of the stable oxygen isotopes 18 and 16 in the calcium carbonate of the foraminiferan shells. Variations discernible by these two tests reflect not simply changes in temperatures, but also oscillations in the continential glaciers””
My point here is that if you are using physical facts to support your argument it is invalid until
all the relative physical and scientific evidence is considered. Here I acknowledge that the mentioned writers may already know this but the reason I bring this up is to inform the readers that global warming is a natural process that has been repeated many times over millions of years. Again I repeat I acknowledge that we need to be concerned about global warning but we need to be aware of false or incomplete information designed by the money grubbers in their pursuit of wealth and power.
I am sorry, this bafflegab of yours doesn’t change the facts. Physics, unlike economics, can make reliable predictions because it is founded on well established and universal regularities and symmetries.
Your belief that “all the relative physical and scientific evidence” has no been considered only speaks your ignorance out loud. I think you must be a secret economist…
This was uncalled for. Ted’s first statement was an admission he lacks [college] education. So what? He was trying to say what he can see, which is further than those who live in ivory towers can. What he didn’t have words for is the difference between macro and micro reasoning, i.e. logical and quantitative. IF one is using quantifiable physical facts then Ted’s point is the legal requirement: one needs “the truth, the whole truth and nothing but the truth”. IF one is using macro reasoning It is indeed true that physics is founded on reliable observations of inertias, operating on timescales relative to which ours are trivial. That doesn’t mean infamous people don’t tell lies and practice “economy with the truth”.
re the nice guys comments thank you very much they are greatly appreciated because I think part of the process of reforming economics is to involve and educate the mass of ordinary people in how to think and inform them in simple terms that their are alternative to the present ruling main stream economists. On this basis I find this post of Lars Syll and the following posts has produced some serious conversations. Also recently in one of the blogs it was mentioned that there is great scope for this in the technological media.
See my comment just now under “What is wrong”, about what people should be being taught at school before they get to university.
A major part of my work is long-range planning. That’s done using scenarios. Scenarios are a tool for helping us take a long view in a world of great uncertainty and many unknowns. Scenarios are a way for articulating the different pathways that might exist for you tomorrow and finding your appropriate movements down each of those possible paths. Scenario planning is about making choices today with an understanding of how they might turn out. Scenarios are organized ways for humans to dream effectively about our collective future. In practice, scenarios resemble a set of stories, either written out or more often spoken. These stories are built around carefully constructed “plots” that make the significant elements of the world stand out boldly. The approach is more a disciplined way of thinking than formal methodology. Scenarios are not predictions. It is simply not possible to predict the future with certainty. An Arab proverb says just that – “he who predicts the future lies even if he tells the truth.” Rather, scenarios are tools for helping people learn by presenting for consideration and comparison alternative images of the future. Not by simply extrapolating current trends. This learning includes a heightened awareness of the many uncertainties and unknowns facing humans in any situation. This is a powerful antidote to the common belief in certainty that makes governments, businesses, and social organizations unprepared to deal with future surprises, which then often overwhelm these organizations’ efforts to deal with upcoming events and actors. The motto of long-range planners is “be prepared for whatever happens.” It is this ability to act with a knowledgeable sense of risk and potential reward that separates the prepared leader and organization from a bureaucrat or a gambler.
I think this is a very helpful way of looking at the issues, Ken. It is really what I am doing myself. I’ve tried to boil that down to concrete aims, but in the end people with different experience need to be persuaded, so the approach has to be presenting a scenario enabling them to mentally experiment with what is being proposed, so they can see for themselves both its advantages and what can go wrong. Isn’t this what Steve Keen is doing at a more technical level with his ‘Minsky’ simulations? It is certainly the key objective at the decision making stage of SSADM systems analysis.
Dave, a proviso I suggest be added to your description is that scenario alternatives to Minsky need to be considered in the planning.
Steve Keen is my choice of probably the best economist on the planet. Maybe co-equal with Michael Hudson. However, his Minsky software is basically “epicycle” research because it resides virtually entirely within the current paradigm of Debt Only. Yes, DSGE/Neo-classical Economics ignore/deny money, banks and debt…but factoring them correctly into theory…isn’t sufficient, and neither is one-off actions like “a modern debt jubilee”. It’s reform, it’s palliation of the actual problem, it is still hypnotization by the current paradigm. Minsky was right that modern economies are financially unstable. So how do we actually stabilize it?
Keen correctly states that whenever the rate of credit creation dips the economy will quite soon go into recession….unless (he thinks) the government injects more money and/or credit into the economy. This is basically new Keynesianism, MMT, his modern debt jubilee one-off and maybe a little QE for the People thrown in. That’s good. It’s better than the present enforced austerity and it looks some of the policies of the new paradigm straight in the face…..but it still doesn’t actually recognize the new paradigm, how it can be implemented so as to saturate the economy as the current paradigm of Debt Only has done and so replace its primacy and dominance. And because it’s only a reform instead of the paradigm change that has been needed for the last 150-200 years if not the last 5000….it will inevitably get morphed back into a tweaked DSGE/Neo-classical economics or worse just like what happened to the reform of Keynesianism Keen and every other heterodox economist is happy to point out.
Find the new paradigm and boldly go with it. Don’t Escape From Freedom. (Erich Fromm) Expand your mind, re-integrate your life….and enjoy!
Ken, the issue of alternatives is important, but for most people the only real alternative is what we’ve got. That however has subsystems, and alternative scenarios usually amount to rearrangements within those. Minsky isn’t a scenario but a program automating the working out and simultaneous display of the subsystems of the current scenario in action, i.e. (comparatively speaking) in “real time”.
Intellectuals predominantly use the language side of their brain and unemotionally work out implications, thus not developing imagination capable of directly (intuitively) seeing and feeling the significance of several things happening and interacting at once. I’ve seen Minsky in action, making good this defect of imagination by automating the simultaneous working out and display of graphs of economic sub-processes with different timescales. The effect is that the emotions alert one to significant changes and interactions. This is proper science. It doesn’t tell you what to do, it reveals what is important and needs looking at.
Dave, I’ve no experience with the modeling tool Minsky, you mention. But the literature about the tool notes that it is like Mathcad, Mathematica, Mathlab and other mathematical modeling/ simulation tools, but optimized for accounting-based, flow-of-funds analysis. This tells me Minsky uses optimization mathematics. Optimization models cannot be dynamic. They cannot handle interval data. In simple terms, optimization models provide snapshots based on the data and time-period inputs. Dynamism requires interval data and differential equations. But the model is certainly capable of modeling Hyman Minsky’s assumptions, and dozens or hundreds of variations of them. Hyman Minsky’s theories are not particularly new or imaginative. But they did capture aspects of how economic networks function that mainstream economists have pushed to the margins or out of economics entirely. One aspect of the model Minsky I find interesting is it translated Hyman Minsky’s theories into mathematics. Something Minsky himself never did.
“This tells me Minsky uses optimization mathematics. Optimization models cannot be dynamic. They cannot handle interval data.”
You are simply wrong, Ken. The issue really is the difference between macro and micro. Macro relationships are logical and all-embracing even if the micro-events are innumerable (as the inclusion of interval data suggests). Your minor premise is wrong because navigation DOES dynamically optimise a course, but what doesn’t show up in the theory of that and does in the Minsky displays is the cumulative effect of losing your bearings (e.g. by misreading of or magnetic influences on your [moral] compass), so that over a period of time your course becomes not optimum but chaotic. Mathematically there are exponential AS WELL AS optimisable processes going on, and their interaction is what I saw Steve Keen’s displays demonstrating. It is also where I saw Keynes heading: when the economy has lost its direction it needs to re-establish its position so that it CAN re-set its direction.
Dave, as I’ve said before, the differences between macro and micro are made-up by people. People decide there ought to be such differences. Ditto, with logic. Dynamic optimization within these parameters requires interval data. Economics is not like physics. The data of sun energy output is interval, continuous. Unemployment, debt, etc. are not. As I’ve also said before these limitations don’t make mathematics in economic useless, just limited. Please stop hawking the idea that economics can be like physics.
Ken, you are driving me to the conclusion you are a twit. Please stop hawking the idea that physics is based on the same scientistic methods as economics.
Dave, I’m not driving you anywhere. If you’re going it’s all on your own. Physics and economics study different parts of the world. If each is a science, however, they operate the same in terms of approach. Observe, experiment, write up, review, repeat. Currently, mainstream economics seems to do little observation or experiment. Economists just seem to write up “off the tops of their heads,” with no regard at all for getting at least some facts about economics and economies right.
If that’s a Monty Python twit, then I’m amused. Otherwise, the word is funnier than the people it depicts.
The idea/claim that economics can be mathematically modeled after physics and the critique that this is wrong is a very consistent theme in all the heterodox literature I have read. Amusingly enough, reading Dave’s comments in RWER I have had the thought that Dave indeed seemed to hold just such a belief: that economics can be like physics.
So, why not just ask. Dave, do you think that economics is like physics? Do you think that the solution to the current crisis in economics lies in “getting the math right”? What exactly do you believe?
No, Rob. I don’t think that economics is like physics, and as I had just indicated, I believe there is something to be learned not from “getting the maths right” (i.e. not making mistakes using the current maths) but by getting the right maths (i.e. topological logic rather than quasi-geometrical functions).
Navigation is not about prediction, it is about three different types of error correction (at the time the errors become evident) via information feedback loops. Its logic applies to all reasoning about flows just as conventional logic applies to the genesis or inclusiveness of sets. But flows occur in physical contexts to which the laws of physics apply, and the control of Adam Smith’s invisible hand can be effected in physical ways as well as via human agency. So I can demonstrate a whole range of different control arrangements, some physical but others involving human agency constrained by the physical, and what we do depends on what we believe we are doing –
which of these paradigm we are taking for granted – insofar as we are not physically confined to traditional practices.
I can show each of these being used in some economist’s argument or working practice, and how the physical models are subsystems of the “physical and informational” models in the same way as a switched-off TV is a subsystem of a working one. As subsystems they each have merit, but as coherent pictures of the whole they are misleading, especially the physical ones which completely ignore the relevance of the communication and storage of information in humans and human social institutions.
I think “the solution to the current crisis in economics” is to rethink its physical architecture along the up-to-date lines of time-sharing, multi-user, multi-function, networked data processing obeying the laws of and using the capabilities of the internet, pre-programmed to enable us to navigate our own economies, rather than encouraging Dad to gamble away the proceeds of a Ponzi racket and telling the kids they must survive on what little Mum has left in her purse.
In reply to Ed Seedhouse February 2918 at 3:29 pm I may be ignorant but that does not alter te fact that I din not try to alter the fact you described. My attempt was to try and point out that when telling a story telling all of the relevant facts reduces the possibility of misinterpretation by those with ulterior motives, such as the elite wealthy manipulators of neoliberal ideology. Apart from that in a democracy where the editors of Real World Economics have invited anyone to make a comment on the blogs and posts is in my humble opinion a step foreword in the right direction.
Then on February 5, 2018 Dave Taylor’s post brings up the issue of education at school level. “In a way we are addressing different audiences. I’m arguing to justify what should be taught at school, so that by the time they get to university, youngsters can see that their is more to economics than going shopping”.
Here in Australia from actually talking to some year twelve student before they went to university they said that they had been told how free trade was a good thing. Then when I asked if they received any explanation of the possibility that free trade might make it difficult for them to find a job after university. Their answer was nothing had been mentioned thus they were being indoctrinated not educated to think. The point I am trying to make here is that teaching students and people how to think so that they can reasonably evaluate the “bafflegab” of some academics. Furthermore the importance of thinking is supported by Fullbrook with::
“the neoclassical monopoly in the classroom and its probation on critical thinking has meant that it has brainwashed successive generations of students into viewing economic reality exclusively through its concepts, which more often than not misrepresent or veil the world, especially todays world. Nearly all of these neoclassical notions have a bearing on judgments about social, cultural and economic policy. Consequently, if society were to learn to think about economic matters outside the neoclassical conceptual system, it would almost certainly choose different policies. One of PAE’s projects has been to expose some of the conceptual lunacies of today’s mainstream, both in the terms of the concepts it uses and the concepts it lacks.” Edward Fullbrook (2005) PAE vol 32 p34.
Once again I thank all those who are trying to improve equality and justice in the world even those who think I am ignorant because they take the part of the devils advocate and make me think.
Any model that rejects the two policies of a universal dividend and a 50% discount/rebate at the point of retail sale consciously or unconsciously refuses to cut the Gordian Knot that holds the tug of war between necessary injection of additional credit into the economy and price inflation. And any model that entitles the dominating and monopolistic paradigm of Debt Only to continue is a sop to Cerberus/Finance.