Many hiring decisions in academia still rely strongly on the reputation of journals where applicants published. This may be less accentuated in computer science, where most research is presented at conferences and preprints are often available on the arXiv, but it's a very important factor in many other disciplines. It's also the primary reason why researchers don't boycott these journals: most of them are simply not in a position where they can afford to without seriously damaging their career prospects.
See also https://en.wikipedia.org/wiki/Price_of_anarchy - a system can get stuck in a terrible state where everyone's best move is to keep going. A much better equilibrium could be constructed (Price of Stability), but how do we get there?
See also https://en.wikipedia.org/wiki/Price_of_anarchy - a system can get stuck in a terrible state where everyone's best move is to keep going. A much better equilibrium could be constructed (Price of Stability), but how do we get there?