the typo of course being “ribeye,” which should read “rib or eye.”

and how much of that acreage is available for sale, what kinds of infrastructure, connections, etc?

this paper doesn’t help you distinguish between lowering market rents and providing affordable housing. affordable housing in this context refers specifically to the kind of housing that by definition falls below the market. the kind it requires subsidy to provide.

if you’re using the data sets in the paper David Vatz has been circulating, you should be transparent that (a) the paper acknowledges the data set is too small to support drawing any conclusions and (b) the authors got a lot of numbers wrong in their analysis of Lawrenceville housing starts.

this is a flawed approach that doesn’t consider the availability of buildable parcels, the time it takes to put together housing development projects, the availability of affordable financing - how are you measuring “land availability” just for starters?

there is no good data to support any of these assertions, and the successes of IZ in the neighborhoods where it’s currently applicable have been huge. If you’re serious about grappling with this issue, would invite you to talk with folks who have actually built and manage affordable housing.

recommend you re-examine this framing. the challenge is not to redeem AI art but to determine if it can or should be redeemed. requires thinking about what the research is intended to deliver.

it’s hard to avoid concluding that creating confusion and disruption was the primary intent.