The most important issue discussed at the September 27th Joint Work Session of the City Council and Planning Commission was the reduction or even elimination of minimum lot sizes. We want to give some sense of how mooted changes in the minimum lot size requirements of the zoning code’s subdivision rules could significantly increase the maximum potential density under the proposed new code.

First, we need to explain the current situation. The existing zoning code stipulates several conditions that need to be met in order to allow for the division of an existing land parcel into multiple parcels. These rules are spelled out in the City Code here. The main requirements relate to the minimum size of a new lot, minimum street frontage, and some rules on lot shape and orientation. The reason for such requirements are not hard to understand: the zoning code could in many respects be easily circumvented if lots were infinitely divisible; lots without adequate frontage or access could present a challenge for emergency services. Right now, to give you a sense of the rules, in R-1 zones, the minimum lot size is just over 8000 square feet (about 1/5th of an acre) and in R-1S zones just over 6000 square feet. Minimum street frontage is, in most cases, 50 feet (though a developer of a very large parcel always has the option of building a street network as part of a new plat and thereby creating more frontage).

So what was proposed at the Work Session? Code Studio, the sub-contractor in charge of drafting the new Zoning Code, queried Commission and Council on what they would like to do about minimum lot sizes. The options were to keep requirements as-is, to unify minimum lot sizes across the city at some new level, or to do away with them entirely. The response of Council and Commission was unanimously against the preservation of the status quo, and a consensus formed around either a very small uniform lot size or none at all. One very pro-density Commission member seemed startled that a full elimination would even be considered and asked how it could be workable — couldn’t a developer just infinitely subdivide a lot into postage-stamps and build infinite units, in theory? Code Studio responded that in the case of full elimination, there might need to be consideration of a some kind of cap on Dwelling Units per Acre (something the Commission has in the past expressed a wish to move away from), but without that, the only limit would be what the Fire Department could live with in terms of access. If a Fire Truck can bully its way to within a 150-foot hose haul of each building envelope, that’s a viable lot.

It is easy to see how this can very dramatically change the outer envelope of development intensity under the Zoning Ordinance. Our reading of the CP — a fair reading, we believe — was, to use R-1 as an example, a given lot would go from a current maximum of one unit to a new GR maximum of 3 units (or MIR of up to 12, ignoring bonuses). It seemed like this new maximum would prevail even in case of subdivisions. Partly, this was about encouraging affordable housing. If it were easy to increase entitlements via subdivision, a developer could get to her desired level of density by subdividing instead of by providing affordable housing units to obtain a density bonus. But now, a developer would be able to subdivide a given plot into potentially very small lots and use each one up to the zoned maximum in terms of units and mass. A quarter-acre lot with a single house might now become not just three units in GR, but could rather turn out as 3 lots with 3 units each, none affordable!

As an aside, though it is not the main issue here, we note that there is simultaneous proposal to allow the division of zoning lots into smaller fee-simple ownership lots. The “sublots” would not create any additional unit entitlements. However, they might distort what gets built on newly subdivided lots in unpredictable ways. Under the existing arrangement, a lot with 3 or 6 units of entitlement would, if the developer wanted to make all the units owned, most likely pursue a condominium plan. Once a developer goes this route, most likely the built form would be a multifamily building. But if a developer can pursue fee simple ownership (more popular, perhaps, among buyers) only by slicing up the lot into sublots, the built form would likely look like a bunch of cottages built in depth. That is why some people have said, this is a recipe for 1 acre to turn into 30 cottages, which would be a strange and environmentally unfriendly result. We don’t think this is an economically probable outcome, but it does show the potential unintended consequences.

We have performed some rough calculations of how many new lots could be created in GR and MIR zones under a range of minimum lot sizes and minimum frontages. To be clear, we did not hear anything explicit about the removal of minimum frontages, but based on the comments about fire code becoming the only constraint, we find little reason to believe the Planning Commission would not also advocate in favor of reducing this criterion as well. They just have to realize it exists — many members of the Council and even Planning Commission are shockingly ignorant of the subject area. We did the calculation in two ways: first, we assumed tear-down. Second, we took out lots where subdivision looks difficult without tearing down the existing house due to the structure’s position within the lot. Bear in mind, the “tear down” algorithm represents a rough cut. You can see details here.

For context, there are just over 10,500 R-1/R-2/R-3 lots (including subcategories) at present. For these lots, the median size is 8900 square feet and the median frontage is 62 feet. The lowest minimums shown above, therefore, represent extreme values in the current context. Fewer than 500 lots have both less than 5000 square feet of area and less than 30 feet of frontage.

You can visualize the potential impact with the map below, which shows results for 5000 square foot minimum lot size and 30 foot minimum frontage. Blue parcels represent parcels where subdivision appears at least plausible without tearing down the existing structure; red parcels are those that do not appear to be easily subdivisible unless the existing structure goes. Click on a parcel to see how many additional parcels could be created out of it.

The map clearly shows how dramatically the move would multiply the potential density increase allowable under the new zoning code. The whole CP process has felt to us like a classic case of “salami-slice” tactics. It started with the suggestion of “gentle density” and little by little restrictions have been further loosened and the entitlement envelope expanded, until we find ourselves with a code that pretty much amounts to “developers can do whatever they like, wherever they like.” Council will finish slicing the last of the salami, and then developers can move on to slicing up lots…

September 2022