UCLA Center for Regional Policy Studies
Building Up the “Zoning Buffer”
Using Broad Upzones to Increase Housing Capacity Without Increasing Land Values
Author: Shane Phillips,
Housing Initiative Project Manager
Excerpts:
U.S. cities spent much of the middle and late 20th century reducing capacity for new housing through extensive downzoning zoning or never allowing higher densities in the first place, leading to a shortage of homes and rising prices in high-demand locations. The nation is now dominated by single-family-only zoning, sharply limiting housing production in the places it’s needed most. This shift has had profound consequences on the cost of development and the affordability of housing.
The gap between a city’s current population (or housing stock) and its estimated [zoned] capacity is what I call the “zoning buffer.” Before 1960, the buffer in both New York and Los Angeles was at least 300%. New York’s fell to roughly 50% after the 1961 zoning update, and it was just 12% in Los Angeles in 2010. In this paper, I explain why small zoning buffers are insufficient to improve long-term housing affordability, and why a large buffer may in fact be critically important for achieving this goal.
In built-out cities most growth occurs by building more densely on already-developed parcels. Rather than being limited by the number of vacant parcels, cities today are limited by the number of parcels on which denser housing is permitted and the level of density allowed on each parcel. They’re limited by zoning. In the real world we don’t need to invent new construction techniques to reduce the cost of new housing; we just need to change our zoning. If we can rezone without changing the land prices dramatically, housing will become less expensive; if we can’t, it won’t. We’ll consider two scenarios.
In the first scenario, 5% of parcels are upzoned to allow three homes each, enough to increase the total number of homes in the town by 10% (accounting for demolition of the original homes).
Because these parcels are in short supply and demand is high, their price rises to ensure that homes built on them meet or exceed the price of existing homes in the neighborhood5 — the price of the upzoned parcels will increase. The landowners hold all the cards, and they’ll capture the value of minor zoning reforms before any home-seekers can benefit.
In our second scenario, the town upzones much more extensively, allowing triplexes on 50% of its parcels, which is enough to double the number of homes in the city. There’s land zoned for higher density everywhere; the right to build more is no longer a precious commodity. In this case, the amount of land zoned for redevelopment — and crucially, the amount available for sale at any given time — exceeds the amount of land demanded by developers by a wide margin.
This is important: Cities can’t do much to increase the number of parcels for sale, but they can increase the number with redevelopment potential. About 7% of U.S. homeowners move each year, so when a large number of parcels have redevelopment potential, natural turnover will ensure that property owners bid the sale price of these parcels down (relative to what developers are willing to pay) rather than developers bidding prices up (relative to what homeowners are willing to pay).
There are only a few modern examples of broad upzoning in the U.S., but their early results are promising. In Minneapolis, where single-unit zoning was abolished in 2018 and triplexes can now be built on residential parcels citywide, home prices on affected parcels increased by only 3%-5% compared to similar unaffected parcels in neighboring cities.
Today, and perhaps only during the past 50 to 75 years of human history, the demand for developable parcels exceeds the supply in many U.S. cities. Rising prices, overcrowding, long commutes, and demographic shifts such as the rise of one-person households all point to a housing shortage. It’s a result of reduced capacity — of zoning, which was invented in the early 1900s and became increasingly restrictive over the ensuing decades. What was once abundant is now scarce, not primarily because of physical, demographic, or economic changes, but because of prohibitions on the legal right to build more homes on most urban parcels.
So long as this artificial land scarcity persists, our cities will continue to provide insufficient housing. Prices will keep rising, or at best they will hold at current levels, unaffordable to too many households.
Value Capture, Windfalls, or Broad Upzones
A lack of zoned capacity is what motivates the practice of “value capture,” a set of strategies employed by planners to capture the increased land value created by zoning changes, and direct it toward public rather than private benefit. Planners can “capture” much of this additional value by tying additional requirements to the redevelopment of the property (e.g., by mandating that one of the four units rent to a low-income household at a below-market price). Because this requirement represents a large financial loss to the developer (relative to the parcel’s potential in the absence of the requirement), they will need to pay less for the land for their project to remain profitable.
We can see evidence of the value unlocked (or revealed, or transferred) by increased zoning capacity — as well as the interplay of windfalls and value capture — in a recent community plan update near the Metro Rail E Line (Expo) in Los Angeles. The area to be upzoned around the Expo/Bundy Station was expanded to include several dozen parcels in a neighborhood known as Bundy Triangle.
One way to interpret the Bundy Triangle case is as a failure of value capture design. Planners underestimated the value that their upzoning would unlock and they failed to capture enough of it, and as a result the property owners captured it instead. If planners had captured all or most of the value, the price of the homes sold after the upzoning would be closer to the price of those sold before.
Imagine, for the sake of argument, that every parcel in Los Angeles currently zoned for single-unit detached homes, duplexes, and triplexes was rezoned to allow up to 10 units in modest three- and four-story buildings. With more than 400,000 such parcels in Los Angeles11, this would increase the city’s zoning capacity by at least 3.6 million units, 2 ½ times the city’s existing stock of 1.4 million homes and more than its estimated capacity in 1960. There are approximately 25,000 single-family homes sold on these parcels each year; if just 5,000 of these were sold and redeveloped to their maximum capacity, the city would add 45,000 units to its housing stock annually.
Recalling our hypothetical town and the scenario in which 50% of parcels are rezoned to allow for triplexes, homeowners in this 10-unit scenario cannot sell their parcels for a premium — there are too many just like them. The capacity for housing has increased, but the land price has not. A large zoning buffer allows properties with redevelopment potential to transact at lower prices, ultimately delivering lower-cost housing to residents. In the long run, windfall- and value capture-based approaches both fail to improve affordability, but for slightly different reasons. Windfalls fail because potential savings on land costs are fully captured by landowners. Value capture fails because potential savings are transformed into small quantities of income-restricted units, with any excess value devolving to landowners.
Value capture is especially poorly suited to creating long-term, marketwide affordability. First, as discussed above, capturing savings in the form of income-restricted units constrains the ability to build lower-cost market-rate units. Second, value capture contains the seeds of its own destruction: If rents and sales prices ever did fall, developers would no longer earn enough from market-rate units to cross-subsidize losses on income-restricted units, and housing production would grind to a halt until prices returned to earlier levels. When prices rise, on the other hand, there is often pressure to increase inclusionary requirements — from 15% to 25%, for example — which puts a new, higher floor on the price of housing at which development is feasible.
Implementing Broad Upzoning
The upzones must also include at least moderate density — triplexes or fourplexes won’t cut it in most urban areas. Minneapolis changed its zoning to allow triplexes citywide in 2018, but thus far only about 70 units have been produced. These underwhelming figures can be blamed partly on barriers to redevelopment such as owner-occupancy requirements, but density is key. Redevelopment is costly: It requires design, entitlement, and permitting; demolition of the existing home; construction, including financing; and the loss of rental revenue or use of the property during construction. Enduring all this to replace a functional single-family home with a triplex or fourplex will rarely be worth the trouble.
[Broad-scale upzoning] allows housing development that is substantially denser than existing land uses, and to allow it on many parcels, throughout the city. Rather than using financial inducements to promote the sale and redevelopment of land, these broad upzonings — and the large zoning buffers that characterize them — rely on an abundance of development opportunities to produce adequate housing. This abundance limits the market power of property owners, reducing or eliminating the land value increase that accompanies more concentrated upzones, and allowing for the development of lower-priced housing.
[Broad-scale upzoning] benefits both market-rate and income-restricted housing developers, favors small-scale developers — the builders of “missing middle housing” , is also more welcome in many places than taller, denser developments.
Comment:
In this research paper Shane Phillips makes a good case: that small zoning buffers are insufficient to improve long-term housing affordability, and why a large buffer may in fact be critically important for achieving this goal. Daniel Herriges, in his article: “What Would Mass Upzoning Actually do to Property Values?”, sets up a similar hypothesis. What’s true: Upzoning a property, all else equal, increases its market value substantially. What’s not true: Upzoning all or many properties will substantially increase the market value of every property.
Currently in many cities where housing production has lagged is suppressed demand for new housing. Now imagine you upzone one small pocket of the single-family residential part of the city for high density development. Typical urban renewal districts illustrate this, by limiting upzoning to targeted areas. What would happen? A few neighborhoods or corridors would absorb all the development pressure from a much larger region around it that has not been upzoned. You would see large-scale construction, and a “speculative feeding frenzy” on land. And you’d likely see property owners not only selling to apartment developers, but others holding out for a similar deal, driving prices up to a point where only luxury units would be feasible, or some units would have to be heavily subsidized.
Now suppose you enlarge the zoning buffer to include many more properties, disperse the rezoning to include predominantly single family neighborhoods, and furthermore restrict the allowed redevelopment to moderate density levels. Would this rapidly drive up land values and trigger speculation? The short answer is no. Those who warn of this outcome are committing a classic “Fallacy of Composition”. This is a logical fallacy in which you assume that something that applies to the individual parts of a whole must also apply to the whole.
As Phillips states: broad-scale upzoning allows housing development that is denser than existing land uses, and to allow it on many parcels, throughout the city. This benefits both market-rate and income-restricted housing developers, and small-scale developers — the builders of “missing middle housing”. The latest wave of zoning reforms allows the next increment above single family housing—typically some version of duplexes through fourplexes. Several states and cities, beginning in Minneapolis, and recently Oregon and California have modified their zoning codes to eliminate exclusive single family zones. Rather than setting off a wave of speculation, conversions to moderate density housing types has been gradual, and widespread.
Minneapolis changed its zoning to allow triplexes citywide in 2018, but thus far only about 70 units have been produced. Since 2019, the Portland Bureau of Development Services has received about 80 applications for 3-4 unit conversions. A new update of the city’s Residential Infill Project zoning code is likely to nudge up these numbers. These underwhelming figures can be explained partly by barriers to redevelopment costs: design, permitting, demolition of the existing home; construction costs, financing; and the loss of rental revenue or use of the property during construction. However, as our research reveals, a change to land value taxation would benefit conversions from single family to moderate density structures, saving owners thousands of dollars in annual property taxes compared to Oregon’s current tax structure that taxes structure value heavily.
As for value capture being poorly suited to creating long-term, marketwide affordability, let’s consider that its prudent use could actually make it an effective tool. First, limit value capture to transit oriented development (TOD) in designated rail transit station areas where land values are rising rapidly. Craft a special levy that captures on an annual basis only the land value premium attributable to the location of the transit station. Use the betterment proceeds for TOD improvements such as transit community amenities, not rail line construction or operations. This would in turn further raise land values in the transit district, capturable in subsequent years by the special levy. Do not impose income restrictions on new dwelling units in the TOD district, thereby “putting a higher floor on the price of housing at which development is feasible”. Instead use some of the levy proceeds to construct below market rate housing owned by public or non-profit entities.
Tom Gihring, Research Director
Common Ground OR-WA
