In every city with political battles over zoning and development, there’s a significant number that gets hashed and rehashed by all sides: zoned capacity. Planners use this measure to assess whether the zoning code allows enough housing to meet future demand. Today, we’ll analyze this measure from the perspective of Desire for Density, understand some of its shortcomings, and suggest ideas for improving on it.
Identified problems and solutions
Zoned capacity is a simple concept: if every single parcel is built as big as zoning allows, how many housing units would there be? Unfortunately, the measure is so simple as to be simplistic. The Sightline Institute has documented many of the problems with the measure as used in Seattle, an analysis that could apply anywhere. Some of these problems include:
- It counts capacity that is allowed by some zoning rules but disallowed by others. For example, a parcel may be zoned to allow duplexes, but a more detailed site survey would reveal that the existence of a protected tree leaves only enough room for one unit.
- It counts capacity that is unlikely to get built due to the presence of existing buildings: while a duplex may be permitted on a new lot, the fact that someone just built their custom dream house means that it is highly unlikely to get built..
These flaws can leave both planners and the lay public confused. Frequently, zoned capacity is cited by development opponents as a reason to oppose development. After all, if a city already allows hundreds of thousands of more housing units than it has built today, why allow even more until those get built? In this way, units that are illegal or infeasible to build are used as ammunition against legalizing realistic, affordable homes.
Cities have attempted some measures to get around these issues. In Austin’s recent attempted code rewrite, City Council requested staff ensure a new zoned capacity of at least three times projected 10-year growth to account for the fact that not every zoned unit is feasible to build. In Seattle, the city takes into account the ratios of land values to building values to try to predict which properties are likely to be redeveloped.
What is Economic Feasibility?
The fundamental problem with zoned capacity as ordinarily calculated is that it doesn’t take into account the economics of housing construction. The question “how many housing units can be built?” requires a second question: “at what price?” If a city desires to ensure a stock of affordable housing for both current and future residents, it must ensure not only that enough units can be built, but that they can be built before prices go sky-high. The type and location of capacity affects housing prices more than the gross number of units theoretically possible to build.
Housing will only get built in places where building it won’t lose money. A theoretical capacity, if it is to be meaningful, has to allow housing where it is demanded, as revealed through land values. Even a fast-growing city such as Austin has areas that are languishing despite being zoned for dense housing. A city should not congratulate itself for zoning an area for dense development where dense development would lose money. It might look good on the zoning map, but the practical effect of this paper capacity is exactly zero.
More generally, a meaningful zoning capacity depends on market conditions and the cost of construction. Limiting the capacity for new housing to mid-rises on commercial corridors or high-rises downtown ensures that new housing will not be realized until housing prices have risen enough to support mid-rise or high rise construction. A reasonable zoning capacity allows density to step up gradually, avoiding the sharp discontinuities in housing prices caused by the only-corridors-and-downtown approach.
Let’s illustrate. Imagine two cities, each with 115,000 single-family homes and 40,000 four-plex units.
Both Exlusiville and Inclusivetown anticipate that growth of an additional 71,000 households is likely in the next thirty years. They want to get ahead of it by adding additional zoning capacity, but they go about it in different ways. Exclusiville preserves single-family zoning everywhere, but upzones some of the areas currently holding fourplexes for mid-rise and high-rise development. Inclusivetown upzones fewer fourplexes for denser development, but upzones some of the single-family areas for fourplexes. As a result, the cities have the following zoned capacities:
Construction techniques matter
Exclusiville appears poised to handle growth at least as well as Inclusivetown: it has the same zoned capacity and allows more large buildings. But Inclusivetown can grow more affordably. As more people want to live in Exclusiville, prices will rise sharply before new housing gets constructed. The reason is that no developer will construct a mid-rise building in Exclusiville until prices have reached at least the cost of construction (say, $300k/unit). Stated more precisely, prices in Exclusiville will rise until they equal the marginal cost of building an additional housing unit. After prices reach that point, Exclusiville will be able to add new housing at the same price until it reaches 196,000 units, at which point new housing can only be added once prices rise to the cost of a downtown high-rise unit (say, $600k).
Inclusivetown can grow more affordably. The cost of adding a new dwelling in Inclusivetown is simply the cost of building a new four-plex, which is much cheaper than mid-rise construction. As in Exclusiville, new homes will only be built once prices equal the marginal cost of construction, but the marginal cost of construction is much lower. Inclusivetown can rapidly add housing without price increases. Inclusivetown can add tens of thousands of units before the zoned capacity limit forces developers to switch to a more expensive method of construction. This is the case at every stage of growth. While the marginal cost of the 196,000th unit in Exclusiville is the cost of building a downtown high-rise unit, in Inclusivetown, it’s just the cost of building a mid-rise unit. Exclusiville allows plenty of paper capacity, but the distribution of that capacity guarantees that people in Exclusiville can’t afford to build new housing until prices rise substantially, while developers in Inclusivetown can keep on churning out new fourplexes even at affordable prices.
In Exclusiville, richer folks moving in will buy their way into single-family houses while poorer folks will move away or not have a chance to move in in the first place. This dynamic will continue until prices rise high enough to allow mid-rise construction to at least break even. As more people want to live in Inclusivetown, on the other hand, prices will only rise gently; instead, single-family houses will give way to fourplexes.
Location matters too
Some might be tempted to conclude that mid-rises and high-rises are “causing” the high prices in Exclusiville. If keeping construction costs low is the key to affordability, isn’t the answer to just ban mid-rise and high-rise construction? Not at all! The issue with Exclusiville is not that it allows a lot of mid-rise and high-rise; it’s that it bans cheaper alternatives, raising the marginal cost for any unit to get built.Sometimes high-priced construction techniques are necessary to limit the cost of land per unit. That $25 million parcel downtown might seem obscenely expensive for residential development, but when spread among 250 units is only $100,000/unit, which is likely a quite reasonable land cost compared to the single-family homes nearby. Limiting that downtown parcel to a mid-rise density of 125 units would double the land cost to $200,000/unit, making downtown living much more expensive.
Land prices will likely vary dramatically within both Exclusiville and Inclusivetown. In both towns, people will be willing to pay more to live close to work and central city amenities. Even in Inclusivetown, where people have the option to buy cheaper fourplex construction, some people will prefer to pay a little more for location, location, location, just as some people will be willing to pay more for a detached house rather than a fourplex. A reasonable zoned capacity will allow land prices to guide the type of construction. Allowing dense construction where land is expensive does not increase the marginal cost of adding housing where it is not. Building high-priced downtown high-rises does not raise the cost for someone looking to buy a cheaper apartment two miles from downtown. On the contrary, giving affluent residents the option to settle downtown reduces competition for that apartment, helping it stay cheap.
Even within a city with zoning that allows for enough units of each type, location still matters. If a city were to upzone 10,000 parcels zoned exclusively for single-family housing to allow fourplexes, which parcels are upzoned matters a lot! Fourplexes may not pencil out on the city’s fringes, but may offer vital relief from high land costs in the central city. Depending on how desirable the areas are, a very different number of additional units might get built.
How to do things differently
In our next blog post, we will cover how to use land value ratios to build a more realistic and useful zoning capacity, that doesn’t just answer the question “how many units can we build?” but also “at what price?” That post will have our practical advice to planners and advocates on building a more economically literate capacity analysis. But to hold you over, let’s take a small detour: do we even need capacity analyses in the first place?
Take Conjecture out of Land Planning
A well-done capacity analysis can be a useful measurement of a city’s housing market and its room for growth. But even the best estimates have lots of possibilities for error and bias. Zoned capacity analyses are less science and more a crude conjecture. Combining a capacity analysis with a population growth projection adds even more room for error, as population growth projections have their own different errors and biases and pressures, not the least of which is they depend on the housing capacity!
Rather than pile conjecture upon conjecture, we can use Land Value Ratios directly to make zoning decisions based on objective data that will minimize price increases. The idea is simple: set an acceptable land value-to-unit ratio, and upzone whenever the ratio exceeds a chosen threshold (say, $150k per unit). By upzoning any area where land values per built unit are higher than the threshold, the land value ratio allows you to target the places where additional density is both most wanted by home shoppers and where it will have the biggest impact on affordability. Redoing this analysis every few years — or using an automatic trigger like the land value trigger — would let a city stay on top of problems as they arise, rather than needing to foresee changes in trend that will happen five, ten, or twenty years into the future.