Saturday, October 16, 2021

On David Card's Nobel

The Sveriges Riksbank prize in economic sciences in memory of Alfred Nobel this year goes to David Card, Joshua Angrist, and Guido Imbens. I cannot say much about instrumental variables, Angrist, or Imbens. Since I have been pointing to Card's work with Alan Krueger on minimum wages for decades, I thought I might say somthing about his half of the prize.

I do not have much new to say. I find both natural experiments and meta-analysis intriguing.

Both Card and Krueger's natural experiments with minimum wages and their meta-analysis have been superceded. Maybe 'transcended' or 'replicated' would be better terminology. That is why, in my 2019 paper in Strucutral Change and Economic Dynamics, I reference Andrajit Dube and his colleagues, not Card and Krueger. Also, David Neumark's quibbles with Card are currently uninteresting. (Any reporter talking to Neumark should note he started out with funding from a consortium of fast food joints.)

I object to attempts to explain the lack of impact of minimum wages on employment by the theory of monopsony. Economists have known, for over half a century, that wages and employment cannot, even under ideal conditions, be explained by the interaction of well-behaved supply and demand curves in the labor market. In marginalist theory, the supply of labor is derived from utility-maximizing households trading off leisure and commodities to consume. The demand for labor is supposed to be derived from profit-maximizing firms. But no such valid derivation goes through if firms produce some commodities with the use of previously produced commodities, that is, capital goods. This well-established result is widely ignored, with no pretence at justification.

4 comments:

Anonymous said...

I think there are like two possibilities for an array being thrown from the Card arch. One is to keep within the neoclassical framework from Steedman and Opocher and relate Card's achievements with the phenomena of recurrence. To be more clear, accepting the framework of well-behaved functions. So there can not be a whole-economy effect of rising employment.

The other is to see it through the optics of non-differential functions and maybe to extend the analysis to the opening of the numeràire or more rightly to the opening of both rates of wages and profits on a non-uniform level so the pattern spaces for global rising wages following rising employment can gain some area on the plot.

Blissex said...

«extend the analysis to the opening of the numeràire or more rightly to the opening of both rates of wages and profits on a non-uniform level»

But but... to abandon the J.B. Clark or Arrow-Debreu-Lucas models is COMMUNISM! :-).

Blissex said...

«"extend the analysis to the opening of the numeràire or more rightly to the opening of both rates of wages and profits on a non-uniform level"
But but... to abandon the J.B. Clark or Arrow-Debreu-Lucas models is COMMUNISM! :-).»

The point I am trying to make here humorously is that “opening of the numeràire” or “opening of both rates of wages and profits” are technical fixes as if the people who use Clark-Arrow-Debreu-Lucas models do so because they cannot think of better technical approaches; of course not, they are clever, capable people.

But it seems to me that they have chosen "internal consistency" with the assumptions of Clark-Arrow-Debreu-Lucas rather than "external consistency" with the actually-existing political economy, and they have done so for ideological and career reasons. So suggesting different technical approaches is futile.

Blissex said...

«no such valid derivation goes through if firms produce some commodities with the use of previously produced commodities, that is, capital goods»

To be pedantic, but here it is of the essence, for the sake of posterity :-)

* If there is a single capital good that can be endlessly subdivided and recycled ("leets", "putty") then such a derivation is possible (but it is still not valid because it depends on other assumptions that are contradictory).

* Therefore "capital" in J.B. Clark and Arrow-Debreu-Lucas model is fantasized as a single *metaphysical* commodity, "capitalness", that turns other commodities into capital goods.

* The fundamental reason why two or more distinct capital goods cause "anomalies" is that introduces multiple non-equivalent potential (essentially timewise) paths in the models, triggered by "interest rates".