Welches Modell haben Sie mein Herr?

bzw. meine Dame. Paul Krugman schreibt über Methodik und Markoökonomie. Er wendet sich gegen den Vorwurf, dass "das Versagen der Makroökonomie" auf die Verwendung von Modellen zurückzuführen sei. Ein (gutes) Modell schafft die notwendige Transparenz um die Interaktion (auch etwas komplexerer) kausaler Mechanismen übersichtlich und explizit darstellen zu können. Dies verbessert auch die Möglichkeit die Aussagen (Prognosen) logisch und empirisch zu überprüfen:

First of all, whenever somebody claims to have a deeper understanding of economics (or actually anything) that transcends the insights of simple models, my reaction is that this is self-delusion. Any time you make any kind of causal statement about economics, you are at least implicitly using a model of how the economy works. And when you refuse to be explicit about that model, you almost always end up – whether you know it or not – de facto using models that are much more simplistic than the crossing curves or whatever your intellectual opponents are using.

Think, in particular, of all the Austrians declaring that the economy is too complicated for any simple model – and then confidently declaring that the Fed’s monetary expansion would cause runaway inflation. Whatever they may have imagined, they were in practice using a crude quantity-theory model of the price level. 

And as I have often tried to explain, the experience of the past six years has actually been a great vindication for those who relied on a simple but explicit model, Hicksian IS-LM, which made predictions very much at odds with what a lot of people who didn’t use explicit models were sure would happen. 

Suppose that you didn’t know about IS-LM and the concept of the liquidity trap. You would (and many did) look at the growth of the monetary base, and predict huge inflation:

And you could (and many did) look at government borrowing, and predict soaring interest rates:

But if you understood IS-LM, you realized that both the relationship between money and inflation and the relationship between borrowing and interest rates break down at the zero lower bound; and so they did. 

If you don’t think these successful predictions are a big deal, go back and read the dismissive, vituperative comments those of us who predicted low inflation and interest rates faced back in 2009. 

And a somewhat related point: when people claim to have a sophisticated understanding that transcends models, what, exactly, would they ever regard as evidence that their sophisticated understanding is, you know, wrong?

Sentences to ponder (econometrics)

Mark Thoma sagt, warum Ökonomen erste ökonometrische Evidenz immer nur als das sehen sollen was sie ist, nähmlich erste emirische Evidenz:

There's a version of this in econometrics, i.e. you know the model is correct, you are just having trouble finding evidence for it. It goes as follows. You are testing a theory you came up with, but the data are uncooperative and say you are wrong. But instead of accepting that, you tell yourself "My theory is right, I just haven't found the right econometric specification yet. I need to add variables, remove variables, take a log, add an interaction, square a term, do a different correction for misspecification, try a different sample period, etc., etc., etc." Then, after finally digging out that one specification of the econometric model that confirms your hypothesis, you declare victory, write it up, and send it off (somehow never mentioning the intense specification mining that produced the result). 

Too much econometric work proceeds along these lines. Not quite this blatantly, but that is, in effect, what happens in too many cases. I think it is often best to think of econometric results as the best case the researcher could make for a particular theory rather than a true test of the model.

Dazu fällt mir immer noch ein, dass die Duhem-Quine-These bekanntlich besagt, dass Theorien im Allgemeinen aus einer Reihe verknüpfter Aussagen bestehen, und durch Beobachtungsdaten nicht vollständig bestimmbar sind. Daher lassen sich Theorien nicht durch einzelne Beobachtungen und/oder Experimente verifizieren oder falsifizieren. Es braucht in der Regel mehr als eine empirische Studie. Allerdings gilt das nicht für Aussagensysteme, die logisch inkonsistent sind, keinen praktischen Nutzen haben oder nicht überprüfbar sind.