Having done financial modeling for a living at one point in my life, I'll chime in briefly. A developer's model backs into the highest dollar amount that they're willing to pay for the land. If a developer decides to take a haircut on a development in order to provide a public good, then they won't be able to pay as much for the land. They'd get outbid by a profit-maximizing developer. Also, people that specialize in multifamily typically are hesitant to venture into retail. Its a totally different business model with its own market dynamics and cost structure. If the retail rents that can be achieved are sufficiently high, then the extra effort and uncertainty may be worth it. The problem is that there are only a very few cases in Houston where the risk-adjusted retail rents exceed apartment rents. Which brings me to my next point: It isn't enough to point out that restaurants at the Post development are busy or that Phoenicia is busy. Many other stores are not busy; that consumers return over and over again to busy stores and witness other consumers doing exactly the same thing probably has a lot more to do with those stores' business models and good management than with their physical plant, but that's easy to forget if you don't go to the places that are suffering for business, which you aren't because otherwise they wouldn't be suffering for business...and besides which, don't tend to last very long anyway. So consumers see other people at the places where consumers are at and figure that everywhere should and could be like that. But the fact is that there are only so many retail prospects out there, and so many fewer still that have viable business models or good management. When rents are being negotiated, landlords are price-takers. They can't necessarily tell whether a store will be successful (although a place like Phoenicia is probably an exception, and in cases like that the TENANT typically holds the cards and the rent is much lower), but whether destined for success or not, the prospective tenant can go down the street and find a landlord that will undercut the other one. Its a competitive market. If the rents aren't obviously high enough or the demand isn't obviously there, then nine times out of ten a multifamily operator isn't going to make the effort to take the risk. The multifamily operator has no doubt that there will be demand for their units, with or without a retail component, and they know that even if the market declines prior to delivery, they can give concessions and fill the units quickly to generate cash flow, cover the note, and make the investment marketable to a third-party buyer. If that retail component sits empty for three or four years before finding an awesome tenant, that's a goddamn long time and there's no cash coming in. One last thing. Having retail at the ground floor can be disruptive to parking designs. It depends a lot on the layout of the site and what the architect can do with it, but any option that requires more concrete to yield less net rentable area skews the model against that option. Mixing uses still requires on-site segregation of those uses for resident convenience and security. So yeah, if its a good model then the model won't break. It'll just indicate a lower land price that can be paid as a maximum bid, and the mixed-use guy gets outbid. There are exceptions, but not many in Houston. Niche, out.