Home > Uncategorized > Paul Samuelson — a case of badly invested intelligence

Paul Samuelson — a case of badly invested intelligence

from Lars Syll

Paul Samuelson claimed that the ‘ergodic hypothesis’ is essential for advancing economics from the realm of history to the realm of science.

But is it really tenable to assume that ergodicity is essential to economics?

The answer can only be — as I have argued

here

here

here

here

and

here — NO WAY!

Obviously yours truly is far from the only researcher being critical of Paul Samuelson. This is what Ole Peters writes in a highly interesting article on Samuelson’s stance on the ‘ergodic hypothesis’:

Samuelson said that we should accept the ergodic hypothesis because if a system is not ergodic you cannot treat it scientifically. First of all, that’s incorrect, although I think I understand how he ended up with this impression: ergodicity means that a system is very insensitive to initial conditions or perturbations and details of the dynamics, and that makes it easy to make universal statements about such systems …

Another problem with Samuelson’s statement is the logic: we should accept this hypothesis because then we can make universal statements. But before we make any hypothesis—even one that makes our lives easier—we should check whether we know it to be wrong. In this case, there’s nothing to hypothesize. Financial and economic systems are non-ergodic. And if that means we can’t say anything meaningful, then perhaps we shouldn’t try to make meaningful claims. Well, perhaps we can speak for entertainment, but we cannot claim that it’s meaningful.

And this is Nassim Taleb’s verdict on Samuelson’s view on how to do economics:

mathscienceThe tragedy is that Paul Samuelson, a quick mind, is said to be one of the most intelligent scholars of his generation. This was clearly a case of very badly invested intelli­gence. Characteristically, Samuelson intimidated those who questioned his techniques with the statement “Those who can, do science, others do methodology.” If you knew math, you could “do science.” … Alas, it turns out that it was Samuelson and most of his followers who did not know much math, or did not know how to use what math they knew, how to apply it to reality. They only knew enough math to be blinded by it.

Samuelson was a formalist committed to mathematical economics. Where did this methodological stance take him? Nowhere I would argue. It is one thing to come up with ‘minimally realist’ models of ‘possibly true’ worlds.’ And something completely different to give us relevant truths about real-world systems. Just giving us a set of possible truthsdoesn’t suffice. We are looking for how-actually explanations and not how-possiblyexplanations. The model world is not the real world. Model worlds may well be ergodic. Real-world economies are not. So how then, could one possibly learn anything economically relevant and interesting using an ergodic model that obviously badly misrepresent the world in which we live? Providing us with knowledge of possibilities is not enough. Genuine understanding and explanation demand knowledge of causes, not knowledge of possibilities. Although mathematics may teach us — without empirical knowledge — why a mother cannot evenly divide her twenty-three apples among three children without cutting any of the apples, we have to remember that most relevant economic problems and issues are of a much more complex and different nature. In real-world economies, purely mathematical explanations do not take us anywhere.

Samuelson’s elaborations on revealed preference theory show this in an illuminating way. The very raison d’être for developing revealed preference theory in the 1930s and 1940s was to be able to ascertain people’s preferences by observation of their actual behaviour on markets and not having to make unobservable psychological assumptions or rely on any utility concepts. This turned out to be impossible. Samuelson et consortes had to assume unchanging preferences, which, of course, was in blatant contradiction to the attempt of building a consumer and demand theory without non-observational concepts. Preferences are only revealed from market behaviour when specific theoretical constraining assumptions are made. Without making those assumptions there are no valid inferences​ to make from observing people’s choices on a market.

But still, ​a lot of mainstream economists consider the approach offered by revealed preference theory ​great progress. As people like Robinson, Georgescu-Roegen, and Kornai have shown, this is, however, nothing but an illusion. Revealed preference theory does not live up to what it claims to offer. As part of the economist’s​ tool-kit, ​it is of extremely limited use.

If we want to be able to explain the behaviour and choices people make, we have to know​ something about people’s beliefs, preferences, uncertainties​, and understandings. Revealed preference​ theory does not provide us with any support whatsoever in accomplishing that.

So why did Samuelson basically fail in his theoretical endeavours?

One reason is that he did not — like most of mainstream economists that have followed in his footsteps — seriously reflect on the fact that limiting model assumptions in economic science always have to be closely examined since if we are going to be able to show that the mechanisms or causes that we isolate and handle in our models are stable in the sense that they do not change when we ‘export’ them to our ‘target systems,’ we have to be able to show that they do not only hold under ceteris paribus conditions and a fortiori only are of limited value to our understanding, explanations or predictions of real economic systems.

Another reason is that he confused epistemology and ontology. The validity of the inferential models we as scientists use ultimately depends on the assumptions we make about the entities to which we apply them. Applying a relevant modelling strategy presupposes far-reaching ontological presuppositions. If we are prepared to assume that societies and economies are like urns filled with coloured balls in fixed proportions, then fine. But — really — who could earnestly believe in such an utterly ridiculous analogy? We are not mainly interested in what could be true, but what is true. In a real world full of ‘unknown unknowns’ and genuine non-ergodic uncertainty, urns are of little avail.

The ‘ergodicity hypothesis’ provides a totally wrong-headed picture of the processes that operate in real-world economies. Ergodicity is not relevant for studying actual economic processes. It doesn’t actually explain anything of what goes on in real-world economies. How it could be claimed by Paul Samuelson to be essential for advancing economics to the realm of science is beyond comprehension.

  1. Paul Davidson
    July 10, 2019 at 4:56 pm

    I am glad to see Lars attempting to attract attention to my in print 1968 analysis that economics involves nonergodic stochastic processes and that Paul Samuelson is largely to blame for making economists think economics is an ergodic process. Samuelson did more destructive things to economic theory than even Milton Friedman.

    • merijntknibbe
      July 11, 2019 at 8:02 am

      David,

      a question about Friedman and Schwartz (1963). It’s a very good book (the empirical results) but a failed endeavour (the analysis of these results) as Friedman and Schwartz did not even try to analyze their results in a Flow of Funds framework (where did money growth come from…) which is, considering that the knew Copeland and his efforts and the fact that Flow of Funds (money flows) approach had already been developed means that they were not up to the state of the economic art. Instead of this a weird and dogged fascination with only one liquid asset class and a neglect of debt. Do you have any direct knowledge why Friedman and Schwartz disregarded the Flow of Funds?

  2. Dominique
    July 10, 2019 at 5:23 pm

    I certainly agree that financial and economics systems are not ergodic. However, the statement:

    “This turned out to be impossible. Samuelson et consortes had to assume unchanging preferences, which, of course, was in blatant contradiction to the attempt of building a consumer and demand theory without non-observational concepts. Preferences are only revealed from market behaviour when specific theoretical constraining assumptions are made. Without making those assumptions there are no valid inferences​ to make from observing people’s that choices on a market.”

    Nor with that one:

    ” If we want to be able to explain the behaviour and choices people make, we have to know​ something about people’s beliefs, preferences, uncertainties​, and understandings. Revealed preference​ theory does not provide us with any support whatsoever in accomplishing that.”

    The idea behind the concept of revealed preference is to bypass the unobservables; in other words, economists cannot do science with unobservables. To say that without making these assumptions, there are no valid inferences to make…. would contradict Set Theory:
    Preferences of economic agents are in :ordinal Space. The only way to transfer them to real space, where mathematical operations are permitted, is to find an output set from the system that is isomorphic to the ordinal space in equilibrium. If there are changes in the preference set, the isomorphic set will respond and the equilibrium point will wobble. If we were to depend on people’s beliefs, we would never be able to do science.

  3. Craig
    July 10, 2019 at 8:24 pm

    The solution to a non-ergodic system’s problems is to find a point of leveraging power and leverage it sufficiently with the most potent factor in the system that an inversion of the problematic condition results. That is the definition of a pattern-paradigm change.

    Modern economies are in a state of continual scarcity disequilibrium regarding total individual incomes in ratio to total costs and prices. The solution is not a stochastic equilibrium, but the higher abundant disequilibrium of incomes to costs/prices.

    Retail sale is the leverage point, money in the form of gifts is the factor and the 50% discount/rebate policy at retail sale is the means of paradigm change.

  4. Mike Ryan
    July 10, 2019 at 10:07 pm

    Paul Samuelson’s job was to produce an effective cultural response to defeat communism – he succeeded with his wildly effective textbook. His mathematical gibberish confused everyone and allowed Academia to create a false science that removed moral decisions from the realm of profit seeking corporations.

    Academia continues to promote the lies of Neoclassical Economics as the means justifies the end – no communism or socialism in America. Universities collect billions each year teaching this nonsense to students that are REQUIRED to take the class.

    The long term effect is increased power and control for corporations as the common populace is still starry eyed about free markets. Meanwhile corporations and wall street cooperate to suck trillions out of our economy with leveraged buyouts. It is very easy to name acquisitions that resulted in lost jobs, inflation and eventually a failed corporation.

    What a shame….

  5. Ikonoclast
    July 11, 2019 at 12:30 am

    We should take note of Ken Zimmerman’s statement in the post above, “There is no separation between culture and economy.” If we take that as a founding principle in social ontology (I certainly would), then we can say confidently that there can be no science of economics. The attempt to turn the management of economics into a science would founder just as would any attempt to turn the management of culture into a science.

    From a pragmatic point of view we can, in modernity, make cultural-economic decisions which are one or both of science-informed and/or ethically informed. The science-informed part is (relatively) easy, in theory. If hard science tells us (as it now does) that endless material growth and continued use of fossil fuels will destroy the capacity of the biosphere to sustain humans and the current complex web of life, then we need to stop growing the economy and stop using fossil fuels. We need to manipulate economic parameters not according to some prescriptive theory (about the sanctity of property, money and markets) but according to the clear requirements of the science. This implies superseding in some way the current operations of property, money and markets. The plain path and only known practical alternative for large, modern, democratic states is to take statist, dirigist action.

    The other requirement is for ethical action. What cannot be achieved by science-informed actions must be achieved by ethical actions via a return to Moral Philosophy. Of course, this begs the question, “Which ethics?” The old recipe for jugged hare starts with “First catch your hare.” First, we must catch, or rather develop, our ethics. We are not starting this chase just now. It’s an age-old pursuit. We have many ethical traditions to hand although these are still developing and still contestable. Decisions on what to do must not be handed off to the autopilot system of property, money and markets. That’s an abdication of the need to make moral decisions. Again, if the operations of property, money and markets lead to mass species extinctions and mass human poverty and die-offs (as examples) then their operations must be progressively superseded and replaced by democratic statist, dirigist action (as the only feasible alternative power to capital at this juncture in history).

    To use a medical analogy, If a severe disease is in progress then “heroic measures” are required. “In medicine, heroic treatment or course of therapy is one which possesses a high risk of causing further damage to a patient’s health, but is undertaken as a last resort with the understanding that any lesser treatment will surely result in failure.” – Wikipedia.

    It is very clear that the course of continuing to permit the operations of market fundamentalism (of the neoliberal corporatist type) will kill the patient (the biosphere and the human race with it). The only medicine currently available, in the potential quantities and proven effectiveness required, is democratic statist action.

    In summary, this would mean demoting economics to a quite subsidiary, fourth order role. Democracy, ethics and science (human and ecological impact science more than production science) would take their place at the top of the cultural decision process.

    • Craig
      July 11, 2019 at 4:04 am

      Love is universally recognized as the pinnacle human ethical concept. Grace is nothing more and certainly nothing less than love in action/policy. Grace as in giving, and viz monetary economic systems gifting, is the broadest, deepest and most specific concept to so apply.

  6. July 12, 2019 at 3:00 am

    An easy explanation of Ergodic Hypothesis from Nicholas Nessim Taleb – Ergodicity means that time ensembles behave like cross sections — here is an example where this fails:
    Consider a more extreme example than the Casino experiment. Assume a collection of people play Russian Roulette a single time for a million dollars –this is the central story in Fooled by Randomness. About five out of six will make money. If someone used a standard cost-benefit analysis, he would have claimed that one has 83.33% chance of gains, for an “expected” average return per shot of $833,333. But if you played Russian roulette more than once, you are deemed to end up in the cemetery. Your expected return is … not computable.

    Based on this dramatic difference between sequential probabilities and cross-sectional probabilities, NNT writes that:

    I effectively organized all my life around the point that sequence matters and the presence of ruin does not allow cost-benefit analyses; but it never hit me that the flaw in decision theory was so deep. Until came out of nowhere a paper by the physicist Ole Peters, working with the great Murray Gell-Mann. They presented a version of the difference between the ensemble and the time probabilities with a similar thought experiment as mine above, and showed that about everything in social science about probability is flawed. Deeply flawed. Very deeply flawed. For, in the quarter millennia since the formulation by the mathematician Jacob Bernoulli, and one that became standard, almost all people involved in decision theory made a severe mistake. Everyone? Not quite: every economist, but not everyone

    The idea I just presented is very very simple. But how come nobody for 250 years got it? NNT blames it on economists lack of insight, and on their lack of skin in the game =

    Adding people without fundamental insights does not sum up to insight; looking for clarity in these fields is like looking for aesthetic in the attic of a highly disorganized electrician.

    For the full essay see: The Logic of Risk-Taking

  7. Ken Zimmerman
    July 18, 2019 at 2:18 pm

    “The ‘ergodicity hypothesis’ provides a totally wrong-headed picture of the processes that operate in real-world economies” is nearly right. The next sentence is “on the mark,” if we omit the “actual.” So, it now reads, “Ergodicity is not relevant for studying economic processes.” Unfortunately, the other social sciences also assume ergodicity for their research standard. However, at least sociologists, psychologists, etc. make a little effort to educate their students on the many cautions when applying methods based on the assumption of ergodicity.

    But still economics and economies can be studied scientifically. Anthropologists do it. Anthropologists consider and locate aspects of people’s individual and collective lives, which is to say their lives and societies, in terms of how these aspects relate to one another in an interconnected, though not necessarily bounded or very orderly, whole. For example, an anthropologist might want to study how household organization among a particular set of people is related to, say, wealth and changes in wealth, and vice versa (in an ideal world an anthropologist would want to know how all the elements of people’s lives and societies are related to one another). As this suggests, anthropologists tend to want to see people’s lives in the round. Anthropologists also want to know about the relationship between what people think and say on the one hand, and on the other what they do. Generally, anthropologists distinguish the two as culture and culture performed (society). Mostly through experience, anthropologists have learned that observing people’s actions provides a profounder and more complete understanding of their culture than just listening to what people say and write.

    Finally, two universal features of anthropology are important. First, the perspective is fundamentally empirical and naturalistic. Extended participant observation, empirical naturalism, has come to define the field. Which means that anthropology is both a semi-humanity and semi-science. My position is that this is the only genuine science. Second, in part because of the importance of extended participant observation and in part because of the concern to approach people’s lives in the round, anthropologists in general, are reluctant to think in terms of social laws and universals. Anthropologists have studied many societies in different parts of the world and have come up with almost no social laws that apply throughout specific regions, much less that apply globally. Put differently, anthropology tends to be an idiographic or particularizing discipline, rather than a nomothetic or generalizing one.

    • James Beckman
      July 19, 2019 at 5:28 pm

      Hi, Ken, it’s been a while. I was trained as both an economist & anthropologist. I lived with the Kpelle Tribe of Liberia, West Africa for 15 months before their civil wars & Ebola. The people look & seem to act the same now as then. They are dissimilar in various ways to tribal people from African nations which had major French or British former colonies; freed American slaves went to Liberia starting in the 1820’s & the American influence has continued. Traits can be discussed between such tribes, but we really hesitate to generalize too much. You hit that nail on the head, my friend.

      • Ken Zimmerman
        July 20, 2019 at 10:25 am

        Good to hear from another anthropologist on this blog, James. Ulf Hannerz quotes a 1960s remark that traditional anthropologists were “a notoriously agoraphobic lot, anti-urban by definition”. Various changes and stresses in the Western World as well as in the “Third World” (the latter being the habitual focus of attention of anthropologists) brought the attention of “specialists in ‘other cultures'” closer to their homes. For example, urban Houston where I did my first anthropological work.

  8. Robert Locke
    July 19, 2019 at 11:33 am

    Australopithecus Prometheus, the most impressive person i learned about in antro i, physical anthropology, in 1953. let’s hear it for the anthpologists! But there is a fuzziness in this science of empirical-naturalness, which monotheism won’t tolerate. and they control this academic culture.

    • Ken Zimmerman
      July 19, 2019 at 1:44 pm

      Robert. Also known as Australopithecus africanus (“little foot”). Thought to be a direct ancestor of the Homo genus. I believe it was paleontologists who unearthed the skeleton. Empirical-Naturalism = observation in natural settings and recording everything. Without, if possible disturbing the setting. Yes, Anthropology’s conclusions are always fuzzy because human culture and society are fuzzy. To which monotheism are you referring? There are several, even in academia. And still more in business and industry, finance and banking.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.