Scott Clark, an Econlog reader, responds to my critique of analytical egalitarianism:

The way I viewed analytical egalitarianism when i was in Levy’s class was not that everyone is the same as everyone else. But when analyzing and making policy decisions, it best to act as if everyone were equal.

This sounds good. But here’s the problem: Why would it ever be best to act on an assumption that you know to be false? The most convincing response: Because the assumption is a very close approximation to the truth, so the marginal benefits of increased accuracy are less than the marginal costs of increased complexity.

By this standard, analytical egalitarianism still fails, because it’s assumptions are grossly wrong. People are very far from identical in talents and preferences; a lot of these differences are due to genetic differences; and saying this does not remotely put us on a slippery slope to genocide.

But isn’t it often “best to act as if” something very false is actually true? Individually speaking, no. If you’re weighing whether to take a walk alone at night, it is foolish to act on the assumption that you WILL get mugged. If you really acted on that assumption, you’d stay home! If there is a .1% chance of getting mugged, the optimal response is take actions and precautions appropriate for a danger with this .1% probability – perhaps going on the walk but carrying a big stick to deter predators.

Now socially speaking, it may be “best to act as if” something very false were actually true. If you have a Prisoners’ Dilemma, it is socially better if everyone believes that defection will make the defector’s head explode. Then no one defects, and everyone enjoys the cooperative payoffs.

Is analytical egalitarianism the kind of belief that helps society even though it is false? Hardly. It’s a mixed bag, at best. Analytical egalitarianism weighs against coercive eugenics, but tilts the scales in favor of social engineering (if all differences are environmental, why not radically change the environment?). Analytical egalitarianism weighs against apartheid, but tilts the scales in favor of anti-discrimination witch-hunts (if all groups are equally talented, then don’t unequal incomes prove the existence of widespread discrimination?). Analytical egalitarianism weighs against dictatorship by experts, but weighs in favor of dictatorship of the majority (if everyone is equally knowledgeable, then why not follow the most popular advice?).

In each case, analytical egalitarianism reduces the risk of one set of bad policies, and increases the risk of another important set of bad policies. We can imagine cases where bending the truth would benefit the world, but this isn’t one of them.