Michael Woodford on models

10 Sep, 2020 at 16:16 | Posted in Economics | 18 Comments

woodfordBut I do not believe that the route to sounder economic reasoning will involve an abandonment of economists’ penchant for reasoning with the use of models. Models allow the internal consistency of a proposed argument to be checked with greater precision; they allow more finely-grained differentiation among alternative hypotheses, and they allow longer and more subtle chains of reasoning to be deployed without both author and reader becoming hopelessly tangled in them. Nor do I believe it is true that economists who are more given to the use of formal mathematical analysis are generally more dogmatic in their conclusions than those who customarily rely upon more informal styles of argument. Often, reasoning from formal models makes it easier to see how strong are the assumptions required for an argument to be valid, and how different one’s conclusions may be depending on modest changes in specific assumptions. And whether or not any given practitioner of economic modeling is inclined to honestly assess the fragility of his conclusions, the use of a model to justify those conclusions makes it easy for others to see what assumptions have been relied upon, and hence to challenge them. As a result, the resort to argumentation based on models facilitates the general project of critical inquiry that represents, in my view, our best hope for some eventual approach toward truth.

Michael Woodford

This is — sad to say — a rather typical view among mainstream economists today. Defending the use of unrealistic and unsubstantiated models with the argument that models make it “easy for others to see what assumptions have been relied upon, and hence to challenge them” is rather far-fetched. It’s like arguing: “We can’t understand what is going on in our complex and uncertain world, so let us set up ‘small-world’ models in which we assume away the complexities and reduce genuine uncertainty to calculable risk, and then let us, with precision and rigour, look at those assumptions and challenge them.” Yours truly fails to see the point. 

Mainstream economic theory today is in the story-telling business whereby economic theorists create make-believe analogue models of the target system – usually conceived as the real economic system. This modeling activity is considered useful and essential. And it’s used both in micro- and macroeconomics. Since everything the economist wants to know is put in to the model, it’s a piece of cake to prove whatever in a ‘rigorous’ and valid way. Deductive certainty is achieved — in the model. Unfortunately, the price one has to pay for getting at ‘rigorous’ and precise results in this way, is making outright ridiculous assumptions that actually impair the possibility of having anything of interest to say about the real world.

Since fully-fledged experiments on a societal scale as a rule are prohibitively expensive, ethically indefensible or unmanageable, economic theorists have to go for something else. To understand and explain relations between different entities in the real economy the predominant strategy is to build models — the preferred stand-in for real experiments — and make things happen in these ‘analogue-economy models’ rather than engineering things happening in real economies.

Mainstream economics has since long given up on the real world and contents itself with proving things about thought up worlds. Empirical evidence only plays a minor role in economic theory, where models largely function as a substitute for empirical evidence. The one-sided, almost religious, insistence on axiomatic-deductivist modeling as the only scientific activity worthy of pursuing in economics, is a scientific cul-de-sac.

Avoiding logical inconsistencies is crucial in all science. But it is not enough. Just as important is avoiding factual inconsistencies. And without showing — or at least warrantedly arguing — that the assumptions and premises of their models are in fact true, mainstream economists aren’t really reasoning, but only playing games. Formalistic deductive ‘Glasperlenspiel’ can be very impressive and seductive. But in the realm of science it ought to be considered of little or no value to simply make claims about the model and lose sight of reality.

Mainstream theoretical economics is still under the spell of the Bourbaki tradition in mathematics. Theoretical rigour is everything. Studying real-world economies and empirical corrobation/falsification of theories and models nothing. Separating questions of logic and empirical validity may — of course — help economists to focus on producing rigorous and elegant mathematical theorems that Woodford et consortes consider as “progress in economic thinking.” To most other people, not being concerned with empirical evidence and model validation is a sign of social science becoming totally useless and irrelevant. Economic theories building on known to be ridiculously artificial assumptions without an explicit relationship with the real world is a dead end. That’s probably also the reason why Neo-Walrasian general equilibrium analysis today (at least outside Chicago) is considered a total waste of time. In the trade-off between relevance and rigour, priority should always be on the former when it comes to social science. The only thing followers of the Bourbaki tradition within economics — like von Neumann, Debreu, Lucas, and Sargent — has given us are irrelevant model abstractions with no bridges to real-world economies. It’s difficult to find a more poignant example of a total waste of time in science.

If the real world is fuzzy, vague and indeterminate, then why should our models build upon a desire to describe it as precise and predictable? The logic of idealization is a marvellous tool in mathematics and axiomatic-deductivist systems, but a poor guide for action in real-world systems, where concepts and entities often are without clear boundaries and continually interact and overlap.

Being told that the model is rigorous and amenable to ‘successive approximations’ to reality is of little avail, especially when the law-like (nomological) core assumptions are highly questionable. Being able to construct ‘thought-experiments’ depicting logical possibilities does not take us very far. An obvious problem with mainstream economic models is that they are formulated in such a way that they realiter is extremely difficult to empirically test and decisively ‘corroborate’ or ‘falsify.’

Contrary to Woodford, I would argue such models have — from an explanatory point of view — no value at all. The ‘thinness’ is bought at too high a price unless you decide to leave the intended area of application unspecified or immunize your model by interpreting it as nothing more than a set of assumptions making up a content-less theoretical system with no connection whatsoever to reality.

18 Comments »

RSS feed for comments on this post. TrackBack URI

  1. Lars says:
    .
    “…. is making outright ridiculous assumptions that actually impair the possibility of having anything of interest to say about the real world.”
    .
    Excellent. So now we can invalidate every model with unrealistic assumptions. Task done.
    .
    Now what do we do?
    .
    Lars has virtually nothing to say on this question. He can emasculate every model known to man yet cannot provide even a vision of how to go about understanding economic behaviour let alone a hard and fast methodological formula.
    .
    Nanikore, I’m sure, will chime in with 500 – 1000 words saying let’s chase facts.
    .
    But which facts? Nanikore totally avoids the question of how we are to discern which facts are relevant. He cannot answer this question.
    .
    The only way to answer the question is to have some sense of the underlying functional relationships. To do that we have to have some form of model. There is no escaping this.

    • Henry, what is your model for the cause(s) of WWI?
      What is your model for the cause(s) of the GFC 2008?
      Or on the video I had earlier Chomsky/Varoufakis

      How did Chomsky arrive at his conclusion to explain why north to East Asia (except the Philippines) industrialised and the rest of the world outside Europe and the Anglosphere did not?

      Did he go to Romer’s or Solow’s growth theory for the explanation? No. He looked at the history of these countries and the facts. And was open minded about where to get those facts. And he could do it because he has had a multidisciplinary education, even if he is not a specialist in the history or economies of those countries. He does not hesitate to look at those things (including the unquantifiable but verifiable facts).

      It was not a model, and certainly not Model, that led him to those conclusions.

      There is no alternative. You have to start with the facts. And the facts must drive the analysis. Of course you are not doing all this on your own. You must also draw on the whole existing literature across disciplines.

      Drawing on philosophy and linguistics it is a case of where the object not the subject leads the analysis.

      Working in a world of abstraction is not the answer. This is also true in micro, as made clear by none other than Ronald Coase:

      https://hbr.org/2012/12/saving-economics-from-the-economists

      Models/Theories are reference points. They should not dictate your analysis or conclusions, including causal mechanisms. If you want to explain why interest rates are flat, you don’t go to ISLM or DSGE, you go to financial and consumer, business associations, governments and central banks that have got the reasons in their reports to explain what consumers, businesses and banks are doing and why. Then you have to put the story together, linking all the information – but not with a model. Just like someone chasing the causes of a major conflict. Or an investigation into a major crime. If you have a theory or model, it comes AFTER you have done this.

      • That’s my 500 hundred words. And no doubt I have repeated myself.

        “But which facts? Nanikore totally avoids the question of how we are to discern which facts are relevant. He cannot answer this question.”

        Go back to the video I put up earlier by Chomsky about what does it mean to be educated, and go in particular to the place where he talks about knowing how to ask the right questions.

        Asking the right questions and then knowing where to go to find the answers is a skill one has to learn. A lot of it is knowing how to access information. Of course no one can do it perfectly. But models like Woodford’s are a barrier in the painstaking process of finding real world answers.

        • According to Chomsky

          (Industrialisation in the global south) = f(whether they were colonised by European powers)

          He did not arrive at that conclusion from a model. His theory or model was arrived at AFTER he had gone through a wide spectrum of historical research.

          If he started and ended with neo-classical growth theory he would not have got very far, and would have wasted a lot of time.

      • Nanikore,
        .
        “Models/Theories are reference points. They should not dictate your analysis or conclusions, including causal mechanisms.”
        .
        I agree, more or less. However, models embody the causal relationships.
        .
        And, how do you organize your facts? How do you make sense of them?
        .
        How do you explain causality without a model? You might think you are not using a model, but there is one or many always lurking in the background.
        .
        Your explanation is tautological and circular.

        • I think the problem is you call explanations models and I don’t. You call causal relations (ie. why something happens or has happened) models.

          Ok, let’s get back to what is, or what should be, a big question in economics.Why did some countries grow, and others did not? (We want to know this because hopefully we would like to bring prosperity to Africa, for example.)

          OK, with big simplification (this is a blog not a book):

          Let’s say someone, went through the history of many Asian and other countries. He found a recurring pattern: countries industrialised by Western powers outside the Anglosphere were not economically successful. Countries in Asia colonised by Japan, and Japan itself, were and was.

          The big question then is why.

          You do not start with a model.

          You start with the historical literature.

          Some historical research will reveal the very different nature of Japanese colonisation compared to that by the US in the Philippines that Chomsky refers to. An industrialisation strategy, brutal, but not commercially exploitative, was put in place by Japan for foreign policy reasons in its colonies, which makes it very unique, and this needs to be well-understood.

          That does not mean that such an industrialisation strategy would work in the Philippines. To understand why not, you need to go into the history of these countries going way back.

          Model will distract you from understanding this and going down there. Or it will take you down a path that ignores important parts of the story.

          I do know something about the former, but I think I have said enough on this for now.

          The point is I do not call these explanations of how and why, and the process of reaching these explanations of how and why, a model. And a pre- existing model should not predispose one to arrive at those explanations of how and why.

          In this example, the only important model I can think of here is List’s, the German development strategy, and the blueprint of Japan’s. But this information is only going to part of the story you need to know. You need to go through the (Japanese language) literature that has already done a lot of the work for you. (And this is the sad thing. Much of this work has been done and the knowledge has been there, potentially invaluable understanding in lifting billion out of poverty, but ignored by the MIT led hegemony for the reasons Varoufakis explains in the video I posted earlier.)

          “And, how do you organize your facts? How do you make sense of them?

          As I have said before, putting a messy story together into a coherent narrative that accounts for all the facts and does not contain irrelevant information or dictated by a model requires skill. It is a necessary skill for humanities and social science subjects. Learning to identify what facts are relevant and irrelevant and what questions need to be asked is part of learning it. But understanding the real world means that in the beginning you also accept the mess.

          “How do you explain causality without a model? You might think you are not using a model, but there is one or many always lurking in the background.”

          I think what you mean by this is how does the subject not influence the object. How do our own educational influences, conditioning, environments, or predispositions not prejudice our conclusions. How, for example, do we avoid things like confirmation bias. If I have implicitly reproduced a model, Marxian or anything else, without explicitly stating it is not that important; the important thing is have I accounted for the facts, are there holes in the analysis? Or is the argument tight?

          That is where a pluralistic education comes in, and why critical reasoning is all-important. There is no magic bullet here, but there are ways we can mitigate it. What is all important is that you read widely, keep going through the evidence and be open to all evidence, and don’t start putting a theory together until the very end. If for example you see the German industrial (List’s development) model in the Japanese literature which is not made explicit, you can put this into perspective this by a pluralistic education, for example, by being aware of the debates relating to the Industrialisation of Britain. It is important you explain things; for example what does capital mean in the Japanese/German case? But understanding the German model does not mean that this has to dictate your analysis and determine your conclusions.

        • Nanikore,
          .
          I think you have taken this discussion to another place, a place that slides the argument away to a distraction.
          .
          We are talking about economic models, not models that explain the kind of history you are looking at, if there are such things. I am not familiar with them and if they exist they would go beyond the sort of questions I am interested in and the sort of questions I thought we were talking about.
          .
          I would agree that an economic model would have little to say about the phenomena you are referencing. The bog standard economic models and their variants aren’t designed to deal with such matters. So I think your argument is irrelevant to the question.
          .
          “You call causal relations (ie. why something happens or has happened) models.”
          .
          As far as I am concerned, models explain causality. It is their fundamental characteristic.
          .
          ” let’s get back to what is, or what should be, a big question in economics.Why did some countries grow, and others did not?”
          .
          Yes, it is a big question. And I would agree it goes beyond economics. But that does not invalidate economic models that might be bought to bear to support other more pertinent factors. As I assert above, you have introduced a slide away, a distraction to the main question.

          • Agree to disagree. I don’t disrespect your views. For me this knowledge is essential. It’s at the heart of the matter in understanding how countries grow or don’t grow. And that is an important question for economists; perhaps the most important (even Lucas thought so). But its about institutions and social structures and everything else that is precisely not in neo-classical theory.

            Someone said that good economics is basically good history. I agree. (You are allowed to disagree.)

            I agree with Varoufakis; to produce universal laws in economics to get a ‘hard science’, time and place had to be taken out of economics.

            I also agree with Varoufakis, you can’t understand capitalism by doing that.

            Theories can be used as reference points, but universal laws don’t exist in social systems. For those reasons models must not frame your analysis. If you adopt such frameworks to explain causation with econometrics to disprove or prove it you’re not going to learn anything about the workings of capitalism – including the actual causation mechanisms. You are merely engaged in a tautologous exercise.

          • Nanikore,
            .
            We are talking about different things.
            .
            You are not talking about the kinds of matters that Woodford is talking about – which is I what I thought we were talking about.
            .
            Seeing you’ve have a strong knowledge of Japan would you not say that, post war, Japan was in some senses of the word, a colony of the USA.
            .
            The surrender treaties allowed the USA to dictate constitutional and political arrangements. Once they were settled, Japan saw an influx of US capital, management and technology. I would argue this was the reason for Japan’s post war growth and economic success. The Americans saw cheap diligent labour and an opportunity to bring Japan into the strategic fold (contra the developing cold war).
            .

            • Henry

              Not from my perspective. Everything I said applies to understanding Japanese macro-policy (and I have read what Woodford has said about unconventional Japanese macro-policy).

              And macro-policy almost everywhere else.

              Thanks, but well, it’s relative. I probably know more than some people who call themselves Japanese experts who do not even speak the language (or engage with people who do), but I am no Ishi Kanji, Hashimoto Juro, or Miwa Ryouichi. Nor am I anywhere a match for the likes of Hugh Patrick or Ronald Dore.

              I agree that you have made an important point. This is the central Danzetsusetsu/Renzokusetsu debate that has long divided Japanese economic historians. This is about how much the evolution of Japanese capitalism really changed during the American occupation and how much it contributed to its success. I would argue it played a role, especially the land reforms; but the foundations of Japanese economic success were laid a long time before that and over a long period. A high level of industrialisation had already been achieved by the late 1930s, by which time Japanese industry was highly competitive.

      • Nanikore,
        .
        Re Chomsky/Varoufakis video.
        .
        Chomsky offered an explanation as to why East Asia industrialized – he said it was related to non colonization by the USA. Firstly, I would dispute this assertion particularly in the case of Japan.
        .
        Now it may be he has developed this theme elsewhere and you have studied his work on this.
        .
        Is this the case?
        .
        Did he offer a detailed explanation for these observations?
        .
        I am not aware of any economic model that explains why East Asia industrialized? (I am not saying these isn’t one.) Are you?
        .
        Why would there be a model that explains it necessarily? Why would you expect any standard neoclassical/Keynesian model to explain these facts? They are not designed to.
        .

        • If I may … cookie cutter belief systems which transcend social constructs and time / space are a largely a projection and not a reflection of fact.

          Per se the example of Japan, do you claim to understand or can speak of its cultural nuances which took epochs to evolve yet some would seek too commodify it, too obscurity, just to lessen the data set to model it.

          Seems some seek purity where it does not exist.

          • If I may it sounds suspiciously like some kinda force …

        • Sjkippy,
          .
          Can you be more specific? Which cookie cutter belief system?
          .
          Are you being rhetorical?

  2. Simply put, the real economy is like shadows on the cave wall in Plato’s classic Cave Allegory. Economists are prisoners seeing only the shadows, ignoring the fire and the figures on the road behind them which originate the shadows. Finance comprises the real figures on the road creating the shadows that are the real economy.
    .
    @ Henry: Finance models can allow contradictions (they can be hidden in linear algebra matrices, for example); and story models allow contradictions. Simple mathematical models that economists typically use ban contradiction because of a fetishization of consistency, and therefore fail to be complete (cf. Godel’s Incompleteness Theorem). There are true facts that the orthodox math models cannot reach through deduction alone.

  3. Henry is right to emphasise that there is no escape from the need for models.
    To illustrate this, let us consider Prof. Syll’s “finest hour”, namely his opposition to the Euro during the 2003 Swedish referendum.
    https://braveneweurope.com/lars-syll-my-finest-hour-swedens-euro-referendum
    https://larspsyll.wordpress.com/2018/05/06/my-finest-hour/
    https://larspsyll.wordpress.com/2013/09/13/this-was-my-finest-hour/
    .
    Prof. Syll openly recognises that the main reason for his opposition to the Euro was based on Keynes’ theories, in particular “the drag on prosperity which can be exercised by an insufficiency of effective demand”.
    .
    He was also influenced by economic theories of optimum currency areas, and a 28 volume special report commissioned by the Swedish government published in 1996.
    All of Prof. Syll’s arguments regarding the Euro are found in these sources.
    .
    Thus Prof Syll’s contribution was essentially an intelligent journalistic interpretation of simplified economic models, namely Keynes’ foundational model of deficient aggregate demand and subsequent models of optimal currency areas.

  4. Henry is right to emphasise that there is no escape from the need for models.
    To illustrate this, let us consider Prof. Syll’s “finest hour”, namely his opposition to the Euro during the 2003 Swedish referendum.
    https://braveneweurope.com/lars-syll-my-finest-hour-swedens-euro-referendum
    https://larspsyll.wordpress.com/2018/05/06/my-finest-hour/
    https://larspsyll.wordpress.com/2013/09/13/this-was-my-finest-hour/
    .
    Prof. Syll openly recognises that the main reason for his opposition to the Euro was based on Keynes’ theories, in particular “the drag on prosperity which can be exercised by an insufficiency of effective demand”.
    He admits that he was also influenced by economic theories of optimum currency areas and a 28 volume special report commissioned by the Swedish government, published in 1996.
    All of Prof. Syll’s arguments regarding the Euro are found in these sources.
    .
    Thus Prof. Syll’s contribution was intelligent journalistic interpretation of simplified economic models, namely Keynes’ foundational model of deficient aggregate demand and the subsequent models optimal currency areas.

    • Sure, Lars (perhaps unconsciously) retreats to models when it serves his gut instincts. When he says employment and stable prices should be the primary goals of public policy, he is relying on orthodox models of income and inflation.


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

Blog at WordPress.com.
Entries and comments feeds.