EconTech

I bring up social welfare functions relatively often. There is not a whole lot written about them these days; they have serious theoretical issues; they have more serious practical issues; and yet, in my mind, it is one of the most important topics in policy oriented economics. Many have accepted the proposition that the social welfare function must include itself.[1] I see the intuitive appeal of the argument, though I haven’t managed to work through the formal argument in its entirety.

However, I think this post by Tyler brings out the problem I have with the argument quite starkly: His suggestion we should not necessarily pay attention to the wishes of the dead with some justification coming from the fact that they would have racist and/or non-democratic preferences. Now, I’ve said before that I don’t really have a problem with, to some extent, ignoring the wishes of the dead. If they wanted something done so badly, they probably should’ve done it themselves, though there are likely situations where this is not possible. Anyway, my issue is not with the dead. It is with the proposition that the inclusion of their social welfare function in the production of society’s social welfare function would be problematic because we disagree with the values inherent in their function. This is exactly how I feel about plenty of the living! Now, there are formulations that I disagree with, and formulations that I find fundamentally incompatible with how a society ought to be run. I don’t know if the aggregation of welfare functions can accomplish this separation task into areas of legitimate disagreement and absolute abhorrence.

I suppose that my function could output zero for some set of inputs (racist/non-democratic social structures), but in the aggregation, what is there to protect human rights from the tyranny of the majority?

As a note, Tyler apparently got lambasted for this post, here is his response.

Don’t burn it

Tom Stoppard weighs in on the burn it/don’t burn it debate about Nabokov’s unfinished work: “It’s perfectly straightforward: Nabokov wanted it burnt, so burn it.”

That is from Bookslut, but Stoppard is wrong. Dead people don’t count in the social welfare function. (If they did, how many of them would prefer non-democratic or racist outcomes? And would we count that? We shoudn’t and we don’t.)

Don’t destroy the output. Nor is there an incentive problem. If we release Nabokov’s papers as a book, maybe the next Nabokov will burn the manuscript in the first place. We’re no worse off, compared to not releasing such manuscripts. Kafka told Max Brod to burn his works, but we’re all glad Brod didn’t. Think of the current generation as a player in the multiple selves game of the author (he could have burnt it himself long ago) and then the right answer is obvious.

Footnotes:

1. Steven Landsburg. “Methodology of Normative Economics.” University of Rochester [↩]