I haven’t really thought about this yet, but I suspect that, much like Arrow’s theorem, it’s framed as a voting constraint but it actually applies to preference aggregation generally.
Yes, I think so too: the theorem basically says that there’s no voting mechanism that takes ordinal preferences and produces one ordinal preference without being dictatorial (one person can decide the outcome alone), allow for more than two options and be strategy-free.
I think this also has a lesson about the gameability of restricted classes of functions in there.
Gibbard’s theorem shows a case in which Goodhart’s law is unavoidable.
I haven’t really thought about this yet, but I suspect that, much like Arrow’s theorem, it’s framed as a voting constraint but it actually applies to preference aggregation generally.
Yes, I think so too: the theorem basically says that there’s no voting mechanism that takes ordinal preferences and produces one ordinal preference without being dictatorial (one person can decide the outcome alone), allow for more than two options and be strategy-free.
I think this also has a lesson about the gameability of restricted classes of functions in there.