Who knows best?

Over at the ThinkMarkets blog Gene Callahan asks does Daddy Knows Best? He writes

Does economics have to assume that Daddy (or anyone else) knows what’s best form himself? I’m reading Dan Klein’s Libertarian Essays, in which he asserts “that the individual knows best in matters that concern her alone” is “one of the most basic precepts of economics.”

Callahan is not happy with this analysis, he argues that all that economics needs assume to explain, say, market prices, is that the individual, if left free to do so, will choose based on what she thinks is best for her, regardless of whether or not anyone knows better.

However, there is the point that to say someone knows better, by some objective measure, what is good for us is to abandon the subjective theory of value. You could reply that whether or not our subjective valuations are right does not matter for prices, since even if wrong, they still form the basis for market trading.

But if we don't know what's best for ourselves then should this not lead to "advice markets"? We are able to shop for advisors who can correct any errors of judgment we may make. Or at least, increase the probability of making the right decision. And I have the strongest incentives to correct errors, to search for good advice, since I pay the costs of wrong decisions.

But I'm not sure that the difference between the Callahan and Klein positions has empirical relevance. How could you ever test which of them is right?