On models and simplicity

from Lars Syll

Quotes about Simplicity (804 quotes)When it comes to modelling yours truly does see the point emphatically made time after time by e. g. Paul Krugman about simplicity — at least as long as it doesn’t impinge on our truth-seeking. ‘Simple’ macroeconomic models may of course be an informative heuristic tool for research. But if practitioners of modern macroeconomics do not investigate and make an effort of providing a justification for the credibility of the simplicity assumptions on which they erect their building, it will not fulfil its tasks. Maintaining that economics is a science in the ‘true knowledge’ business, yours truly remains a sceptic of the pretences and aspirations of  ‘simple’ macroeconomic models and theories. So far, I can’t really see that e. g. ‘simple’ microfounded models have yielded very much in terms of realistic and relevant economic knowledge.

All empirical sciences use simplifying or unrealistic assumptions in their modelling activities. That is not the issue – as long as the assumptions made are not unrealistic in the wrong way or for the wrong reasons.

Being able to model a ‘credible world,’ a world that somehow could be considered real or similar to the real world, is not the same as investigating the real world. Even though all theories are false since they simplify, they may still possibly serve our pursuit of truth. But then they cannot be unrealistic or false in any way. The falsehood or unrealisticness has to be qualified.

Explanation, understanding, and prediction of real-world phenomena, relations, and mechanisms therefore cannot be grounded on simpliciter assuming simplicity. If we cannot show that the mechanisms or causes we isolate and handle in our models are stable, in the sense that when we export them from are models to our target systems they do not change from one situation to another, then they – considered ‘simple’ or not – only hold under ceteris paribus conditions and a fortiori are of limited value for our understanding, explanation, and prediction of our real world target system.

The obvious ontological shortcoming of a basically epistemic – rather than ontological – approach, is that similarity’ or ‘resemblance’ tout court does not guarantee that the correspondence between model and target is interesting, relevant, revealing, or somehow adequate in terms of mechanisms, causal powers, capacities or tendencies. No matter how many convoluted refinements of concepts are made in the model, if the simplifications made do not result in models similar to reality in the appropriate respects (such as structure, isomorphism, etc), the surrogate system becomes a substitute system that does not bridge to the world but rather misses its target.

Constructing simple macroeconomic models somehow seen as ‘successively approximating’ macroeconomic reality, is a rather unimpressive attempt at legitimizing using fictitious idealizations for reasons more to do with model tractability than with a genuine interest in understanding and explaining features of real economies. Many of the model assumptions standardly made by mainstream macroeconomics – simplicity being one of them – are restrictive rather than harmless and could a fortiori anyway not in any sensible meaning be considered approximations at all.

If economists aren’t able to show that the mechanisms or causes that they isolate and handle in their ‘simple’ models are stable in the sense that they do not change when exported to their ‘target systems,’ they do only hold under ceteris paribus conditions and are a fortiori of limited value to our understanding, explanations or predictions of real economic systems.

That Newton’s theory in most regards is simpler than Einstein’s is of no avail. Today Einstein has replaced Newton. The ultimate arbiter of the scientific value of models cannot be simplicity.

As scientists, we have to get our priorities right. Ontological under-labouring has to precede epistemology.

  1. yoshinorishiozawa
    November 25, 2022 at 4:59 am

    Of course, “[t]he ultimate arbiter of the scientific value of models cannot be simplicity.”However, in my opinion, “simplicity” in economics should be argued much differently either from Krugman’s or Lars Syll’s way.

    Simplicity is often necessary and it is not wise to reject simplicity in favor of more realistic complicated systems theory. For example, if one wants to obtain a model that can predict the (near) future development of an economy, the complication is inevitable. The example of Laurence Klein’s Project Link showed that increasing number of variables could not increase the accuracy of its prediction beyond a certain limit. If we contemplate why this happened, we should know that a large complicated simulation does not give a more accurate prediction nor a better understanding of how the economy works.

    We must recognize that we are not in the stage of knowing the macroeconomic behavior more than a certain degree of accuracy. It is necessary to change our research strategy. If we leave macroeconomic prediction, there are many relations in which we can detect a causal link between events. Medicine developed in this way. Typical examples are microbial pathogenesis. Koch postulated three or four criteria to identify whether a microbe is the cause of a disease. Although, at Koch’s time, we did not know the microbe level mechanism on how the microbes proliferates and how they infect human bodies, the pathogen theory was extremely useful. Even modern physics gave birth by discoveries of simple rule like Galileo’s law of falling bodies. It was a law whose the range of validity was very narrow.

    Economics did not arrived at the stage of knowing even the law of falling bodies or the second law of motion (law of acceleration). In stead of searching Newton’s stage of economics, we should better make efforts to find simple but fundamental causal relations in the economy. [This is a question of epistemology.] These efforts must be classified as a part of microeconomics, but there are clear differences between these efforts and the actual microeconomics. Neoclassical microeconomics is now a “macroeconomics” in the sense it tries to explain all what happens in the economy as a whole. See for example general equilibrium theory. This is now the device that combine almost everything into an equilibrium state of the economy. We should abandon that kind of frameworks.

    Lars Syll is right when he states that “As scientists, we have to get our priorities right.” But, the option is not between either epistemology or ontology. We must be aware of what is necessary and possible, given the state of our knowledge in economics. Generally speaking, both epistemology and ontology are necessary, but we should not stay at this abstract level of questions.

  2. Steven Klees
    November 25, 2022 at 5:28 pm

    Simple quantitative models may be useful in the physical and natural sciences but not in the social sciences. I agree with Lars on most of what he says but if I connect his (and mine) previous critiques of regression (RWER Blog 10/11/22) you realize simple (macroeconomic or other) quantitative models are never useful to describe what happens in the real world since they are all so misspecified. The upshot is, e.g., that macroeconomists argue endlessly over whose simple model is right and all sides find significant coefficients to justify their results. The results of the regression may inspire interesting theorizing but the empirical basis for using it that way is empty.
    Steve Klees, University of Maryland

  3. yoshinorishiozawa
    November 27, 2022 at 12:25 pm

    Steven Kleen,

    I read your Presidential Address
    Klees, Steven. 2008. “Reflections on Theory, Method, and Practice in Comparative and International Education.” Comparative Education Review, 52 (3) (August): 301-28.
    https://www.researchgate.net/publication/237777843_Presidential_Address_Reflections_on_Theory_Method_and_Practice_in_Comparative_and_International_Education

    I loved it very much. It was based on wide experience and deep reflection. I also agree with you when you pointed out in the paper in RWER that

    [U]nfortunately, regression analysis methodology is a dead end, no better than alchemy and phrenology, and someday people will look back in wonder at how so many intelligent people could convince themselves otherwise. (Kleen 2022 RWE #72, p.92)

    It seems we have the same (or at least similar) judgements on the actual state of economics. Please read my first comment above (still hidden under moderation although it was posted before your post). A true paradigm change is requested.

  4. Steven Klees
    November 27, 2022 at 5:40 pm

    The previous comment uses examples from medicine and physics to talk about the state of our “knowledge” in economics. These examples are irrelevant. Social science is fundamentally different from the physical and natural sciences. The author says we should “make efforts to find simple yet fundamental causal relations in the economy.” We’ve been trying to do this forever with little, if any, success. If particles had intentions, there would be no physics. We have 8 billion people with different intentions and “causal relations” or regularities may not exist and, if they do, we certainly have no agreement on them. What we have in economics and in all the social sciences is endless debate about these relations. Mathematical models – simple or complex – have given us no agreed upon knowledge, nor will they. We need to accept that and turn our attention to how to better bring our debates into the policy sphere. Instead of offering expert knowledge, we must find ways of brokering our debates in widespread, democratic, participatory, policymaking processes.

    • rsm
      November 28, 2022 at 10:32 pm

      From the wikipedia article on Bruno Latour:

      “In the laboratory, Latour and Woolgar observed that a typical experiment produces only inconclusive data that is attributed to failure of the apparatus or experimental method, and that a large part of scientific training involves learning how to make the subjective decision of what data to keep and what data to throw out. Latour and Woolgar argued that, for untrained observers, the entire process resembles not an unbiased search for truth and accuracy but a mechanism for ignoring data that contradicts scientific orthodoxy.”

      Also, from a physics.stackexchange discussion, “About ‘the worst prediction in all of physics'”:

      “As it is presently stated QFT predicts a ludicrous cosmic vacuum energy. You can of course choose to ignore that prediction.”

  5. Steven Klees
    November 27, 2022 at 10:06 pm

    Thanks for the compliment. I think many of us agree that a true paradigm change is needed — in economics and in research methods!

  6. yoshinorishiozawa
    November 28, 2022 at 1:16 am

    Dear Steven, it seems our posts are crossing each other.

    > Social science is fundamentally different from the physical and natural sciences.

    All right. Let us take an example from economics, which is a part of social science(s). Moreover, we are talking in economics blog and we are both economists in a broad sense.

    The principle effective demand is often taken as the demarcation line between mainstream and many strands of Keynesian or heterodox economics. Even among the Keynesians, there are many interpretations on it. One of the reasons of this confused state is due to the bad formulation in Chapter 3 of The General Theory by Keynes himself. What I want to argue here is not the principle itself. Let us consider why we are so confused with regards to the principle of effective demand.

    I published a paper on the principle of effective demand last year:
    The principle of effective demand: a new formulation (open access)
    https://www.jstage.jst.go.jp/article/revkeystud/3/0/3_67/_article/-char/en

    My interpretation (or formulation) is quite simple. It reflects (describes) firms supply behavior vis-à-vis the demand for a product of a firm. The demand for the product in near future is not known. We only know the past series of the demand. Then, a possible behavior is to use the moving average of the past series as a surrogate and determine the volume of production for today or for the week. If we interpret the principle of effective demand in this way, it would be much easier than to argue whether the principle (in any form of interpretations) holds for the economy as a whole. We can enquire how the firms are behaving with this regard. We will be able to estimate how many firms in the economy (or how many firms in an industry) are behaving in this way.

    After publishing my paper, I came to acknowledge that many economists are thinking differently. They were thinking that the principle of effective demand is a hypothesis on how the total of the demand (the aggregate demand) is determined. Although this may be an important question when we want to know the total number of employment, it is quite complicated and would be practically impossible to establish a general rule, because so may factors intervene.

    Even in the case of the economy in which we (or 8 billion people) live, my interpretation as a behavioral assumptions of firms may provide an example that is “simple yet fundamental causal relations in the economy.” Of course, it is evident that we must change our points of enquiry. The first step for a paradigm change must be to ask different questions from different viewpoints.

    I do not side with the preconception that “social science is fundamentally different from the physical and natural sciences.” I believe economics has many things to learn from the history of natural sciences. One of the troubles with us is that we did not learned from it appropriately. Mere formal imitation is not the way we should take.

  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 )

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.

%d bloggers like this: