Thursday 9 February 2012

Rationality = Methodological Individualism.

I wasn't planning writing on this, but Robert Viennau came up with a very interesting post on decision theory.

In the comments to his article, Viennau mentioned favorably a piece by Mario Rizzo. Upon reading it, I agree with Viennau: although I have a problem with a few secondary points, the piece is indeed quite thoughtful (readers could find it rewarding reading it).

After arguing the limitations of decision theory, Rizzo reached the conclusion:
"What are we to conclude from this? Are we to say that the behavioralists are right after all? The behavior illustrated here is not rational. It certainly is not rational in the sense meant by those who espouse the standard axiomatic version of rationality. But since it is quite plausibly rational in a broader sense of the term why should the failure to conform to the axioms bother us?  It might bother us if our only task were to explain or predict behavior. The model does not do the job". (Emphasis added).
As it happens, I agree with the conclusion.

So, what's the problem?

The key here is the implications one is to derive from the conclusion: how one answers the question posed (i.e. "Why should the failure to conform to the axioms bother us? It might bother us if our only task were to explain or predict behavior").

In other words, if the narrow neoclassical formal (i.e. axiomatic) model of human individual, micro behavior is deemed to be faulty, what is the alternative?

Most of economics is based on the notion of methodological individualism: the aggregate, macro phenomena are to be explained and predicted from purposeful individual, micro behavior (that is, action, in Misean terminology).

If the current model is inadequate, one answer is to replace it with an informal broader model of human individual, micro behavior. This is what Austrians already do (at least Austrians in the Misean variant), who base their model on the action axiom:
Human action is purposeful behavior. Or we may say: Action is will put into operation and transformed into an agency, is aiming at ends and goals, is the ego's meaningful response to stimuli and to the conditions of its environment, is a person's conscious adjustment to the state of the universe that determines his life. Such paraphrases may clarify the definition given and prevent possible misinterpretations. But the definition itself is adequate and does not need complement of commentary. (See here)
In my opinion, this is a false way out, because it gives the analyst the role of ultimate arbiter of what is rational behavior: the analyst "rationally" determines what the rational behavior is, on an ad hoc basis.

An altogether different answer, that I favor, is the abandonment of methodological individualism, at least in its most extreme forms.

12/02/2012. Although Bill Mitchell doesn't mention methodological individualism, at least one of his posts ("Fiscal austerity – the newest fallacy of composition", 06/07/2010) seems complementary to the views just exposed.

Mitchell starts by claiming that mainstream macroeconomics is largely affected by a form of the fallacy of composition: assuming what is rational for individuals is necessarily rational for the collective of individuals.

Mitchell explains that initial assessment:

"The origins of this logical error lie in the way in which mainstream economics developed. It was largely concerned with microeconomics and started its a priori reasoning from the perspective of an atomistic individual. The single consumer or single firm. I won’t go into detail here but this body of theory soon got into trouble via the so-called Aggregation Problem.
"So to make statements about industry or markets or the economy as a whole, the mainstream had to aggregate their atomistic analysis. Of course this proved to be impossible using any reasonable basis and so they fudged the task and assumed things like the 'representative household' to be the demand side of a product market and the 'representative firm' to be the supply side."
A few comments: what Mitchell calls "a priori reasoning from the perspective of an atomistic individual" is the deduction of individual behavior, from rational individual behavioral assumptions (in neoclassical microeconomics) and what Mises called action axiom (in Misean Austrian economics).

From this individual behavior and according to the principle of methodological individualism, the collective, aggregate, behavior is to be deduced. When a collective, macro behavior can be traced back to individual behavior, it is said to have microeconomic foundations.

The difference between Mitchell's views, as expressed in the quote above, and the views expressed by me here is that Mitchell emphasizes several macroeconomic problems inherent in the methodological individualism programme (what he refers to as the Aggregation Problem) and leaves out of consideration (at least in the text mentioned) the intrinsic difficulty in defining rationality. Therefore, with the caveat expressed that I am scrutinizing this particular text, the methodological individualism/microfoundations paradigm is not necessarily ruled out.

I, at the other hand, implicitly place greater emphasis in the latter aspect (i.e. rationality definition), and leave out the macroeconomic aspects (i.e. Aggregation Problem) altogether, because I consider that macroeconomics can hardly be based on detailed microeconomic behavior, when we are unable to express what rationality is.


  1. Methodological individualism can occasionally be a useful approach, but as a general rule, for economics or politics, it is simply nonsensical. It's a bit like "methodological atomism" applied to biology.

    How can it make sense to (try to) explain why railway gauges, road quality, field layouts, etc. etc. change at borders in terms of individual decisions, rather than the properties and histories of states?

  2. Although I can't say that I have gone into other forms of atomism, in what concerns economic/social sciences methodological individualism seems to me an over ambitious position.

    If you think about it, the 19th century saw the birth of statistical mechanics/thermodynamics, where the basic Newtonian laws of mechanics, with an statistical treatment, were found by Boltzmann to explain the empirical regularities known as the gas laws (Boyle-Mariotte, Charles, Gay-Lussac, etc.).

    This is the explanation of macro behavior, empirically observed, in terms of micro behavior. And it's undeniably a great achievement.

    The whole methodological individualism, marginalism, micro foundations thing, seems like an attempt to replicate in economics and the other social sciences the same research agenda.

    (Incidentally, this attempt would explain the adoption of the same or similar mathematical tools: if they were useful in physics, surely they should be useful in economics?).

    The idea of replacing a formal model of human behavior with an informal model, in this context, would seem at least partially justified: we get rid of part of the problem (notice that we are still working within the methodological individualism paradigm).

    The problem with this is that the use of mathematics (or more precisely formalism), is not necessarily erroneous: it allows for clarity.

    So, the problem is not the use of formalism per se, but the attempt to use a formal micro model to explain macro behavior.

    Frankly, I haven't gone through the whole relevant philosophical discussion, but I will make just one observation: it was the determination of the laws of gases that provided a goal for statistical mechanics. Do we have anything close to the gas laws in economics and the social laws?

    Curiously, Ludwig Boltzmann, father of statistical mechanics, was an Austrian.