I hear this a lot but don’t fully agree to the premise of ‘having no choice’. I published an article once and asked the professor if he was willing to publish open access. His answer was that he was open to it, but that if I was looking for a career in science, publishing OA in the current journals would require more effort from our side to justify the importance of our work. (The other author wanted to stay in science, and wanted to take the easier route, so in the end we decided to publish the old-fashioned way)

Just to say: you have a choice, the OA path just requires more effort. And depending on your area of expertise, the OA journals can actually be very good.

I agree with your conclusion though: change at scale requires universities to change their behavior. Before boycotting Elsevier, a good start would be to move evaluations away from ‘impact factors’ to e.g. h-index.