My best attempt: without journals charging fees for access, there is a free rider problem where only a few institutions will support journals. By allowing Elsev

My best attempt: without journals charging fees for access, there is a free rider problem where only a few institutions will support journals. By allowing Elsevier to charge, you force everyone to share in the costs. In a weird sort of way, the fact that universities haven’t yet put Elsevier out of business is evidence for the collective action problem. I don’t really think it works, and the publication fees model might already be a good enough counter-argument, but it’s my best try.