3 Comments

I agree that honest evaluations of experts by other experts would be useful to customers, but it is very hard to create incentives to create and publish such things. There's too much of a temptation to instead be paid to say what experts and customers want to hear.

Expand full comment

Also, I'd add that in many markets it's totally reasonable to infer that high price is an indicator of quality. I mean I know that when I buy a recently produced video game I can use that as a quite reliable indicator of quality.

Of course, the market needs to have sufficiently many people making choices based on fundamentals for that to work. It needs to be the case that enough people would stop buying from them if they were charging more than what their services were worth but whether or not a market has that property isn't something most people can perceive.

Expand full comment

Well, first I should say that I do think what you are saying is substantially correct in that often people are really minimizing post-hoc regret (e.g. seeing data after the fact that makes them realize they choose wrongly) than maximizing outcome but I think there are other factors at play here too..

Regarding the guides I still think that part of the issue is that guides are more likely to be aimed at exactly those situations in which clear, easily interpretable outcome data isn't available.

I mean isn't it true that the very occupations for which guides exist (doctor, lawyer etc..) are often the same occupations that seem to have some kind of prohibition (perhaps informal) on making clear signals of quality available.

I think if you could find a ranking of surgeon quality by other surgeons or of lawyer ability by other lawyers that would be consulted by many people but various factors discourage those from being available. Even surgery success rates aren't something most people are able to access in most cases and when they are available maybe those same norms discourage other practitioners for giving out the info that would turn that data into a quality measure (e.g. really all surgeries of X are equally hard or whatever).

Expand full comment