Saturday, March 21, 2015

A Quick Point on Models

According to Keynes the purpose of economics is "to provide ourselves with an organised and orderly method of thinking out particular problems"; it is "a way of thinking ... in terms of models joined to the art of choosing models which are relevant to the contemporary world." (Quoted here.)

I want to amplify on that just a bit. The test of a good model is not whether it corresponds to the true underlying structure of the world, but whether it usefully captures some of the regularities in the concrete phenomena we observe. There are lots of different regularities, more or less bounded in time, space and other dimensions, so we are going to need lots of different models, depending on the questions we are asking and the setting we are asking them in. Thus the need for the "art of choosing".

I don't think this point is controversial in the abstract. But people often lose sight of it. Obvious case: Piketty and "capital". A lot of the debate between Piketty and his critics on the left has focused on whether there really is, in some sense, a physical quantity of capital, or not. I don't think we need to have this argument.

We observe "capital" as a set of money claims, whose aggregate value varies in relation to other observable monetary aggregates (like income) over time and across space. There is a component of that variation that corresponds to the behavior of a physical stock -- increasing based on identifiable inflows (investment) and decreasing based on identifiable outflows (depreciation). Insofar as we are interested in that component of the observed variation, we can describe it using models of capital as a physical stock. The remaining components (the "residual" from the point of view of a model of physical K) will require a different set of models or stories. So the question is not, is there such a thing as a physical capital stock? It's not even, is it in general useful to think about capital as a physical stock? The question is, how much of the particular variation we are interested is accounted for by the component corresponding to the evolution of a physical stock? And the answer will depend on which variation we are interested in.

For example, Piketty could say "It's true that my model, which treats K as a physical stock, does not explain much of the historical variation in capital-output ratios at decadal frequencies, like the fall and rise over the course of the 20th century. But I believe it does explain very long-frequency variation, and in particular captures important long-run possibilities for the future." (I think he has in fact said something like this, though I can't find the quote at the moment.) You don't have to agree with him -- you could dispute that his model is a good fit for even the longest-frequency historical variation, or you could argue that the shorter frequency variation is more interesting (and is what his book often seems to be about). But it would be pointless to criticize him on the grounds that there isn't "really" such a thing as a physical capital stock, or that there is no consistent way in principle to measure it. That, to me, would show a basic misunderstanding of what models are.

An example of good scientific practice along these lines is biologists' habit of giving genes names for what happens when gross mutations are induced in them experimentally. Names like eyeless or shaggy or buttonhead: the fly lacks eyes, grows extra hair, or has a head without segments if the gene is removed. It might seem weird to describe genes in terms of what goes wrong when they are removed, as opposed to what they do normally, but I think this practice shows good judgement about what we do and don't know. In particular, it avoids any claim about what the gene is "for." There are many many relationships between a given locus in the genome and the phenotype, and no sense in which any of them is more or less important in an absolute sense. Calling it the "eye gene" would obscure that, make it sound like this is the relationship that exists out in the world, when for all we know the variation in eye development in wild populations is driven by variation in entirely other locuses. Calling it eyeless makes it clear that it's referring to what you observe in a particular experimental context.

EDIT: I hate discussions of methodology. I should not have written this post. (I only did because I liked the gene-naming analogy.)  That said, if you, unlike me, enjoy this sort of thing, Tom Hickey wrote a long and thoughtful response to it. He mentions among others, Tony Lawson, who I would certainly want to read more of if I were going to write about this stuff.


  1. Vacuonomics II
    Comment on ‘A Quick Point on Models’

    Economists are in the state of manifest self-delusion. They are convinced that what they do is science.

    “Suffice it to say that, in my opinion, what we presently possess by way of so-called pure economic theory is objectively indistinguishable from what the physicist Richard Feynman, in an unflattering sketch of nonsense ‘science,’ called ‘cargo cult science’.” (Clower, 1994, p. 809)

    Keynes provides as good an example as any classical or neoclassical economist. JW Mason writes:

    “According to Keynes the purpose of economics is ‘to provide ourselves with an organised and orderly method of thinking out particular problems’.” (See intro post)

    Let us accept this for the moment and see how it plays out in practice. As a centerpiece of his General Theory Keynes formulated the foundational syllogism of macroeconomics.

    “Income = value of output = consumption + investment. Saving = income - consumption. Therefore saving = investment.” (1973, p. 63)

    The fault in Keynes's two-liner is in the premise income = value of output. It can be formally demonstrated that this equality holds only in the limiting case of zero profit in both the consumption and investment good industry (2014b). Keynes himself felt that something was wrong with profit but he could not identify it.

    “His Collected Writings show that he wrestled to solve the Profit Puzzle up till the semi-final versions of his GT but in the end he gave up and discarded the draft chapter dealing with it.” (Tómasson and Bezemer, 2010, p. 12)

    Keynes was first and foremost a political economists and this means -- as a matter of principle -- that methodological trivia are never allowed to derail the good cause.

    “For Keynes as for Post Keynesians the guiding motto is ‘it is better to be roughly right than precisely wrong!’” (Davidson, 1984, p. 574)

    The real scientist's motto is of course ‘It is better to be precisely right than roughly wrong.’ When von Neumann considered the overall poor performance of economics he came to this conclusion.

    “I think it is the lack of quite sharply defined concepts that the main difficulty lies, and not in any intrinsic difference between the fields of economics and other sciences.” (quoted in Mirowski, 2002, p. 146 fn. 49)

    Indeed, when profit is not correctly defined, income is not correctly defined, and then saving is not correctly defined. It is with profit where the confusion about saving ‘equals’ investment starts. In addition, when income and profit are not correctly defined then distribution theory goes down the tube. And this in turn means that Piketty's study lacks sound conceptual foundations already long before he comes to the definition of capital.

    The representative economist is not much disturbed by all this. He builds one zero profit and I=S model after the other (2014a). Economics has its own methodology.

    “The test of a good model is not whether it corresponds to the true underlying structure of the world, but whether it usefully captures some of the regularities in the concrete phenomena we observe.”

    The sun goes up: this model ‘usefully captures some of the regularities we observe.' Indeed. Another regularity since more than two hundred years is that the representative economist cannot tell the difference between income and profit. In can hardly be denied that this distinction is fundamental for all economic models. Yet, neither in the Keynesian nor the Walrasian context it is carried out correctly.

    The quickest point on economic models is therefore that they do not satisfy the most elementary methodological requirements. Or, in Joan Robinson's unsurpassable tweet format: Scrap the lot and start again.

    Egmont Kakarot-Handtke

    References see

  2. The fault in Keynes's two-liner is in the premise income = value of output. It can be formally demonstrated that this equality holds only in the limiting case of zero profit

    Or, we can treat profits as income. I certainly agree with you that the identity doesn't hold if we limit income to money payments received by households. And I agree with you that this is a serious problem if we naively try to apply Keynes' model to real data. But if we either attribute all business earnings to households, or explicitly include a business sector with positive net income, then the identity still holds, no? The first solution is the orthodox one, I prefer the second one, but either one works, as far as I can tell.

    when profit is not correctly defined, income is not correctly defined, and then saving is not correctly defined.

    I happen to feel that "savings" as conventionally described is not an economically meaningful category -- it combines several different flows that really have nothing to do with each other. Conventional savings groups together net acquisition financial assets, debt service payments, and acquisition of investment goods out of current income by households and businesses, plus the first two flows for governments. I think there are almost always more useful ways to group those different flows. But you seem to be saying something stronger than that. Are you really saying that it is impossible to meaningfully define savings in such a way that it is identically equal to investment? Or are you saying there is only one usage of the word "saving" that is ever permissible, in any context? Neither claim seems very plausible to me.

    I've looked at your papers in the past and I admit I can't get much out of them. I'll try again with the linked paper when I have a chance. But I have to admit, this is why I'd prefer not to talk about methodology (and wish I hadn't written this post). I'm perfectly happy to start from the directly observable facts of experience ("the sun rises in the morning") and build up from there. I'm not sure there are any distinctions that are "fundamental for all economic models." I think what you distinguish, what you aggregate and what you abstract from, all depends what you're trying to do.

  3. The part that is "notcontroversial in the abstract" is actually the core error at the heart of economics. Ironically (given your example), it is biology that demonstrates the problem. 

    When Newtonian physics is used as the archetype of science, Keynes's view does seem correct. But as Tony Lawson has pointed out, physics is an outlier in human empirical endeavors. The sort of "If A then B" regularities that appear in the Newtonian view are not present in the vast majority of our empirical investigations. 

    Biology is the perfect counter-example. What sorts of "If A, then B" regularities did Darwin draw together into evolution? In the infinite variety of life every event is one of a kind. The regularities occur on time scales beyond our observation and at a level of complexity beyond our calculation. The science of biology is about the patterns and regularities of life, but in a radically different sense than that of physics referenced by Keynes. 

    Notice how the analogy you've drawn fails: the "model" in biology is a negative regularity, not positive understanding. Take away one protein and the whole multi-factor process that ultimately results in an insect eye being built does not take place. The actual analogous situation in economics is not IS-LM, but, eg, an analysis of steel production in an economy that has no iron. If A (no iron), then B (no steel). A regularity has been observed. And yet steel production cannot be understood in a positive sense as the sum of mechanical "If A, then B" propositions. (Neither can be genetics.) That is because steel production, like all subjects of economic study, is a human activity, and no one has ever gotten anywhere applying Newton to any living being or collection of beings. 

    See also,

    1. "When Newtonian physics is used as the archetype of science, Keynes's view does seem correct."

      I think you totally misunderstood Keynes (and Mason's) point.

      In the ancient world, various philosophers (e.g. Plato) and various philosophical/religious schools believed that there was a "pure spiritual" world, and that our material world is just a reflection of the pure world.
      In (a large part of) christian philosophy, God is the pure source of all material world, and creates the material world through the Logos, that means rationality and language (and also, in some traditions, sperm, as the idea was that the mother gives the "body" to the kid, while the father gives the "form", including soul). Also breath is often linked to the idea of Logos, soul etc., presumably because it is immaterial and linked to language.

      This train of though, that is very ancient, gave us the philosphical family of "idealism", that broadly speaking means that the material world is just a reflection of some metaphisical rationality somewhere.
      At the opposite of idealism there is "materialism", that means that the world is just a caotic mass of events and stuff, and that rationality is just something we use to understand the world, but is not proper to the world in itself - in short, that all knowledge is just a model that we make in our mind of a fundamentally unknowable and irrational world.

      Idealism and materialism are likely the two main currents of western philosophical thought (e.g. the dispute about nominalism[materialists] and realism[idealism] in middle ages scholastic philosophy), and most philosophers can be seen as attempt at mediation of these two poles.

      Today, we tend to see science as a materialist view of the world, and idealism of a big load of BS.

      However, historically, many scientist had a different, way more idealistic view, and believed that they were discovering "ojective thruths", and in some instance that discovering this Thruth was something that lead them somewhat closer to God.

      This idea of discovering the Thruth with a capitalised T is still present in many people, and might sound somewhat believable if we apply it to "hard sciences" like phisics.

      This "idealistic" view of science is, I think, the approach that you criticise as Newtonian.

      But Keynes and Mason (and Krugman) are saying exactly the opposite, that there aren't Thruths with a capitalised T in economics, so everithing we can know is just a model in our brain of a fundamentally caothic and unknowable world, so it is useless to search for eternal Thruths (a materialist approach).

    2. That you don't understand is demonstrated by your generalizing about all of science. It's understandable, because physicists are like economists in their imperial pretensions. Physicists work today on developing a "theory of everything" that would not explain a single event in my day from dawn to dusk. That's Gary Becket level arrogance and its field wide (and so, more imperious than econ!).

      Two points about idealism:
      1. Krugman and Keynes have not left it nearly so far behind as they think. See, Venlen, Why Econ Is Not Evolutionary Science.
      2. Read up on "emergence" in order to learn about the ontological complexity required to develop an empirical (materialist) understanding of human society. You'll learn that Newtonian regularities are useless for understanding humanity, and thus a deep coal pit in which economics finds itself trapped.

    3. Damn autocorrect. "Becker" and "Veblen".

    4. I scanned the wikipedia page on "emergence", a term that I didn't know:

      Why do you think that the kind of empirism Mason, Krugman and Keynes propose is "Newtonian regularities" and not "emergence"?

  4. It is absolutely crucial that heyerodox economists engage in methodological discussions. If they don't, then the whole discipline has no hope whatsoever.

    It is clear that the problems is econ are unconscious mental habits. Conscious errors would have been noticed from within and corrected by now.

    To uncover these unconscious habits requires that economists submit themselves to the vulnerability of methodology discussions.

    My belief is that this will reveal deep seated assumptions that human interactions can be systematized in a Newtonian fashion. The second unconscious error is a two headed utilitarianism problem: 1) a failure to see how utilitarianism pervades economic thinking, and 2) a failure to understand that utilitarianism ultimately fails to match our strong moral intuitions and also fails as description of actual humsn motivation and behavior.