welfare, consequentialism, morality, rationality
A governmental decision is "ex ante efficient" if it maximizes the satisfaction of everyone's preferences ex ante, relative to other possible decisions. Equivalently, each affected person would be rational to approve the decision, given her preferences and beliefs at the time of the choice. Does this matter, morally speaking? Do governmental officials - legislators, judges, regulators - have a moral reason to make decisions that are ex ante efficient? The economist's answer is "yes." "Ex ante efficiency" is widely seen by welfare economists to have moral significance, and often appears within law-and-economics scholarship as a criterion for evaluating legal doctrines.
The intuitive attractiveness of ex ante efficiency is grounded, I suggest, in a broader intuition: namely that "rational approvability" has moral weight. Imagine that person P would be rational to approve some particular governmental decision A*, given P's preferences and beliefs at the time of governmental choice. Then it seems plausible that the governmental decisionmaker has a prima facie moral reason to choose A*. Considerations of welfare seemingly come into play here. A governmental decision that P is rational to approve is (seemingly) a decision that enhances her welfare, at least given a preference-based account of welfare and rationality. Consent, too, seems relevant. To say that a governmental decision is rationally approvable by P is just another way of saying that she would consent to it. And the moral force of ex ante efficiency follows quite immediately from the moral force of rational approvability: an ex ante efficient decision is one that everyone rationally approves.
In this Article, I seek to deflate the intuitive case for ex ante efficiency just described. The fact that P rationally should approve some official decision, given her preferences and beliefs at the time of choice, has no moral significance whatsoever for a governmental decisionmaker. Neither welfare, nor consent, nor the fact that governmental decisions are made under conditions of uncertainty (producing a divergence between what's best ex ante and what's best ex post), nor the need for stable governmental decisions, nor other considerations, actually lends moral weight to rational approvability. Choices rationally approvable by P might, unluckily, lead to outcomes that are worse for her welfare. Hypothetical consent is not the same as actual consent; further, the rational approvability by P of some choice implies that she would hypothetically consent to the choice given her current beliefs, not that she would hypothetically consent under full information or under Rawlsian conditions of morally structured ignorance. And the probabilistic beliefs that prompt P's rational approvals can readily deviate from the probabilistic beliefs of the governmental decisionmaker; it is those beliefs or the objective probabilities, not P's beliefs, that determine what the decisionmaker morally ought to do ex ante.
Matthew D. Adler, The Puzzle of Ex Ante Efficiency: Does Rational Approvability have Moral Weight?, 151 University of Pennsylvania Law Review 1255-1290 (2003)