“I have yet to see any problem, however complicated, which, when you looked at it in the right way, did not become still more complicated.” – Poul Anderson

Upzoning and reduced regulation are not magic bullets. Indeed, almost 20 years of data from across the country consistently shows that zoning reforms alone end up producing only a fraction of the new housing the country needs.  Restrictive zoning is correctly understood to cause higher housing costs and less construction.  But the opposite turns out not to be true; reduced regulation cannot on its own create enough housing to lower costs.

Projections for new housing created through zoning reforms have been overly optimistic.  On-going research finds that over the long-term, areas untouched by zoning reform seem to have produced almost as much new housing as upzoned areas.  According to the Urban Institute’s 2023 Land-Use Reforms and Housing Costs: Does Allowing for Increased Density Lead to Greater Affordability, rental housing supplies in 1,136 cities across the country only increased by an average of 0.8% where zoning restrictions were loosened.

It’s not that zoning reform doesn’t work – it does.  The problem is that the financial viability of small-scale housing development and the value of existing properties, when combined with other regulations/processes still on the books, make infill housing projects difficult, if not impossible.

The data from that and many other sources shows that the upzoning of densely built-out neighborhoods does not increase supply because the potential profit from individual projects does not justify the additional costs of small buildings on in-fill sites.  Studies from Oregon to Virginia have come to this conclusion: rezoned parcels are either too small for any kind of redevelopment (including ADUs), or their existing houses are already worth so much that adding density does not make financial sense for either developers or homeowners.

Buying, demolishing, and adding more units on infill lots in existing neighborhoods remains highly controversial, whether or not as-of-right zoning is in place.  Examples from across the country show that it is hard to justify redevelopment when:

• Even “as-of-right” projects face difficult and lengthy community processes including site plan approval, conservation commission, fire department review, etc.;
• Construction costs increase exponentially with height and density on small sites;
• Infill construction is far more challenging than greenfield building, and;
• The price of homes is so high right now that adding units to existing properties creates little-to-no additional, residual value.

Los Angeles streamlined its process for homeowners to create duplexes or subdivide lots starting in 2017, leading to an impressive increase in ADUs (Accessory Dwelling Units), from 93 in 2016 to 3,100 in 2020.  However, those ADUs only accounted for 14.8% of LA’s 20,903 housing permits in 2020, which represents barely 1.1% of the 270,000 additional housing units the city says it needs.

A February of 2022 report to the Planning Commission of San Francisco outlined major hurdles the city faced in effectively reforming zoning to create more housing, finding that:

• The residual development value of demolishing existing single-family homes to develop new triplexes and fourplexes ended up below current single-family home prices;
• Low financial viability made requiring affordability in these projects difficult and/or impossible;
• Financial feasibility was challenging in every neighborhood studied, and;
• Smaller additions or ADUs, while more feasible, would likely be successful only on a limited number of lots that have sufficient land.

In Chicago, the zoning reforms of 2013 and 2015 increased permissible densities and reduced parking requirements near transit stops.  Yet the upzoning appears to have only impacted development on about 6% of the land deregulated, creating only slightly more development when compared to lots where the zoning was left alone.

In Portland, Oregon, economists estimate there is a need for 140,000 more homes right now, and another 440,000 over the next 20 years.  The city upzoned residential areas in 1996, yet in the following 20 years, only 22% of the rezoned parcels were actually redeveloped.  And while the changes did lead to a 150% increase in permits for two-to-four-unit buildings, those 814 permits accounted for only 4.7% of the 17,417 new housing units approved in 2023 – less than 4% of the 22,000 units needed each year.

Minneapolis changed its zoning in 2018 to allow moderate density on single-family properties citywide.  And indeed, that increased the number of building permits for duplexes, triplexes and fourplexes from 13 in 2015 to 53 in 2021.  But like Portland, that seemingly large 408% increase represented only 0.7% of the that year’s 7,811 housing permits.

The Virginia Housing Commission received a clear analysis of the challenges to upzoning earlier this year.  That report concluded that upzoning, while effective, does not necessarily produce housing at the desired pace or scale, finding that on its own zoning reform can only:

• Create improvements to housing stock and affordability in limited settings;
• Be a partial solution to affordable housing issues;
• Has the greatest effect on moderating the costs of higher-end rental units, and;
• May not produce additional affordable rental housing for low-to-moderate income households.

Larger scale zoning changes have a better history of producing housing at the scale required.  Locally, the Assembly Square area in Somerville was an underutilized, industrial area that until 2010 literally provided housing for no one.  By 2020, redevelopment created 1,006 apartments for 1,660 people, which was almost 40% of Somerville’s total new housing production in that period.  Seen another way, that is a full 12% of the city’s total housing unit growth since its population peaked at 102,351in 1950 (which is almost 22,000 more residents than Somerville has today).

Assembly will eventually have 1,843 units housing around 3,000 people if the area’s current people per household trend continues.  For people concerned with school impacts, the 2020 Census showed only 138 of those 1,660 new residents were under 18, which is basically no impact on a school district whose enrollment dropped by more than 1,400 students from 2000 to 2020.

Keeping up with demographic changes and housing demand has proven to be very difficult.  That is because there are many, many impediments on the road to new housing, and upzoning is only one tool in the larger toolchest of the changes needed.   Zoning reform has proven highly successful in underutilized areas or those zoned for outdated uses, but it is proving less successful in creating the large numbers of new units we need.

Existing, dense neighborhoods with small lots provide limited opportunities to create viable, larger parcels and projects that can cover today’s expensive construction market.  In established, urban in-fill markets the availability, size, and value of the existing land may matter as much or more than zoning.

The level of effort expended on regulatory reform has not, unfortunately, been matched by new housing construction.  Only ADUs have turned out to be truly viable, and they seem unlikely, due to their limited size, to provide the sheer scale of new housing needed countrywide.  Upzoning alone, without due consideration of other impediments and the value of the existing buildings and land, has proven incapable of solving the problem people thought it could solve.

Scott Pollack is founder of SRPlanning (SRPlan.net) and serves as Co-Chair of ULI Boston/New England’s Housing Roundtable. Susan Connelly is Chief Operating Officer of Housing Opportunities Unlimited. Please send any reactions, comments, or ideas to Scott at [email protected].