Obviously, for the second scenario to be plausible, we have to believe that lots of people labeled "expert" are not actually experts.
To a certain extent yes, but from my understanding a lot of it is that the education industry got double helpings of rent-seeking and goodheart's law. For example in many districts unions make it effectively impossible to fire teachers once they have tenure(which is relatively easy to obtain at the k-12 level).
-edit also http://lesswrong.com/lw/le/lost_purposes/
I've wondered for some time now what the effects of online education might be on gender and income inequality, specifically as online education interacts with IQ and Conscientiousness (compared with offline education). I ran into a study of a course done online and offline that found correlations with Conscientiousness, which prompted me to start writing out my thoughts: https://plus.google.com/103530621949492999968/posts/aKa3qLatwZ3
The model/argument I give (towards the bottom) is logically trivial, and the basic idea seems pretty intuitive - offline classrooms remove some need for self-discipline/Conscientiousness and performance is more g-loaded - that I'm sure I can't be the first person to think of it.
Does anyone have statistics or citations handy which might help in any essay I write on the topic?