Eli Dourado

Genetic lotteries and egalitarian redistribution

I wrote a comment on Karl Smith’s post about genetics and egalitarianism, but I thought it might be worth elaborating on the point.

Let’s suppose that a genetic lottery is about to be held. You have a fifty percent chance of ending up as Eli Dourado, gifted economist, chick magnet, high lifetime-earner. You also have a fifty percent chance of ending up as Friedrich Dourado, loyal beagle-German shepherd mix, chick magnet (though without the testicles to partake), low lifetime-earner.

Right before the lottery is run, someone (shall we call him Rawls?) offers you a deal. Rawls has a million dollars to give you, but you must allocate it between your two possible genetic outcomes. If you allocate the money to Eli and you end up as Friedrich, you get nothing, and vice versa. You can also split the million dollars any way you wish between the contingent persons. How do you allocate the money?

In Karl’s calculus, the decision is easy. Friedrich Dourado has much lower lifetime earnings than Eli Dourado, so most of the allocation should go to him. “If we could we would like [to] buy insurance against having been born with bad genes…[P]urely for the sake of economic efficiency it makes sense [for] society at large to insure you against risks for which the free market cannot possibly create insurance.”

There is of course a problem with this argument. Friedrich is a dog. He has very little use for money. As long as he is well stocked with food and chew toys and given plenty of exercise and socialization, he is happy. A million dollars would not change his utility that much. Eli is a human. He has quite a lot of use for a million dollars.

The point is that you do not buy insurance to transfer money from states of the world in which you are richer to states of the world in which you are poorer. You buy insurance to transfer money from states of the world in which you have lower marginal utility of wealth to states in which you have higher marginal utility of wealth. The optimal allocation of the million dollars between Eli and Friedrich is where their marginal utilities of wealth are equal—most of it should go to Eli.

Most people who advocate egalitarian redistribution do not extend it to include dogs (it’s genuinely puzzling why not, but that’s beyond the scope of this post). Let’s take for granted that if you’re in the genetic lottery, you’re going to be human. The same logic nevertheless applies. You want to transfer money to your high MU state. Other things equal, if you’re poor, you have high MU of wealth. But other things are not necessarily equal.

Suppose you are born with “impulsive genes.” This dramatically lowers your earning potential. You have lower lifetime income and this raises your MU of wealth. However, you also have dramatically lower appreciative capital. You never studied in school or read books outside of school. You know nothing about history, foreign languages, architecture, food, etc. What use would you have for, say, a trip to Europe? Or dozens of other expensive uses of money? Your lower appreciative capital decreases your MU of wealth.

The problem of redistributing money so as to mimic an efficient insurance market is much harder than egalitarians think. You cannot assume with certainty that someone who is poor has higher MU of wealth than someone who is rich. An efficient insurance market would probably not transfer wealth from poor people to billionaires, but it is entirely possible that it would transfer money from someone who is relatively poor to someone who is relatively rich at some point on the income spectrum. If you want to be an egalitarian, in other words, you should be in search of firmer ground to stand on than this particular market failure argument.