Home > Uncategorized > Econometrics and the difficult art of making it count

Econometrics and the difficult art of making it count

from Lars Syll

Modern econometrics is fundamentally based on assuming – usually without any explicit justification – that we can gain causal knowledge by considering independent variables that may have an impact on the variation of a dependent variable. This is however, far from self-evident. Often the fundamental causes are constant forces that are not amenable to the kind of analysis econometrics supplies us with. As Stanley Lieberson has it in his modern classic Making It Count

LiebersonOne can always say whether, in a given empirical context, a given variable or theory accounts for more variation than another. But it is almost certain that the variation observed is not universal over time and place. Hence the use of such a criterion first requires a conclusion about the variation over time and place in the dependent variable. If such an analysis is not forthcoming, the theoretical conclusion is undermined by the absence of information …

Moreover, it is questionable whether one can draw much of a conclusion about causal forces from simple analysis of the observed variation … To wit, it is vital that one have an understanding, or at least a working hypothesis, about what is causing the event per se; variation in the magnitude of the event will not provide the answer to that question.

Causality in social sciences – and economics – can never solely be a question of statistical inference. Causality entails more than predictability, and to really in depth explain social phenomena requires theory. Analysis of variation – the foundation of all econometrics – can never in itself reveal how these variations are brought about. First when we are able to tie actions, processes or structures to the statistical relations detected, can we say that we are getting at relevant explanations of causation. Too much in love with axiomatic-deductive modeling, neoclassical economists especially tend to forget that accounting for causation – how causes bring about their effects – demands deep subject-matter knowledge and acquaintance with the intricate fabrics and contexts. As already Keynes argued in his A Treatise on Probability, statistics and econometrics should not primarily be seen as means of inferring causality from observational data, but rather as description of patterns of associations and correlations that we may use as suggestions of possible causal realations.

  1. December 6, 2012 at 11:51 pm

    Correlations between A and B might be due to a cause-and-effect relationships or they might be due to a third cause C. Without a realistic narrative and body of theory, statistical analysis is of limited value or can lead to false conclusions being drawn.

    • sergio
      December 10, 2012 at 2:07 pm

      Such empirical-modeling “logic” can be best explained as “trees make the wind blow”. Such model includes what can be observed, measured, calculated and for which there is a data. Everything else, even if it is the cause are considered as non-economics.

  2. Bruce E. Woych
    December 7, 2012 at 2:04 am

    There is a very relative discussion here:


    There is a very relevant presentation for causality integrating the social sciences here:

  3. Bruce E. Woych
    December 7, 2012 at 3:39 am

    Note: in the comment stream from your alternate article:( https://rwer.wordpress.com/2012/12/04/put-mindless-econometrics-and-regression-analysis-where-they-belong-in-the-garbage-can/ )…Asad Zaman poste a link to his own work that also appears appropriately relevant to the issue: If I might quote directly from his post:

    Econometric Methodology is based on logical positivist principles. Since logical positivism has collapsed, it is necessary to re-think these foundations. We show that positivist methodology has led econometricians to a meaningless search for patterns in the data. An alternative methodology which relates observed patterns to real causal structures is proposed.”

    The full article can be downloaded from the journal website:

  4. December 7, 2012 at 7:25 am

    Bruce, thanks for the links, especially to the Zaman paper. Looks interesting. I’ll hake a look at it :)

  5. December 7, 2012 at 7:31 pm

    Asad’s paper is indeed a valuable contribution to the discussion of econometric methodology, not least by how well he has been able to articulate the different issues. There is only one area where I disagree with him, which is where he looks behind logical positivism only as far as Kant, and fails to discover that Kant was “stirred from his slumbers” not by traditional thinking but by the then very new arguments of David Hume, which are almost identical to those of logical positivism. I realised that way back in 1958, and around 2000 found Critical Realists Tony Lawson and Roy Bhaskar saying much the same thing. I have to hand a little book by A D Lindsay on “The Philosophy of Kant” dated 1919 (somewhat before the publication of Russell’s lectures on Logical Atomism in 1924) from which a couple of brief extracts show how Asad is mistaken not about Logical Positivism but about blaming Kant.

    “Hume’s answer [to Locke’s difficulties with universals] is ingenious. The principle of causation cannot be rationally justified, and the necessary connection we predicate of changes in the outside world is not in the things; it is only a feeling in ourselves, and is the result of custom. After seeing the same succession several times, we come somhow to feel different about it, and that feelin of difference we express by saying that we have before us an instance not of simple succession, but of cause and effect. This is not the place to discuss the difficulties in Hume’s position; it is enough to notice how enirely passive it makes the mind, and how alien such an exlanation is from the spirit of inquiry and discovery”. [p.25].

    “This discovery [that reason is about understanding, not observing reality], and its main purport, is expressed in the distinction Kant so often makes between things in themselves and phenomena. … The doctrine it implies is very easy to misunderstand, partly because idealism is used in a very different sense from that in which Kant uses it, partly because Kant’s statement of the distinction between things in themselves and phenomena depended on a view of knowledge which he was very much concerned to refute, but with which we are not now familiar” [Cartesianism and Representationalism; p.79].

    “[I]t was Hume’s criticism of causation which first led Kant to formulate the critical problem.” [p.79]

    I hope to deal with this in the “Framework” discussion, along the lines that in computers the equivalent of Kantian concepts provide an index to the type of program necessary to focus the senses so they can see the data. Hume is thus answered by the brain acting somewhat like a closed circuit TV.

  6. December 9, 2012 at 6:04 pm

    How many people would play Golf if the only acceptable outcome of teeing off would be that the ball roll into the cup, where anything short of that is an absolute failure?

    It seems to me that the success of Economics and Econometric’s is to get “on the green” where choices are more clear and where hard work and talent can work.

  7. Jeff Z
    December 10, 2012 at 5:29 am

    True, but we need to know the categories and the social context. The banking system can not be the source of an economic collapse in a society that does not have money or banks. Defining these in a way that is both operational and acceptable to most practitioners is difficult, but necessary work.

    To take tcoss’ golf analogy, the criticisms are more like this. We accept the idea of the green as the appropriate target, but for many here the focus of econometrics seems to be on the small details and refinements in the model, not the social importance of the phenomena being investigated. The social importance is the actual green, where the practice of econometrics seems akin to a wicked slice off the tee, and then magically moving the green to where the ball landed and declaring the model (golf shot) a success.

    For examples, see other comments in another discussion on econometrics.



  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