How good are they, especially with dealing with the unknown unknowns to quote a famous contemporary American philosopher (/s)? Even for the known unknowns, they

How good are they, especially with dealing with the unknown unknowns to quote a famous contemporary American philosopher (/s)? Even for the known unknowns, they have to make assumptions to start somewhere. Yes, they create a lot of models for all kinds of stuff, and I'm sure all those models work out well. I must say though that I have doubts about the parent's "they largely ignore externalities", I think that is too negative a claim.