Research study
Where ADUs are actually being built — and how concentrated that is
The equity question was answered by Terner and the Center for Community Innovation, and we credit them for it rather than restating it as ours. What we add is a current, maintained, tract-level map of one city, with its match rates on the page.
Written by HyreADU Research Desk Primary-source research and data analysis
Audited by HyreADU Research Desk Attribution, geocoding-quality and scope-of-claim review
The finding
ADU permitting in the City of Los Angeles is concentrated, but less extremely than the phrase “ADU hotspots” suggests. Of 39,279 non-supplemental ADU-adding building permits issued between 1 January 2020 and 2026-08-30, 39,279 carry a census tract. They are spread across 1,142 distinct tracts.
The busiest 231 tracts — 20% of the tracts with any activity — hold half the permits; 526 tracts hold 80%. The Gini coefficient across tracts is 0.4731.
The single largest geographic fact is the San Fernando Valley. The North Valley and South Valley area planning commissions together take 53.8% of the city’s ADU permits, against 10.7% for West Los Angeles and 2.6% for the Harbor. Council District 12 alone accounts for 11.1%.
The shape has been stable. The Valley’s combined share has moved within about 5.9% across seven years of issuance. Whatever produces this geography is not a fashion.
Zoning is most of the visible mechanism. 29.9% of these permits are on land zoned R1-1, and the top ten zones account for 67.5%.
An ADU needs a lot with a house and space behind it, and that is a description of the Valley’s post-war single-family tract housing.
On who lives in those places, we defer. The Terner Center for Housing Innovation and the Center for Community Innovation established the demographic finding, and their August 2022 report states that homeowners who build permitted ADUs “are disproportionately White, high-income, and highly educated” and that earlier work of theirs “found that ADUs in the state are disproportionately built in higher-resource neighborhoods.” That is their finding and we cite it as theirs. What we could not do in this edition: join these tracts to American Community Survey demographics ourselves.
The Census API was unreachable from our environment on 2026-09-05. Rather than publish a demographic claim we did not compute, we publish the geography, publish the geocoding match rates that bound it, and say plainly that the join is outstanding.
Read this first
- The equity finding is Terner’s and the Center for Community Innovation’s, not ours
That permitted ADUs skew toward higher-resource neighborhoods and toward White, higher-income and more highly educated homeowners was established by Karen Chapple and colleagues at the Terner Center for Housing Innovation and the Center for Community Innovation.
We quote them, we link them, and we do not restate their conclusion as a HyreADU finding. Our contribution is a maintained, current, tract-level map of one city — and that is all it is.
- We did not run the ACS join in this edition, and we are saying so
The topic this page answers calls for permit records joined to American Community Survey tract demographics.
We could not do it: the Census API was unreachable from our environment on 2026-09-05, and the WebSearch budget for the session was exhausted.
Publishing a demographic result we did not compute — or restating somebody else’s as though we had — would be worse than publishing the geography and the gap.
The join is outstanding and will be added when we can compute it from the Census’s own service.
- One city
Every count and share here is the City of Los Angeles. Los Angeles is the only one of the five jurisdictions this desk holds ADU permit evidence for that publishes a departmental ADU flag together with census tract, council district and area planning commission on every permit row.
This is not a California map and we are not going to imply that it is.
- Geocoding quality is published because it bounds the conclusion
22,991 permits (58.5%) are located by the department to a street address, 8,249 to a parcel identifier and 8,011 to a building; 28 carry no location type. All of these are LADBS assignments and none is a HyreADU geocode.
A parcel- or building-level match is fine for a tract-level count and would not be fine for a block-level one, which is one reason we stop at the tract.
- A permit is not a completion, and a tract is not a neighborhood
These are permits issued, not units finished — our duration study shows how long the gap between those runs. And a census tract is a statistical unit drawn for enumeration, not a place anybody lives in as such. Both matter for how far a reader should push these numbers.
Whose finding this is
The most cited fact about ADU geography in California is that permitted units cluster in wealthier, whiter neighborhoods. That finding is not ours and we have no interest in behaving as though it were.
Fact, quoted. ADUs for All: Breaking Down Barriers to Racial and Economic Equity in Accessory Dwelling Unit Construction, published August 2022 by the Terner Center for Housing Innovation and the Center for Community Innovation — authors Julia Greenberg, Hannah Phalen, Dr Karen Chapple, David Garcia and Muhammad Alameldin — states that “homeowners who build legally permitted (or formal) ADUs—like homeowners in the state overall—are disproportionately White, high-income, and highly educated,” and that “[a] previous study authored by the Terner Center and Center for Community for Innovation found that ADUs in the state are disproportionately built in higher-resource neighborhoods.”
The same report also records a distinction that is easy to lose and important here: “unpermitted ADUs are more highly concentrated in communities of color and lower income areas.” A permit file, by construction, cannot see those units at all.
Any map of permits is a map of the formal sector, and the informal sector it omits is not distributed the same way.
What we add, and only this. A current, reproducible, tract-level count for one city, refreshed from the department’s own live extract, with the geocoding quality published on the page. Terner’s work is a study; this is a maintained measurement. Those are different jobs and we are doing the second one.
By council district
The geography most likely to matter politically, and the one least often published. Every ADU permit in the frame carries a council district on the department’s own record.
| Council district | ADU permits | Share of city total | Ratio to the least active district |
|---|---|---|---|
| District 12 | 4,373 | 11.1% | 4.28× |
| District 3 | 4,106 | 10.5% | 4.02× |
| District 2 | 3,806 | 9.7% | 3.72× |
| District 7 | 3,426 | 8.7% | 3.35× |
| District 6 | 3,129 | 8.0% | 3.06× |
| District 11 | 2,987 | 7.6% | 2.92× |
| District 4 | 2,932 | 7.5% | 2.87× |
| District 8 | 2,748 | 7.0% | 2.69× |
| District 5 | 2,447 | 6.2% | 2.39× |
| District 10 | 2,213 | 5.6% | 2.17× |
| District 13 | 1,757 | 4.5% | 1.72× |
| District 14 | 1,489 | 3.8% | 1.46× |
| District 9 | 1,454 | 3.7% | 1.42× |
| District 15 | 1,390 | 3.5% | 1.36× |
| District 1 | 1,022 | 2.6% | 1.00× |
HYRE calculation on LADBS resource pi9x-tg5x, 39,279 non-supplemental ADU-flagged permits issued 2020 to 2026-08-30, retrieved 2026-09-05. District boundaries are as recorded on the permit at the time of issue; the city redistricted during this window and the department’s field is not retroactively restated.
The busiest district issued 4.3 times as many ADU permits as the quietest.
That is a large gap between fifteen districts of roughly equal population, and it is the single most policy-legible number on this page.
It is a description, not a criticism: districts differ enormously in how much single-family land with rear-yard space they contain.
How concentrated is it, really?
“Concentrated” gets used loosely, so here is the measurement. Across the 1,142 Los Angeles census tracts that recorded at least one ADU permit in the window, the busiest 231 hold half of all permits and 526 hold 80%.
The top ten tracts hold 4.0%; the top hundred hold 27.5%. The Gini coefficient across tracts is 0.4731.
HYRE analysis. That is real concentration and it is nothing like extreme.
A Gini of 0.4731 across tracts is roughly what you would expect from a process driven by housing stock and lot geometry rather than by a handful of hot streets.
The busiest single tract in the city recorded 190 permits over roughly six and a half years — about 28 a year — while the median active tract recorded 26 in total.
There is no tract in Los Angeles where ADU construction is happening at anything like industrial scale.
The denominator matters and we chose the conservative one. We measure concentration across tracts that recorded at least one permit.
Including every tract in the city — many of which are commercial, industrial, or otherwise incapable of hosting an ADU — would produce a much higher Gini and a much more dramatic sentence.
It would also be a worse measurement, because a tract with no single-family housing is not a tract that declined to build ADUs.
What the concentration does not tell you. Whether the busy tracts are wealthy, poor, majority-White or majority-Latino. That is the ACS join, and it is the join we could not run this time. The tracts themselves are listed below so that anybody who wants to run it can, against the same identifiers.
The fifteen busiest census tracts
Published as identifiers so the analysis can be replicated or extended by anyone with an ACS extract. These are the department’s own tract codes on the permit record.
| Census tract | ADU permits | Share of city total | Multiple of the median active tract |
|---|---|---|---|
1393.01 | 190 | 0.5% | 7.3× |
1316.00 | 189 | 0.5% | 7.3× |
2760.00 | 175 | 0.4% | 6.7× |
1112.02 | 167 | 0.4% | 6.4× |
1210.10 | 149 | 0.4% | 5.7× |
1220.00 | 146 | 0.4% | 5.6× |
1352.02 | 146 | 0.4% | 5.6× |
1031.02 | 146 | 0.4% | 5.6× |
1097.00 | 143 | 0.4% | 5.5× |
1330.00 | 140 | 0.4% | 5.4× |
1098.00 | 139 | 0.4% | 5.3× |
1112.06 | 136 | 0.3% | 5.2× |
1237.00 | 131 | 0.3% | 5.0× |
2110.00 | 130 | 0.3% | 5.0× |
1112.01 | 130 | 0.3% | 5.0× |
HYRE calculation on LADBS resource pi9x-tg5x, retrieved 2026-09-05. Tract codes are as recorded in the department’s ct field. Median active tract: 26 permits.
We publish the codes and not a characterization of the neighborhoods, deliberately. Naming a tract as “the ADU hotspot” invites a story about the people who live there, and we have not measured anything about the people who live there.
By community plan area
The geography Angelenos actually recognize. Community plan areas are the department’s own field and map roughly onto neighborhoods.
| Community plan area | ADU permits | Share of city total |
|---|---|---|
| Canoga Park-West Hills-Winnetka-Woodland Hills | 2,902 | 7.4% |
| South Los Angeles | 2,508 | 6.4% |
| Reseda - West Van Nuys | 2,328 | 5.9% |
| Northeast Los Angeles | 2,225 | 5.7% |
| West Adams - Baldwin Hills - Leimert | 2,141 | 5.5% |
| North Hollywood - Valley Village | 2,068 | 5.3% |
| Van Nuys - North Sherman Oaks | 2,040 | 5.2% |
| Wilshire | 1,885 | 4.8% |
| Sun Valley - La Tuna Canyon | 1,647 | 4.2% |
| Southeast Los Angeles | 1,647 | 4.2% |
| Mission Hills - Panorama City - North Hills | 1,603 | 4.1% |
| Granada Hills - Knollwood | 1,377 | 3.5% |
| Arleta - Pacoima | 1,284 | 3.3% |
| Hollywood | 1,229 | 3.1% |
| Palms - Mar Vista - Del Rey | 1,229 | 3.1% |
| Sunland-E LaTuna Cyn-Lakevw Ter-Shadow Hls-Tujunga | 1,047 | 2.7% |
| Northridge | 958 | 2.4% |
| Encino - Tarzana | 939 | 2.4% |
| Chatsworth - Porter Ranch | 885 | 2.3% |
| Sylmar | 878 | 2.2% |
Top twenty of the department’s community plan areas by ADU permit count. HYRE calculation on pi9x-tg5x, 39,279 permits, retrieved 2026-09-05.
The top twenty plan areas hold 83.6% of the city’s ADU permits. Eight of the top ten are in the Valley.
The zones ADUs are permitted on
The most mechanical explanation of the map. An ADU needs a lot that already has a house on it, and the zoning field says which lots those are.
| Zone | ADU permits | Share of city total | What it is |
|---|---|---|---|
R1-1 | 11,730 | 29.9% | The standard single-family residential zone. The core ADU zone by a very wide margin. |
RS-1 | 4,276 | 10.9% | Suburban single-family — larger minimum lots. |
R1-1-RIO | 2,180 | 5.6% | Single-family within a River Improvement Overlay district. |
R2-1 | 1,618 | 4.1% | Two-family residential. |
RA-1 | 1,599 | 4.1% | Suburban / agricultural residential, largest lots. |
R1V2 | 1,197 | 3.0% | A single-family variation zone under the city’s re:code framework. |
R3-1 | 1,139 | 2.9% | Multiple dwelling. |
RE11-1 | 970 | 2.5% | Residential estate, 11,000 sq ft minimum lot. |
RD1.5-1 | 915 | 2.3% | Restricted density multiple dwelling. |
R1-1-CUGU | 887 | 2.3% | Single-family with a Clean Up Green Up supplemental district. |
Top ten zones by ADU permit count, from the department’s zone field. HYRE calculation on pi9x-tg5x, retrieved 2026-09-05.
R1-1 alone carries 29.9% of the city’s ADU permits, and the ten zones above carry 67.5% between them. HYRE analysis: this is the mechanism behind the map.
California’s ADU statutes attach the right to a lot with a proposed or existing single-family or multifamily dwelling. Where that housing type is, ADUs are.
Nothing about the geography of ADU permitting needs a preference explanation until the housing-stock explanation runs out, and in Los Angeles it does not run out.
Has the map changed?
Share of the city’s ADU permits taken by each area planning commission, by year of issue. If the geography were shifting, it would show here.
| Area planning commission | 2020 | 2021 | 2022 | 2023 | 2024 | 2025 | 2026* |
|---|---|---|---|---|---|---|---|
| South Valley | 30.8% | 30.9% | 28.0% | 28.4% | 29.2% | 28.7% | 27.8% |
| North Valley | 26.1% | 25.4% | 23.0% | 24.5% | 26.0% | 25.7% | 23.9% |
| South Los Angeles | 14.7% | 15.5% | 18.5% | 16.3% | 16.3% | 15.0% | 14.4% |
| West Los Angeles | 9.2% | 10.0% | 10.1% | 10.1% | 9.2% | 12.1% | 14.5% |
| East Los Angeles | 9.0% | 8.1% | 8.6% | 8.9% | 7.9% | 8.5% | 8.8% |
| Central | 8.4% | 8.1% | 9.1% | 8.8% | 8.3% | 7.6% | 7.9% |
| Harbor | 1.8% | 2.0% | 2.8% | 3.1% | 3.0% | 2.4% | 2.6% |
HYRE calculation on pi9x-tg5x, retrieved 2026-09-05. *2026 is a part year, permits issued to 2026-08-30. Columns sum to 100% within rounding.
No commission has moved by more than a few percentage points in seven years. The most notable single movement is West Los Angeles, which was 9.2% of the city in 2020 and 12.1% in 2025.
That is a real drift and it is small. The stability is the finding: seven years, four statutory amendments and two recodifications, and the map barely moved.
Method
One public resource, the department’s own geographies, and no HyreADU geocoding at all.
- 1 Download the whole resource
All 409,619 rows of LADBS Socrata resource
pi9x-tg5x, downloaded 2026-09-05. The department’srefresh_timefield runs to 2026-08-30. - 2 Use the department’s ADU flag and drop supplementals
adu_changed > 0orjunior_adu = 1, with supplemental permits removed by permit-number middle segment. 39,279 permits issued from 2020. This is the same frame as our duration study and our conversion study, on purpose. - 3 Take the geography from the department, not from a geocoder
Area planning commission (
apc), council district (cd), community plan area (cpa), census tract (ct), zone (zone) and ZIP are all fields LADBS publishes on the permit row.We ran no geocoding of our own, which removes an entire class of error and also means our coverage is exactly the department’s coverage: 1,142 distinct tracts and no unmatched rows on the tract field.
- 4 Publish the location-quality field
The department also records how it located each permit:
ADDRESSfor 22,991,PIN(parcel) for 8,249,BUILDINGfor 8,011, and nothing for 28. We publish that split because it is what bounds how fine a geography the data supports. It supports tracts. It does not support blocks. - 5 Measure concentration on the conservative denominator
The Lorenz curve and Gini are computed across tracts that recorded at least one ADU permit, not across all tracts in the city. Including tracts that cannot host an ADU would inflate the concentration figure and would measure the city’s land use rather than ADU behavior.
- 6 And stop where the data stops
No demographic variable is computed anywhere on this page. Where a demographic statement appears, it is quoted from Terner and the Center for Community Innovation with an author list and a date, and it is theirs.
What is outstanding on this page
The ACS join. The tract identifiers are published above and the intent is to join them to American Community Survey tables — median household income, tenure, race and ethnicity, units in structure, year built — and publish permits per owner-occupied single-family unit by tract decile.
We could not reach the Census API from this environment on 2026-09-05 and we will not publish a demographic result we did not compute.
The other four jurisdictions. San Francisco, Sacramento, San José and unincorporated Marin all publish permit extracts, and none of them publishes an ADU flag with a tract on the same row.
Extending this map beyond Los Angeles means classifying ADUs from free text in four more feeds, and doing that badly would be worse than not doing it.
Completions. This is a map of permits. A map of finished units would be a better answer to the question and Los Angeles is the only jurisdiction we have found that publishes a completion date at all. That is a future edition, not a footnote.
Questions
Where are most ADUs built in Los Angeles?
Why the Valley?
R1-1 — standard single-family residential — and the top ten zones account for 67.5%. An ADU needs a lot that already has a house on it and either rear-yard space or a convertible accessory structure. Post-war Valley tract subdivisions are close to the ideal case, and the Valley also has the city’s highest garage-conversion share.Are ADUs concentrated in a few neighborhoods?
Do ADUs get built in wealthier, whiter neighborhoods?
Which council district issues the most ADU permits?
Has the geography changed as the ADU laws changed?
Does this include unpermitted ADUs?
Can I get the underlying data?
pi9x-tg5x on data.lacity.org, which is public and downloadable. The frame is adu_changed > 0 OR junior_adu = 1, supplementals excluded by permit-number middle segment, issue year 2020 or later. Every geographic field used — apc, cd, cpa, ct, zone — is published by the department on the same row.Written and audited by
HyreADU Research Desk
Primary-source research, data analysis and fact checking
We are a research desk, not a builder. We read the permit extract, the statute, the HCD return or the fee schedule ourselves, and publish each figure with its source and retrieval date.
Where a number cannot be traced to a primary source, we leave it out and say what we could not verify. Our store-based claims cover California only.
- CA
- the only state this desk will make store-based claims about
- 5
- jurisdictions with extracted ADU permit evidence
- 735
- CSLB-verified companies in the California store
- 0
- national claims from a one-state store
How this desk works
- Primary sources only. Permit counts come from the city or county that issued the permit. Production counts come from HCD’s Annual Progress Report. Rents come from HUD or the Census. We do not cite an article that cites a source; we download the source and compute the figure ourselves.
- This is a California site. The company store is 734 California firms and one New Mexico firm. Permit evidence exists for five named jurisdictions: Los Angeles, San Francisco, Sacramento, San José and unincorporated Marin. A number from that store is titled to those places, never to the United States.
- A permit is not a completion, and a license is not an ADU grade. California licenses no ADU classification. Being named on an ADU permit is evidence of engagement in that jurisdiction, not of quality, completion, or work anywhere else. Owner-builder permits are excluded from contractor counts.
- Calculation is labeled as calculation. Figures we derive are never presented as something HCD, HUD, the Census or a city published. Terner Center research is cited as Terner’s, never restated as ours.
- We do not design, permit or build ADUs, and we take no payment for placement, ranking or a favorable mention. Pages that look like rankings are not: they publish public-record counts and let the reader decide.
- Nothing here is legal, tax or financial advice. Zoning, underwriting and appraisal practice vary by jurisdiction, lender and appraiser. The useful next step on a specific lot is the planning counter and a licensed professional.
Data as of LADBS Socrata resource pi9x-tg5x (409,619 rows) downloaded and analyzed 2026-09-05; the department’s own refresh_time field runs to 2026-08-30. Terner Center and Center for Community Innovation report retrieved and text-extracted 2026-09-05.. Authorship on this site is organizational: the analysis belongs to the desk rather than to a named individual, and we do not publish credentials we do not hold.
Our editorial policy sets out how we source, date and correct what we publish.
Sources & retrieval dates
- City of Los Angeles Department of Building and Safety — Building Permits Issued from 2020 to Present (Socrata resource pi9x-tg5x) , Full resource download, 409,619 rows, retrieved 2026-09-05. Geographic fields used: apc (area planning commission), cd (council district), cpa (community plan area), ct (census tract), zone, zip_code, lat, lon and type_lat_lon. ADU frame from the departmental adu_changed and junior_adu flags. No HyreADU geocoding was performed. Retrieved 2026-09-05.
- City of Los Angeles open data portal — dataset landing page for pi9x-tg5x , Publisher description and column definitions, used to confirm that the geographic fields are departmental assignments rather than derived values. Retrieved 2026-09-05.
- Terner Center for Housing Innovation and Center for Community Innovation — “ADUs for All: Breaking Down Barriers to Racial and Economic Equity in Accessory Dwelling Unit Construction”, August 2022 , Authors: Julia Greenberg (Research Manager, CCI); Hannah Phalen (Graduate Student Researcher, Terner Center); Dr Karen Chapple (Director, School of Cities, University of Toronto; Professor Emerita, City and Regional Planning, UC Berkeley); David Garcia (Policy Director, Terner Center); Muhammad Alameldin (Policy Associate, Terner Center). Retrieved as a PDF and text-extracted on 2026-09-05. Quoted on this page: “homeowners who build legally permitted (or formal) ADUs—like homeowners in the state overall—are disproportionately White, high-income, and highly educated”; “[a] previous study authored by the Terner Center and Center for Community for Innovation found that ADUs in the state are disproportionately built in higher-resource neighborhoods”; “unpermitted ADUs are more highly concentrated in communities of color and lower income areas”. This is their finding and it is cited as theirs. Retrieved 2026-09-05.
- Terner Center for Housing Innovation — “ADUs in the Coastal Zone” , Terner’s measurement of average days from planning application to building permit inside and outside the Coastal Zone in Los Angeles, Orange and San Diego Counties, 2018–2022. Cited as an example of a geographic effect on ADU permitting that this page does not itself measure, and as Terner’s work rather than ours. Retrieved 2026-09-05.
- California Government Code § 66314 — the standards a local ADU ordinance may impose , The statutory attachment of the ADU right to a lot with a proposed or existing dwelling, which is the mechanism behind the zoning concentration reported on this page. Retrieved 2026-09-05.
A city-level map does not tell you about your lot
A high ADU permit count in your area says nothing about whether your own lot can take one. Setback, lot coverage, height, access and the shape of the yard decide that, and they decide it lot by lot.
HyreADU does not design, permit or build accessory dwelling units, and takes no payment for placement or a favorable mention.
Every count and share on this page is the City of Los Angeles only, computed from a named public permit extract retrieved on 5 September 2026; it is not a California or national statistic.
These are permits issued, not units completed. No demographic variable is computed anywhere on this page. The findings that permitted ADUs skew toward higher-resource neighborhoods and toward White, high-income and highly educated homeowners are the Terner Center for Housing Innovation’s and the Center for Community Innovation’s, are quoted with their authors and date, and are not restated as HyreADU research; an American Community Survey join was not run in this edition because the Census API was unreachable from our environment on the retrieval date.
A permit file cannot see unpermitted units, which the same authors record as being more concentrated in communities of color and lower-income areas. This page is informational and is not legal advice.