Zoning for More Housing
Zoning for More Housing
A. Rezone underutilized land currently zoned for industrial and commercial uses to encourage moderate to high density housing.
A large majority of the vacant and underused land in San Francisco is currently zoned for industrial uses (M-1 light manufacturing or M-2 heavy manufacturing), with low height limits, making housing a disfavored land use. Only live/work projects, which are currently supposed to be reserved for occupancy by artists only, are permitted in these industrial lands, leading to a boom in the construction of loft-type live/work units, but a shortage of land available for more conventional housing construction.
All M-1 and M-2 areas of San Francisco should be re-examined to determine whether some portions of those zones can be changed to residential or mixed use zoning, with appropriate adjustments in height limits. The zoning designations for these areas date back to 1935, when San Francisco's job base was largely industrial. The job base has changed over time, but the zoning has not, leaving many hundreds of acres of land zoned exclusively for heavy industrial uses that in all likelihood will never return. The types of "industry" that are attracted to San Francisco (electronics, multi-media, printing, business services, etc.) are not the types of smoke-stack industries of the past that needed to be segregated from residential development. Rather, housing can co-exist with these newer, cleaner industries, which often require less land than older industries.
An example of how this strategy might provide incentives for private developers to provide housing on a 10,000 square foot lot currently zoned M-1 with a 50-foot height limit is as follows: Under current zoning and height controls, up to about 30 units could be built, but would require conditional use approval (which is uncertain and thus risky for any developer), and the new housing could be located next door to a future noxious heavy industrial use, since those uses would continue to be permitted in the M-1 zone. Most developers would not take these risks. Were the same lot zoned RC-4 with a 84-foot height limit, about 50 units could be developed, and the developer would be assured that future neighbors of the project would be compatible with residential development. Intervention by the Redevelopment Agency often will be necessary to create the types of infrastructure improvements necessary to support new residential neighborhoods. The Agency is well-equipped to orchestrate the creation of new neighborhoods through its powers of land assembly and the issuance of tax increment bond debt to finance essential infrastructure. MUNI service to these areas may also need to be augmented.
SPUR's March 1998 report, The Future of the Central Waterfront, was a case study of this notion. It examined the Central Waterfront between Mission Bay and Islais Creek, one of the largest underutilized areas of San Francisco, and concluded that up to 12,000 housing units could be developed in that area with the right combination of rezoning, public intervention, and private investment. These strategies apply equally to other areas of San Francisco. The Planning Department has just embarked on a land use inventory study of San Francisco. That study must address these issues.
B. Enact a City-wide density bonus for affordable units.
State law requires local jurisdictions to provide density bonuses to developers in exchange for providing affordable units. Yet, San Francisco has never implemented this legal requirement except on a case-by-case basis, requiring a time-consuming legislative action for every density bonus considered, resulting in very few additional affordable units. We cannot afford to stifle the production of additional affordable units in this way.
To meet State law, the City should enact a City-wide density bonus program that would permit builders, within the existing height and bulk limits, to increase allowable density provided a certain percentage of the additional units (above the base density limit) are affordable units. Developers would be willing to voluntarily construct additional affordable units when their marginal costs of doing so are equal to or less than the price at which they could rent or sell the units to low-income households. A density bonus would also make it more feasible for developers to meet the City's current policy requiring 10% of units in market-rate projects to be affordable.
This reform would also permit non-profit developers of low-income housing to produce more units on the a particular piece of land. When density is increased, the marginal costs per unit are lower, since the land prices, soft costs, and foundation costs can be amortized over more units.
Decision-makers will need to abide by this bonus program and permit the additional units to be built in the face of the opposition that certainly will be expressed by some neighbors, who frequently oppose any new developments, much less higher density low-income units. To avoid potential opposition, perhaps this Code change could at least initially not apply in very low density zoning districts, excluding, for example, all single family zoning districts.
C. Increase height limits and densities along neighborhood commercial corridors and major transit routes, encouraging housing over retail.
There are numerous opportunities for the provision of higher density housing along moderately-scaled neighborhood commercial corridors and major transportation routes, such as Geary Boulevard and Third Street, both stand-alone and above ground-floor retail. These areas already have in place all the urban infrastructure needed to support relatively intense residential development (transit, retail, community facilities, etc.) and are generally at least slightly removed from surrounding low-density neighborhoods. Moreover, San Francisco's General Plan calls for such increased density along transit routes, but those policies have never been implemented.
It is time to begin implementing these General Plan policies. With even a 10-foot increase in height limits in many neighborhood commercial districts (from the current 40 feet to 50 feet), significantly more residential units could be created with less expensive wood-frame construction methods, which allow four floors of residential construction above a retail and parking podium.
In mixed retail/residential projects, the costs of land and construction overall could be cheaper. Existing retail areas should be studied individually to determine suitable locations for encouraging the addition of several floors of residential on top of commercial uses, including neighborhood shops, supermarkets, big box and shopping center-type retail, and institutional structures, such as post offices.
Given the fact that most commercial areas are located on good transportation routes, a reduction in parking requirements should be encouraged, particularly given the fact that commercial and residential parking can often be shared.
D. Revise height and bulk limits in certain residentially zoned areas so that the height and bulk limits are better synchronized with Building Code height restrictions and building envelopes that are feasible.
Currently, there is not a rational relationship between the height limits set forth in the zoning maps and Building Code height restrictions. Applicable height limits should be revised to better reflect Building Code standards.
For example, the current 80 foot height limits in some residential zones should be changed to 85 feet. The San Francisco Building Code allows buildings with a height to their top occupied floor of 75 feet or less to be designed with less expensive low-rise standards. As residential buildings typically have floor heights of ten feet, an overall permitted building height of 85 feet (75 feet to the last occupied floor plus ten foot story height) would be required to fully maximize the provisions allowed in the building code. An 85 foot height limit in lieu of 80 feet would permit one additional floor at maximum cost efficiency.
Certain bulk limits also strongly discourage new residential construction, by requiring buildings to be so slender that it is infeasible to construct them. Typically, at least 10 units per floor (about 12,000 gross square feet) must be built in a high-rise residential structure to be feasible, since each floor requires two elevators and two enclosed exit stairs. Yet, the bulk limits in the many districts of the City that permit high-rise housing require floors as small as 6,400 square feet, about half what is practical to build. The City should re-examine the complex bulk controls set forth in the Planning Code so that those controls do not unduly inhibit needed residential construction.
E. Permit planned unit development approvals on lots smaller than a half-acre.
Currently, the Planning Code permits projects to be considered as planned unit developments (a flexible planning category not bound by strict zoning rules) by the Planning Commission only if the lot size is a half-acre or larger and outside the downtown. The PUD process frequently permits more units with a better, more feasible design, because the strict density, rear yard, setback, parking and other requirements of the Planning Code, that were designed for the typical 25 x 100 foot lot, can be relaxed. At the same time, a PUD must be approved as a conditional use by the Planning Commission, giving the commission discretion to examine and regulate the overall design of each PUD, and to impose conditions requiring about 10% of the units to be affordable.
Given the small size of many development lots in San Francisco, limiting the PUD process only to lots of a half-acre or larger is not necessary or desirable. The threshold should be reduced below a half-acre, perhaps to lots of 10,000 square feet or more. Similarly, the rather arbitrary prohibition on PUD's in downtown C-3 districts should be lifted.
An example of how this Code change could result in more housing units (and more affordable units) is as follows: Under the current Code, on a 10,000 square foot lot zoned RM-3 (a typical zoning category for moderate density districts), up to 25 units could be built. If the builder were allowed to apply for a PUD on the same lot, up to 49 units could be developed, 5 of which would be required to be affordable units. The builder would need to go before the Planning Commission for conditional use approval, so that massing, design, parking and other neighborhood concerns could be addressed.
F. Eliminate the requirement in downtown (C-3) zoning districts that housing projects exceeding the base density limit purchase TDR's (transferable development rights).
When the Downtown Plan was enacted in 1985, it intentionally limited the base density allowed in the downtown C-3 districts so that developers of large projects would need to purchase transferable development rights (TDRs) from historically significant structures that were protected from demolition. Unfortunately, this requirement applies to housing as well as commercial space, so that any housing project exceeding the relatively low base FAR (floor area ratio) in a C-3 district can be built only if the developer buys TDR's from a nearby commercial building that has been rated as historically significant.
This requirement applies only in C-3 districts and has discouraged the construction of any new high-density housing in these districts, the exact location where high density construction is encouraged, by adding a significant cost to each project. The Planning Code should be amended to not impose density limits on residential uses in C-3 districts, but simply limit density through height and bulk regulations (such as is employed in the Rincon Hill and Van Ness Plan areas). This Code change would eliminate the disincentive that now exists to build housing in C-3 districts.