Rent harmonization upzoning

I dunno if any place has ever done this, but I was toddling around Athens today thinking about land use (as one does) and the idea that with a general upzone, wealthy areas will develop first. I’ve seen this claimed along with the “learn how the market works, stupid” riders, and in response, my heart says “learn how politics works, pumpkin. In nearly 30 years as an urban planner and policy analyst, I’ve never—not once—seen a wealthy neighborhood have to suck up anything they didn’t like, with the possible exception of Beverly Hills and the Purple Line, and even then those people did an unbelievable amount of harm to the rest of the region’s transit investment fighting. 

IOW, there are good market reasons to develop in both high and low-cost areas, and very good political reasons to develop in lower cost areas.  Regular readers of the blog, the ones not yelling at me, understand that I hold two positions that make policy hard in California: 
1) We desperately need more housing units; 
2) Building on low-income communities and communities of color at this point is going to be hard on them, and potentially disastrous to renters there. 
It’s one thing to put lower-cost housing near high-cost housing (hard politically, see above, and hard market wise because of land costs), but that’s where you’d like to be able to do so because of amenity and service-sharing. So it bring downs local rents—goodie.  It’s another to put higher income people on top of lower-income people; that gives landlords an additional reason to get to thinking they are sitting on goldmines, etc. 
I’ve generally always maintained that regional average rents are pretty useless. It’s no consolation to know that overall regional rents went down overall if rents in your nabe went up (yes, both those things can happen at the same time.)
So what about if you do look at the various districts in cities (I have no idea how to divide this up, but let’s just play with the concepts for a bit before we firehose it)…..and you find a distribution of regional rents. You find a median. And when you hit locations that are 1 or 2 standard deviations above the mean, there is an automatic upzone, and places 1 deviation and lower get a pass for upzoning, and everybody in that middle interval takes development as it happens? 
Just mucking with ideas here, with jet lag and heat stroke. (Too much sun in Athens today).