Payback for referees

There is a recent letter at Nature saying

I have discovered a negative correlation between the number of papers that a scientist publishes per year and the number of times that that scientist is willing to accept manuscripts for review  … I therefore suggest that journals should ask senior authors to provide evidence of their contribution to peer review as a condition for considering their manuscripts.

While I agree with the overall observation, I disagree with any conclusion. It may not be unexpected if an editor of a dying old-style journal cannot find reviewers anymore.
Why not just paying referees? Journal production is a commercial business. So pay me for any review. As I already get paid, I would give the extra income back to my employer.
If there are no commercial interests involved in some open source publications, why not leaving the decision to potential referees what they want to review? Most papers are published mainly for career interests – it will be an enormous waste of time to review them all.