Developers can profit by building affordable housing almost anywhere
citylab.com
citylab.com
I am pretty sure you only have to build enough houses for three hundred million people.
If the lot will only support X thousand square feet of heated living space, you can build 1 X kft^2, 2 X/2, 3 X/3, or 4 X/4 units. It seems overwhelmingly likely that the last will be more affordable than the first.
You can have a shortage of below-market-rate housing by not having enough variance.
Also, they didn't do a side-by-side comparison of building affordable units and not. How much does the profit change?
It's important to remember that just because you turn a profit on an investment, doesn't mean you should make it. You need to look at what else you could invest in.
If constructing a building in SF with affordable units turns a profit of 10%, but constructing the same thing without affordable units in, I don't know, Houston, gives you 20%, where do you think the investments will flow?
Unlike VC, a project that goes south has a massive impact on a portfolio. There are not 1000:1 real estate projects to make up for losers.