The ‘nobody wants to do job X’ problem has the usual solution: compensation.

That compensation would come at the expense of research grants. This is not the same world as commercial development.

Some journals do offer compensation - I think the American Economic Review used to offer £100. The solution hasn't been widely adopted, though, perhaps because academics enjoy the kudos of being a reviewer (and the chance to savage other people's work anonymously :-P... ). And while you can pay for reviews, it is hard to pay for review quality, which is very subjective.

Also because a price that would significantly incentivise a senior academic is a price that is way too high for most journals to pay. If you have prestige, a decent if not exorbitant salary, bulletproof job security, a project you're passionate about and want to spend time on, and way too many obligations to actually spend time on your project, how much would you take in exchange for adding yet another obligation to the pile, taking time away from what you really want to spend your time doing?

Maybe that's OK? Going a bit slower but with more solid footing seems like a reasonable tradeoff to me, though I'm a layman.

My husband reviews papers. He works a 40h/wk industry job; he reviews papers on Saturday mornings when I talk to other people or do personal projects, pretty much out of the goodness of his heart. There is no way he would ever have time to download the required third party libraries for the average paper in his field, let alone figure out how to build and run it. We're not talking about a highly in demand reviewer here -- he's an average expert with a PhD.

That system essentially can't work unless 1) you make 1st-year grad students do the review and 2) you somehow make people OK with publication delays being extended by months.