1) AERA is too big and too long.
2) Because there are too many people on the program, the quality of the average paper is low.
3) Because the quality of the average paper is low, many scholars who do high quality work don't submit their work to AERA.
4) Because there are too many concurrent sessions, most sessions are sparsely attended, so the feedback quality is low unless you have an unusually insightful discussant.
5) As a result of the above, AERA continues to suck.
Can AERA become something more than a convenient punching bag that we pull out each spring? I'm pessimistic about the potential for fundamentally reshaping this meeting, but perhaps you aren't, or have ideas.
For more punching, see the Hessinator here and here.