Sunday, December 2, 2012

Is Deirdre McCloskey trying to be ironic?

She writes: "But treating others as “inputs into a self’s utility function,” as Becker puts it, is to treat the others as means, not as ends. Immanuel Kant said two centuries ago in effect that your mother, if she is truly and fully loving, loves you as an end, for your own sweet sake. You may be a rotten kid, an ax-murderer on death row in Texas. You’re not even a high-school graduate. You give her “nothing but grief,” as we say. In all the indirect, derivative ways you are a catastrophe. And yet she goes on loving you, and stands wailing in front of the prison on the night of your execution. Economists need to understand what everyone else already understands, and what the economists themselves understood before they went to graduate school, that such love is of course commonplace. It is common in your own blessed mother, and everywhere in most mothers and fathers and children and friends."

I don't know if the bolded sentence is meant to be ironic or not.

Becker has an idea that's actually called "rotten kid theorem" that discusses the impact on child behavior of what he assumes to be a commonplace occurrence: that parents love their rotten kids unconditionally.

The point of the theorem is to demonstrate that that can change rotten kids' behavior for the better. But the assumption about parents is McCloskey's assumption. Parents want to be good to their children.

In full disclosure, the theorem is found only to apply under certain conditions. But that doesn't really matter. Unconditional love is central to Becker-style models of the family, as well as the more sophisticated models that follwed Becker. They even talk about different types of unconditional love: deferential, non-deferential, etc.

I like McCloskey a lot. I hope this doesn't confuse that point. But I think she often goes overboard in critiques of "MaxU".

If you interpret things in utility functions as being about means, not ends, maybe it's better to just drop the utility function name altogether. Use the exact same math and just refer to it as the "objective function" (which is its more general name anyway). An objective - i.e., an end - of parents, according to Becker, is unconditionally loving their children. They try to do as much unconditional loving as they can. They face constraints and trade-offs of course (when it comes to the economics of caring a big constraint is the time constraint... you want to spend time with them but you have to be able to feed them - and yourself! - too).

It's the exact same economics. If McCloskey has issues with the utility language perhaps we should just say "Max Objective" rather than "Max U".

To me, it's a distinction without a difference - but maybe she'd like that better.


  1. Her point is that utility (or objective) maximization can only inform you how to act prudentially. And there is something lost when speaking about human motivation when you reduce everything to prudence. Is that a distinction without a difference? I don't know.

    1. I've never been sure I buy the MaxU is all prudence claim.

      Take a central banker's objective function. Surely my non-prudential sympathies with the unemployed, etc., inform what goes into that. I have deferential altruistic utility w.r.t. Kate because I love her. It's the only reason why that's in my objective function. Faith informs my expectations which feed into the constraints that I'm optimizing against. My faith in the architecture defending justice in this society similarly informs my sense of the strength of property rights and thus the constraints I'm optimizing against.

      It's hard for me to understand the way economists view altruism WITH talking about love and faith. Of course if it were all prudence I would have concerns too. But I don't think it is.

    2. You know you could take the way it's talked about the first time someone encounters optimization in a textbook and it sounds awfully prudential. But no one really thinks that's anything other than a simple way of spoon-feeding the logic.

      We treat undergrads that complain about the lack of realism in intro micro as sophomoric, not insightful, right? "Don't worry", we reassure them "I'm just giving you the basic outline so you can get acquainted with the way we talk about this sort of thing".

    3. MaxU is making the best you can out of competing ends given your scarce resources. Sounds a lot like prudence to me! The question is really whether rational choice theory can really explain everything.

      If your talk of the other virtues is to mean that you reject De Gustibus, then you are closer in agreement with her than Becker. Becker implies much more radical things than people realize. Can you really reduce love and faith to information and something that can be thought of mathematically as "capital"? Can you explain all changes in behavior to what amount to changes in cost curves?


All anonymous comments will be deleted. Consistent pseudonyms are fine.