Why every econ paper should come with a warning label!

29 Feb, 2020 at 20:43 | Posted in Economics | 3 Comments

e8f9b7fec248157b1989085deaa05dde-d7bu3k6It should be part of the academic competences of trained economists to be able to be clear about what their models are for; what the models are about; what the models are capable of doing, and what not; how reliable the models are; what sorts of criticisms have been levelled against the models and how the criticisms have been responded; what alternative models there are; etc. The challenge is not easy, and it is clear that it has not been met with sufficient exuberance and success. The capacity of writing “warning labels” would be part of the needed professional competence. Such warning labels would alert the relevant audiences to the capabilities and limitations of the models …

Exceptional amongst the social sciences is the role of the economics discipline in contemporary society, the intellectual and political authority economics enjoys regardless of its failures. Above, I cited Colander’s confession, “we pretend we understand more than we do” and we could add that economists do so in order to – or with the consequence of – protecting and promoting their socially acknowledged authority. In the worst case, there is a nightmarish scenario on which the more economists are consulted for policy advice, the more they need to pretend to know, and so the higher the likelihood of policies going astray. Avoiding the nightmare would require some smart restructuring of the institutions of the economics discipline.

Uskali Mäki

3 Comments

  1. “Perhaps most important, research will be seen as a process leading to reliable and relevant conclusions only very rarely, because of the noise that creeps in at every step.” – Fischer Black, “Noise”

  2. what their models are for; what the models are about; what the models are capable of doing, and what not

    .
    “the models” are almost always analytic, which is to say, theoretic, models. which is to say, a priori
    .
    no extension of geometry will make a map. if you want to know the lay of the land, you have to apply your theory to actually surveying the land. one may find in such practice that the theory as received fails the task of making a good map, a good map being one that usefully abstracts features of the landscape and their relations. if you can use the map to successfully navigate, that is evidence for a good map and a good map is evidence of a good theory.
    .
    but, the geometry — the theory — is not a model of any land. it does not matter how “realistic” are the assumptions; the theorems proven are not statements about how the world is.
    .
    the notion that analytic theories are nomological analogues of social, institutional mechanisms in the world and it isn’t necessary to study the actual and particular mechanisms — that it is thought enough to wave one’s hand out the window — leaves economics without any but stylized facts derived from the pseudo-empiricism of econometrics.
    .
    and, not incidentally, economists fail to appreciate — as Fischer Black appreciated it — the immense noise generated by the actual mechanisms of the economy in motion. so many practical problems of economics come down to the frictional losses and inefficiencies generating that noise, but these can not be studied in theory alone; they must be measured and attributed to the features of mechanism design and management. that can only be accomplished thru a genuine empiricism of observation and measurement.

    • “if you can use the map to successfully navigate, that is evidence for a good map and a good map is evidence of a good theory.”
      .
      But epicycle theory provided good enough navigational and predictive success. Was it therefore better than Aristarchus’s heliocentric model which had no empirical support (because instruments were not sensitive enough to measure parallax motion of the stars)?
      .
      To me, Aristarchus had an intuitive insight that we’ve since validated with data. The Greeks let themselves be fooled into thinking their good map was evidence of a good theory. Instead they should have improved the sensitivity of their instruments …
      .
      “that can only be accomplished thru a genuine empiricism of observation and measurement.”
      .
      Even if you look at basic figures like $23 trillion Treasury debt outstanding compared to less than $2 trillion in bank reserves, you see that basic assumptions are broken, such as that Treasuries can only be purchased with reserves. If you really started looking at individual balance sheets and their interactions, it would quickly become apparent that a lot of fudging is going on. The rules are bent and broken easily enough. Personalities trump assumed economic constraints. Economists are afraid to say this openly. IMHO


Sorry, the comment form is closed at this time.

Blog at WordPress.com.
Entries and comments feeds.