Now Playing

How Supervisor Kim’s New Affordability Law Lets Developers Off the Hook

Fuzzy language and fuzzier math helped ensure landmark affordability.



Last month, in an act of legislative panache, Supervisor Jane Kim introduced a ballot measure mandating that developers produce a minimum of 25 percent affordable housing in medium and large projects—essentially double what they’re now required to do. 

In a city in which rents and real estate prices are climbing assuredly to infinity and beyond, this was a potential masterstroke: This measure is polling well with voters. Kim notably led negotiations on the high-profile San Francisco Giants waterfront development and the Hearst Corporation and Forest City’s 5M project—extracting 40 percent affordability pledges from both. “Forty is the new 30” is a hell of a catchphrase, and Kim, who is running against Supervisor Scott Wiener to succeed state senator Mark Leno, can do well by doing good here.

Her audacious ask of 25 percent affordability is, deservedly, getting plenty of attention thus far (and, behind the scenes, is providing a lot of negotiating leverage with the mayor, who proposed a similar but less stringent ballot measure). But if you peruse the fine print of Kim’s proposal, you’ll find something counterintuitive: It exempts projects in which “a height limit increase has been approved by a vote of the electors...prior to January 12, 2016,” or “has entered into a development agreement or other similar binding agreement with the City as of January 12, 2016.” That would grandfather out the Giants’ project and Hearst’s 5M project. 

Why would two projects lauded as achieving 40 percent affordability be specifically exempted from a subsequent requirement to merely achieve 25 percent affordability?

Perhaps because, when you crunch the numbers, it’s uncertain either would qualify. 

San Francisco land-use politics
is an aneurysm-inducing affair. And, you know, math is hard. But the equation the city uses to calculate how much affordable housing a project creates is straightforward. You divide the number of affordable units by the number of total units and come up with a percentage. Simple. If you have 29 affordable units in a 100-unit project, then you’ve hit 29 percent affordability.

That’s how it’s supposed to work. In the world of San Francisco development agreements, strange things happen. Math gets…fuzzier. The development agreement for the 5M project calls for “the creation of affordable housing units anticipated to equal to forty percent of the total market rate units for the project...” (emphasis ours). 

Very cute: Instead of affordable units composing 40 percent of total units—as the city and any fourth grader would calculate it—5M is promising 40 percent as many affordable units as market-rate units. That’s not the same thing: In our above hypothetical, instead of dividing 29 affordable units by 100 total units and reaching 29 percent, you’d be dividing 29 affordable units by 71 market-rate units—and, jarringly, claiming 40-plus percent affordability. 

And so, when the 5M promotional materials crow about achieving 40 percent affordable housing, they got there by dividing the 241 built or funded affordable units by 601 market-rate units. But if you divide the 241 affordable units by total units on-site (688) you come up with 35 percent. 

Why do this? Kim told us simply that “it’s a deal” and that each deal is structured differently. A planning department source told San Francisco that this methodology goes “beyond the standards” planners use, and is “complicated.” When asked if it’s a bit like an angler holding a fish closer to the camera to make it appear larger, however, the reply was “You seem to get this.” 

Now, 35 percent affordability is still significant, considering the relatively paltry requirements this city currently imposes on developers (12 percent affordable on-site; 20 percent off-site). But it’s not 40 percent; the victory cry of “40 is the new 30” isn’t quite accurate. More like “35 is the new 40.” 

But this weird math is only one of several ways in which the numbers have been spun. 

“Affordable” is a term of art in San Francisco. But, for the most part, it’s unusual to affix “affordable” to units set aside for people earning more than 120 percent of area median income, or AMI. (Rather conveniently for demonstration purposes, 120 percent of AMI in San Francisco is currently right around $100,000.) 

Kim’s legislation calls for a slice of affordable housing to be reserved for “middle-income” earners banking no more than 100 percent of AMI for renters and 140 percent of AMI for buyers. And yet, no fewer than 35 of the 241 “affordable” units in the 5M development are earmarked for residents earning 150 percent of AMI (about $125,000). Subtract them from the total, and your 35 percent affordable tally now becomes 30 percent. (30 is the new 40?).

But wait—there’s more.
Let’s say you were going to build an office tower or hotel or other retail structure in San Francisco. Well, it’s no straightforward operation. There is a lot of process. Your office tower stands to attract people to this city. They need to live somewhere. So the city has, for decades, mandated that you pay a “jobs-housing linkage fee” into an affordable housing fund. 

Even if the 5M project weren’t producing a single housing unit, it would be required to pay this fee because of its heavy retail component. And, in fact, the project is ponying up $15.2 million in linkage fees. But this money isn’t going to the city’s fund. The developer is, essentially, paying itself to fund and/or build affordable housing. 

In order to meet standards like those now being proposed for residential projects, 5M is pulling in money from what is, essentially, a separate retail project—money that would have to be disgorged regardless of the project’s residential element. There’s nothing bad about this per se; millions of dollars are being put toward affordable units, after all. But reaching an affordability goal by funneling millions of office-generated dollars into a residential development is something most projects can’t do. 

All of which is to say: If 5M were a more typical residential project trying to meet the standard of 25 percent affordability laid out in Kim’s ballot measure, it would fail. If you also factor out the 60-odd units funded by the jobs-housing linkage from the 5M affordable tally, you come up with 21 percent affordability. Which is, plainly, less than 25 percent. Or 40 percent. 

And yet, perhaps this is still a good deal for San Francisco. In the end, 241 affordable units are slated to be built. That’s not nothing. But it’s not 40 percent. To use all of this magic to conjure up the politically palatable 40 percent tally is not just sketchy math; it could have unintended consequences down the line. If a developer offers 25 percent affordability—without fudging the numbers, without stretching the definition of “affordable,” and without funneling money in from a veritable separate project (and, perhaps, without being gifted a massive upzoning)—he or she won’t appear generous. This will now be viewed as a miserly offer. 

That’s problematic. But so much about San Francisco’s housing situation is. We are told that negotiations are hot and heavy regarding just what we’ll be voting on in the forthcoming election. But talks aren’t centered on altering the 25 percent requirement, no. It’s more about what other projects can be grandfathered out. 

The next few weeks figure to be noteworthy. We’re more than 40 percent sure of it. 

Have feedback? Email us at
Email Joe Eskenazi at
Follow us on Twitter
Follow Joe Eskenazi at