Why Permitting Time Has Doubled Since 2000
The cry for more housing is getting louder, but the public process is getting slower.
It can take nearly four years to build a house in San Francisco. But nearly five to build a small, 20-unit multi-family project.
That’s the finding of a working paper that sheds new light on what developers have complained about for what seems like forever: Not only does it take ages to build apartments—during which the market can turn—the time it takes to build varies substantially from city to city.
For example, while San Francisco sits at one extreme, Raleigh sits at the other. It takes just 18 months from start to finish for a standardized 20-unit property in the fast-growing North Carolina capital. In the Bay Area, it takes upwards of 48 long, slow-moving months.
The paper from Princeton University and the University of Chicago’s Booth School of Business found that multi-family permitting time alone doubled nationally in the 14-year period between 2000 and 2023 before ameliorating somewhat in the years since.
It estimates that in 2000, the 20-unit project had a 71 percent probability of approval within six months of applying for clearance from the powers that be. In 2025, though, it had just a 44 percent chance of approval within 180 days.
Despite the renewed focus by big city mayors on housing, the authors write, “there is little systematic evidence on the amount of time required to approve new housing or to complete it … There are no representative estimates of permitting and development durations for new housing, either across cities or over time.”
LISTEN ALSO: What Multifamily Should Watch as ROAD to Housing Takes Effect
So they set out to build a data base for development durations by collecting comprehensive permit records for applications to build new housing across 60 cities from 2000 to 2025, recording the dates at which permits are submitted, issued and completed. Then, they separately estimated the time required to obtain regulatory approval and the total time from permit submission to completion. And finally, they compared these outcomes across cities and over time, adjusting for differences in observed project characteristics and censoring among incomplete projects.
The result is a set of standardized measures that show where housing development is slow, whether it is getting slower and at what point observed delays arise in the development process.
The authors—Leonardo D’Amico of the Booth School and Evan Soltas and Audrey Wang of Princeton—found, as most developers already know, that average project duration for similar projects varies greatly across locales. And they attribute 30 percent to 43 percent of the difference between the slowest and fastest cities to differences in the time to receive a permit rather than in the time to actually build.
At the extreme is Charleston, S.C. It takes about two and a half years to gain approval there for the factitious 20-unit project but less than a year to build it. In Reno, Nev., and Kansas City, Mo., on the other hand, it takes only week for approval. In Philadelphia, Durham, N.C., and Jersey City, N.J., it takes maybe a few months.
Time equals money
The paper also concludes that permitting duration is positively associated with housing costs and negatively associated with housing production and estimates of housing supply elasticities. “An additional year of permitting time for multi-family housing is associated with a 39 percent increase in median rent,” they maintain.
The “strength of this relationship with housing costs rivals that of prior measures of land-use regulation,” the paper says. “Furthermore, it survives controls for neighborhood characteristics that are associated with housing costs, such as per-capita income and population density.”
The authors admit their findings have several limitations. For one thing, some projects studied may complete additional stages of review before applying to their local buildings department. For another, some variables—such as some cities do not issue certificates of occupancy—are not measured identically. And some properties may obtain multiple permits.
Nevertheless, they conclude policy to accelerate the permitting and development of new housing “has proceeded, to a striking degree, without basic facts about the typical durations of permitting and development, their variation across citiesand their evolution over time.”
As a result, they say, “permitting duration has risen steadily over time, whereas project duration is cyclical.”
Noting that permit records are widely available but reported inconsistently across cities, the paper says that standardized reporting, such as through the Census Building Permits Survey or Survey of Construction, could facilitate data collection to document the process and learn where time lines could be improved.
