Draft Compliance Guidelines for Multi-family Districts under Section 3A of the Zoning Act
Agency: Massachusetts Department of Housing and Community Development
Comment Period Opens: December 15, 2021
Comment Period Closes: March 31, 2022
Comment Submitted: March 17, 2022
Thank you for the opportunity to comment on the Draft Compliance Guidelines for Multi-family Districts under Section 3A of the Zoning Act. In that draft, the Department of Housing and Community Development (DHCD) lays out praiseworthy and attainable goals for the cities and towns of eastern Massachusetts that comport in spirit and letter with Section 3A of the Zoning Act.
In certain aspects, Section 3A is crystal clear: every MBTA community must zone a district for at least 15 units per acre by right. But the law very quickly retreats from important questions about details: Can municipalities maintain ordinances that make multifamily housing impracticable without explicitly banning it? What objective metrics correspond to suitability for families with children? And, most important, what is a district of reasonable size?
DHCD has to make choices on all these questions, and it has largely made defensible and fair choices. In many cases, there are several possible approaches that are equally valid. For instance, the “reasonable size” calculation could have been scaled relative to jobs, home prices, population, land area, number of transit stations, or—as DHCD chose—housing units.
DHCD also had to strike a balance between giving flexibility to municipalities which are trying to follow the law in innovative ways and applying tough scrutiny to municipalities that might try to game the system without actually allowing any new housing. For example, DHCD requires municipalities to take into account barriers other than zoning that impact housing potential, including environmental restrictions, covenants, and historic district designations.
Section 8(c) offers ample flexibility to municipalities that propose to comply with Section 3A but have unique circumstances that make their MBTA stations poor sites for development. In particular, some towns have a small semicircle of land within a half mile of an MBTA station in a neighboring community. Shrewsbury, for instance, is about a quarter mile from the Grafton MBTA station. The area nearest the MBTA station has a scattering of light industrial businesses quite remote from the rest of the town. DHCD has made clear that it would welcome a proposal from Shrewsbury to rezone land in a walkable, residential part of town rather than the remote station area.
DHCD has also shown flexibility, perhaps to a fault, by allowing the establishment of districts that extend beyond the half-mile radius around a transit station.
In terms of policy substance, DHCD has to strike another balance within the bounds set by Section 3A, requiring municipalities to make substantial changes but not requiring so much that the majority of municipalities fail to comply.
The Multifaceted Goals of Section 3A
The legislation that created Section 3A reveals multiple goals for the geography of new housing that are often in tension. It clearly intends that housing should be transit accessible, open exclusive suburbs to renters, and be distributed fairly across communities.
Focusing on only one of these goals would cause policymakers to neglect the others. If increasing transit use were the only goal, the legislation would have included the City of Boston and focused on high-density rezoning in the handful of cities with high public transit ridership.
The legislation’s other goals are equally important, including allowing moderately priced housing near more workplaces and expanding access to excellent schools.
The high demand for housing in eastern Massachusetts is largely a consequence of job opportunities. Those jobs are scattered across the state. Although Suffolk County accounts for 29 percent of commute destinations of Greater Boston residents, the majority of workplaces are in the suburbs.
Another vital reason for allowing more “multifamily housing . . . suitable for families with children” in suburban towns is to open excellent educational opportunities to families that cannot yet afford a detached home on a large lot. It is often said that a child’s future should not be dictated by his or her zip code. I submit that people should stop blaming the US Postal Service; the arbitrary lines drawn by zoning boards are the more likely culprits for educational segregation.
An indicator of the legislature’s intent to improve housing along several dimensions is that the statute includes even far-flung towns with no transit station. In Lancaster, for example, the Census Bureau estimates that 1 percent of commuters use public transit. No housing there will plausibly be transit oriented, but Lancaster is nevertheless included to maintain fairness, to give people who happen to work in or near Lancaster a chance of finding an attainable home there, and to open the opportunity of attending Lancaster’s excellent schools to families who have been excluded by local regulations.
The multiplicity of goals puts DHCD in an unenviable spot. Every critic will recommend a different ideal balance across the dimensions of fairness, transit, educational opportunity, and proximity to jobs, as well as perhaps other dimensions. DHCD should justify and explain its approach, amend it if necessary, and remember that no formula can be perfect.
Balancing priorities across several dimensions, DHCD has selected from among a range of reasonable choices. Taken as a whole, the result is a draft that will effectively implement Section 3A and expand housing opportunities across eastern Massachusetts. As zoning reforms enacted under these guidelines take effect, new construction will take some pressure off the tight housing market, and residents across the region will find more options to shorten commutes, switch schools, or upgrade their housing.
In the remainder of this comment, I offer several suggestions to clarify and refine the draft guidelines before finalization.
Offer Flexibility in Exchange for Net Density
By statute, the minimum density of a qualifying district is 15 units per acre. However, in practice, 15 units per acre is low for modern multifamily construction. Thankfully, DHCD’s draft guidance allows municipalities to trade district size for greater density on a unit-for-unit basis.
I recommend going even further in allowing tradeoffs between district size and density: municipalities should be allowed to decrease the size of their district below 50 acres on the same unit-for-unit basis to a minimum of 15 acres. This can potentially perform two functions:
- It eases the political difficulties of finding a qualifying site large enough; the smaller the site, the fewer people’s proverbial “backyards” is it in.
- It enables the use of modern midrise construction methods.
On the second point—modern construction methods and apartment management are optimized for midrise apartment buildings, which are typically of four to seven stories and contain over 50 units. But such buildings do not work at moderate densities such as 15 units per acre. Although some communities will prefer low-rise multifamily across a larger area, DHCD should allow smaller, denser qualifying districts for communities that want them.
One caveat is in order: DHCD should not allow municipalities to decrease the size of their qualifying district to less than 50 acres when the proposed district covers an area that is already densely built. The additional flexibility should be available only where the existing average density is less than 15 units per acre (or the rough equivalent in occupied nonresidential space).
Make It Easy to Follow the Law
After DHCD’s guidelines are finalized, the real work will begin in scores of town halls: crafting zoning amendments that will bring municipalities into compliance with the new law. This work will be difficult, especially in towns. Town staff have the unenviable job of studying and crafting potential responses to DHCD guidelines without knowing how town meeting attendees will react to the proposals. Even if most attendees want to comply with Section 3A in theory, some of them will, in practice, object to any specific rezoning that achieves that objective, preferring an alternative. Unlike a city council, the attendees of a town meeting do not have the time and venue to work through such disagreements.
Many town meeting attendees may approach Section 3A with skepticism, unsure how much the state is asking of them. They are unlikely to vote for a zoning amendment without certainty that it will meet DHCD’s requirements.
Thus, DHCD should put town meetings at the center of its agenda and assist town meeting members and town staff in every way possible, including through the following two ways:
- Offer to preapprove (or prereject) proposed zoning amendments drafted by staff.
- When practicable, offer to attend town meetings to help resolve interpretive questions.
The most important principle is to keep things simple: DHCD should not give in to those who want to add bells and whistles to a program that is already asking a great deal of town meetings. Although inclusionary zoning and high environmental standards are attractive to many people, including them in the Section 3A guidance would go beyond the statutory mandate and create another political hurdle for municipalities trying to comply.
Municipalities retain the authority to impose such additional restrictions and may use such additional restrictions to weaken Section 3A. DHCD should reject bad-faith restrictions that go beyond normal practice and appear calculated to make multifamily housing economically infeasible.
Clarify Mixed-Use Zoning
In an op-ed, Sarah Khatib and I note that the statute leaves unclear whether mandatory mixed-use zoning qualifies as multifamily for the purposes of the statute. Walpole, where Khatib serves on the planning board, has successfully built mixed-use buildings in its town center, creating an attractive model of how other communities might look after the success of Section 3A.
To achieve this outcome, however, Walpole has had to grant variances from its zoning code, which requires a high ratio of commercial to residential space in the downtown zone. Planners typically overestimate how much retail a building can support, and municipalities could use mixed-use requirements to make development uneconomical.
I recommend that DHCD clearly require that multifamily housing be allowed as a stand-alone use in qualifying districts. DHCD should also clarify that municipalities can allow mixed uses in their qualifying districts. Both types of development are desirable, but planners cannot reliably predict where and in what proportions mixed-use development will succeed.
Clarify the Contiguous Land Rules
In order to prevent towns from gaming its rules, DHCD has wisely required that at least one portion of a qualifying overlay district comprise 25 contiguous acres and that no portion be less than 5 acres. The language in this section raises a few questions that DHCD can easily clarify:
- Do the same rules apply to base zoning? Or can a town apply qualifying base zoning to scattered sites in order to comply? One hopes not.
- How does the contiguity requirement interact with public rights of way, parks, or other areas that may be immune from zoning? A commonsense definition would consider parcels that face each other across a street to be contiguous for these purposes. If the parcels are separated by a river or railroad track, however, it is not obvious that they should be considered contiguous.
Ease the Requirements Facing the Least Populous Towns
In keeping with the goal of ensuring that all towns participate in providing multifamily housing, DHCD sets a minimum qualifying district size of 50 acres for all communities. However, the result is that the potential number of units that can be built in a handful of small towns is very large relative to those towns’ baseline housing stock. In Princeton, for example, the 750 homes that could theoretically be built on 50 acres zoned at the statutory minimum would represent a 54 percent increase in the town’s housing stock.
This is not necessarily unjust—Princeton’s nearly 23,000 acres make it twice the size of Newton and nine times the size of Watertown, each of which has a far greater population. Princeton could easily find 50 acres to rezone for moderate-density multifamily housing without any visible change anywhere else in town. In fact, Princeton’s small population derives in part from its maintenance of strict zoning, requiring two acres per house.
However, the long-term viability of Section 3A requires that the state take political considerations into account. Princeton should not be able to claim that it is bearing a heavier burden than any other town.
To avoid the appearance of unfairness, I suggest capping the requirement at 25 percent of a town’s housing stock. This could be done in two ways: (a) allowing towns to zone for a smaller qualifying district or (b) allowing a town to cap the number of permits it must issue under Section 3A. In the latter scenario, Princeton would still be required to zone 50 acres. But after issuing by-right permits for 346 units in those zones, it could declare a moratorium until some distant future date.
Relatively few housing units are at stake in the 29 towns that have fewer than 3,000 housing units. Ensuring that DHCD’s guidelines are viewed as fair and reasonable is likely of greater importance to Section 3A’s long-term viability.
Prioritize Grants That Ease the Growing Pains
Although the grant-making power of the Executive Office of Housing and Economic Development (EOHED) is not directly related to DHCD’s draft guidance, the two are pragmatically related. If a large share of municipalities fails to promptly comply with Section 3A, compliant municipalities (as well as those in other parts of the state) will find much less competition for MassWorks and other grants conditioned on compliance with Section 3A. Although these grants fund many worthy goals, EOHED should prioritize those that are aimed at easing the growing pains that can come with housing growth.
The specific needs vary—sewers, sidewalks, or stormwater, for example—but in all cases, community members can resent the addition of new neighbors when they feel that common goods are being depleted. Softening that criticism, especially with grants made in the coming year as zoning reforms are being debated, would help unlock the many positive spillovers of growth. Thus, I recommend shifting grants toward meeting broad-based needs in communities that plan to grow rather than supporting buildout on specific sites or managing problems unrelated to growth.
DHCD deserves thanks for the effort it has already put into producing reasonable and balanced guidelines that faithfully implement Section 3A. When DHCD’s drafting concludes, the real work begins—of implementing new, inclusive zoning bylaws and of building the homes of thousands of new residents.