I’ve never seen the slightest relationship between the charge to read a paper and the quality of review.
Yelp, TripAdvisor, wire cutter, hell even Google results themselves.
Once you start poisoning that well, it's difficult if not impossible to claw it back.
But who watches the watchers? I guess review fraud will need to be considered as well.
Not really. There would be perverse incentives where the publisher benefits from accepting more articles. For good journals that would be a conflict of interests at best where they would optimise the marketing-to-acceptance ratio. I can’t believe I am writing something good about scientific publisher, but at least when the reader pays they are incentivise to publish things that have an audience. Otherwise, they are going to cut corners, and I mean more than they currently do. And it’s not hypothetical, there are already terrible publishers doing this.
Sponsored reviews of products are basically this. If you are paying a reviewer for a stamp of approval and the reviewer sets the bar too high, why would you want to pay that reviewer? On the other end of the reviewer, it's easy to get more money by providing that stamp of approval to more people--not fewer--so they're incentivized to make it fairly easy to achieve.