Portland’s inclusionary zoning requirement is a slow-motion train-wreck; apartment completions are down by two-thirds, and the development pipeline is drying up
This will lead to slower housing supply growth and increasing rents for everyone over the next two to three years
Inclusionary Zoning (IZ) creates perverse incentives to under-utilize available land
In December 2016, Portland’s City Council enacted a strong inclusionary housing requirement. Henceforth, all new apartment buildings in Portland would have to set-aside a portion of their units for low- and moderate income housing. Unlike other cities that either made compliance voluntary, or largely (or entirely) offset the cost of the added units with density bonuses or subsidies (or other quid pro quo), the Portland ordinance applied to nearly all apartment buildings larger than 20 units. The new requirement didn’t kick in until February 2017, and there was a land rush of developers who filed under the old rules. That produced a temporary flood of new apartment buildings, that have, over the past four years, mostly been built.
Investment markets work with lags for a variety of reasons. It takes time to plan, obtain permission for, and actually build new housing, and multi-family housing takes longer than single family housing. As a result, there’s a multi-year pipeline. When there are housing shortages, as there were in the early days of the recovery from the Great Recession, supply can’t expand as rapidly as demand, and rents get bid up. The reverse is also true; a glut of building in good times produces new apartment supply that holds down rents, at least for a while. That effect has concealed the negative consequences of Portland’s inclusionary zoning policy.
As we observed in May of 2019, the initial implementation of inclusionary zoning resulted in a kind of counter-intuitive acceleration of apartment construction.
. . . the first two years of inclusionary zoning in Portland have been a game-theory win-win for housing affordability. The threat of tougher future requirements prompted a whole lot of investment to happen much earlier than it otherwise would have, and new developments, added to those already under construction, have helped deliver a lot more new apartments in Portland.
Portland reaches its Wile E. Coyote moment
Back in 2019, we said that Portland’s apartment market was in the midst of the “don’t look down” portion of its Wile E. Coyote experience. The momentum from pre-IZ housing applications filled the construction pipeline, and led to a steady increase in the number of new apartment completions. And, as we’ve noted, the increase in supply pushed up vacancy rates, and rent increases, which had been in the double digit range in 2015, fell to just 1-2 percent per year, according to Apartment LIst data.
But now, Wile E. Coyote has looked down, and seen nothing holding him up. Data compiled by local economic consulting firm ECONorthwest tracks the number of apartment permits issued in Portland over the past 15 years. It shows the surge in new apartments in 2017 largely holding up in 2018 and 2019, but then plummeting by roughly two-thirds in 2020, from an average of 4,500 new apartments per year to fewer than 1,500.
The ECONW analysis of the building permit data is echoed by other market analysts. Noel Johnson’s website, EnvisionPDXtrends also has data on Portland’s development pipeline showing a diminished volume of new apartment construction activity since the inception of the city’s inclusionary housing requirements.
Similarly Patrick Barry, of the Barry Apartment Report, says that there’s been a sharp fall off in the number of new multi-family building permits applied for in Multnomah County (which contains the City of Portland). New apartment permits in the county have fallen more than 60 percent in the past year, from 5,165 units in 2019 to 2,043 in 2020.
By all measures, Wile E. Coyote is plummeting to the desert floor.
Lean years ahead for apartment deliveries
What’s even more ominous, though, is a parallel decline in new projects in the application process. It takes time (two or even three years) for a project to go from permit application to “for lease,” so much of Portland’s apartment supply for 2022 and 2023 (when, by all accounts the economy is expected to be booming again), is essentially already baked into the cake of filed permit applications. Again, ECONW tracks these new permit applications using city data. Entry into the pipeline is defined as “set up” activity, when an applicant pre-files or files for a new building permit. The number of set ups for apartment units peaked in 2016 and 2017 at slightly more than 6,000 new units per year. Since then, new setups have declined by a third in 2018 and 2019 to about 4,000 per year. In 2020, new setups were about 2,600, less than half their 2017 level.
Just as the past two to three years have produced a kind of temporary win-win of greater apartment deliveries and slowing rent inflation, the next couple of years seem almost certain to have exactly the opposite: declining numbers of new apartments, and rising rents. Already, national forecasters like Zillow are predicting a rapid rebound in demand for urban markets in the post-pandemic period.
Developers will likely wait for the City of Portland to realize the devastating effects of these burdensome IZ requirements and to relax them, or wait for rents to rise enough to support the costs of building new apartments and covering the cost of subsidized units, or simply resign themselves to building smaller, 20-unit buildings, that do much less to expand housing supply and may mean permanently under-utilizing sites that are well-situated to accommodate even greater density.
The Mansard effect
Some of the effects of Portland’s IZ requirements will be quirky and permanent changes to the building stock For example, one key feature of Portland’s inclusionary zoning rule is that it exempts buildings with 20 or fewer apartments from the inclusionary housing requirement, apparently based on the assumption that such a requirement would make these smaller projects uneconomical. But what that exemption has done is to prompt many developers to shift to these smaller buildings. Over the past couple of years, data from the city show that the number of 16-20 unit apartments has (red line) while the number of 21-25 unit buildings (green line) has disappeared. The number of 16-20 unit buildings in 2020 was 143 percent higher than the 2014-2016 average; the number of 21-25 unit buildings was 100 percent lower (zero) than its 2014-16 average.
For reference, as noted above, total apartment completions declined about 67 percent over this time. The 20-unit and under exemption essentially incentivizes developers to build fewer apartments, and because residential structures tend to be long-lived, once a site is built out at 20 units when it could have been 25 or 30 units, that additional housing will be foreclosed for 50 or 100 years.
Also, developers have noted that the inclusionary provision applies on a building-by-building basis, so by dividing a development project up into a series of 20-unit or smaller buildings, a developer can build many apartments without having to comply with the inclusionary requirement. That shows up, for example in a recent development in Northwest Portland, where developer Noel Johnson is building eight five-story residential buildings, with a total of 145 units on a small urban infill site. Again, because each building is 20 or fewer units, the inclusionary requirements don’t apply.
Doing this development as eight different buildings may not the most efficient arrangement, but (to this economist’s eye) the result isn’t unaesthetic. Johnson points out that dividing the project into multiple buildings assures that each apartment is a corner unit (dual-aspect to you English housey types), which make them more desirable. It’s an example of how regulation can prompt innovation.
The effect of restrictive land use regulations on urban architectural form has a long history.
While the Mansard roofs are endearing, they’re a visible and enduring symbol of the power of regulation to alter the housing market. And more significant than the changes to the housing that gets built, are the ways that regulations cause new housing not to be built at all can have even greater, but unfortunately largely unseen impacts. The new apartments that aren’t being built now in Portland will almost certainly lead to higher rents and less affordability in the years ahead—exactly the opposite of the expressed intentions of those who enacted this policy. It’s the apartments that aren’t being built that are the real legacy of inclusionary zoning.
Editor’s Note: Thanks to Mike Wilkerson of ECONorthwest and Noel Johnson of EnvisionPDXtrends.com for sharing their tabulations of Portland apartment permit data. Opinions and analysis presented here solely reflect the views of City Observatory.