Statistics
ADU property value statistics
23 of 52 figures are HyreADU calculations. What has actually been measured, what the appraisal guides actually require, and the arithmetic that does not depend on a coefficient travelling somewhere it was never estimated.
Written by HyreADU Research Desk Primary-source research and data analysis
Audited by HyreADU Research Desk Claim-transfer and provenance-class audit
How to use this page
There is no defensible national ADU value-add percentage, and this page does not publish one. The single peer-reviewed California measurement is Brueckner and Thomaz, Real Estate Economics 52(3), 2024: on City of Los Angeles tax-assessor parcels from 2013 to 2019, ADU presence is associated with a 7–9 per cent higher assessed value or selling price — 7.9% in the column-1 all-property specification, 7–8 per cent with block-group fixed effects, and “almost 9 per cent” on the just-sold subset where the estimate is closest to a sales-price effect.
Those are ordinary least squares; the authors’ instrumental-variable strategy was unsuccessful.
The 20–30 per cent figure that dominates search results cannot be traced to a study with a method, and where it is attributed to the Terner Center, that center’s catalog does not support the attribution.
Cost, appraised market value, assessed value and a capitalised rent are four different numbers, and the category’s whole failure is collapsing them. On contractor-named Los Angeles ADU permits the median declared valuation is $40,000; the indexed Los Angeles County construction median is $153,567; the appraised contributory value is whatever an appraiser concludes from a comparable set the guides themselves describe as thin.
Here is the arithmetic that does not require the coefficient to travel. The indexed statewide construction median of $230,351 is about 30.3% of the California median home value of $759,500 — and about 63.9% of the United States median of $360,600, which is why a California cost figure cannot be read as a national one.
And, inverting the Los Angeles coefficient where both instruments actually exist: a 9.0% differential would equal the indexed Los Angeles County construction median only on a host property worth about $1,706,300.
23 of 52 figures on this page are HyreADU calculations (23 hyreadu calculation, 10 peer-reviewed, 14 agency published, 5 trade / survey estimate). Nothing here is financial, tax or legal advice, a guaranteed return, or a reason to build.
Read the provenance, not just the number
ADU property value is the topic where the gap between what circulates and what has been measured is widest, so this digest uses four classes rather than three, and the extra class is the point.
Peer-reviewed measurement — a published paper with a sample, a method and a stated identification strategy.
There are three on this page and only one of them measures a California sale-price differential.
Agency published — the current appraisal guide topic, the current fact sheet, the Handbook, a Census table, a named county assessor.
HyreADU calculation — the ratios between two published figures, the inversion of a published coefficient inside the city it was estimated in, and the permit-valuation distributions.
Trade / survey estimate — the 2021 owner survey’s occupancy and cost findings, and, labeled explicitly as circulation rather than evidence, the untraceable percentage itself.
The source value study computes almost nothing on purpose. Its stated method note says: “This study does not compute a HyreADU value-add index. The finding is that the inputs will not support one.” This digest holds that line exactly.
It does not interpolate a statewide percentage, it does not apply the Los Angeles coefficient to a California median home value, and it does not publish a dollar “value added” for any unit.
What it does add is arithmetic that survives the transfer problem. Two published figures divided by one another do not require anything to travel: an indexed construction median as a share of a published median home value is true wherever both are true.
And an inversion — what host value would be required for a measured differential to equal a measured cost — is a statement about the internal consistency of two Los Angeles figures, not a prediction about Fresno. Both are labeled every time they appear.
The figures most worth knowing
An appraisal is an opinion of market value for a named intended use. A county assessment is a tax base. Neither is a forecast of what a buyer will pay you, and neither is a reason to build.
What has actually been measured
Three published papers, and the untraceable band, set against each other. The last column is the one that matters: what each result will and will not carry.
| Source | Geography and sample | What it measured | Headline result | Class | Why it is not a national percentage |
|---|---|---|---|---|---|
| Brueckner and Thomaz, Real Estate Economics 52(3), 2024, 885–907 | City of Los Angeles parcels, 2013–2019 tax-assessor panel | OLS regressions of assessed value — and, for just-sold parcels, selling price — on ADU presence, with Proposition 13 structure in the specification | ADU presence associated with 7–9% higher assessed value or selling price. Column 1 of the all-property specification: 7.9%. Just-sold subset: “close to 9%.” All-property with block-group fixed effects and lagged ADU: 7–8%. | Peer-reviewed measurement | One city, on a panel that predates the post-2020 boom. OLS treats ADU presence as exogenous, and the authors’ instrumental-variable attempt was unsuccessful. For parcels that have not recently sold, the assessed increment “may instead rely on the assessor’s estimate of the construction cost,” which is not a market price. |
| Brown and Watkins, The Appraisal Journal, Spring 2012 | Portland, Oregon; 14 properties with permitted ADUs | Income-capitalisation valuations compared with actual sale prices | Income approach 7.2% or 9.8% above actual sale prices, depending on formula. ADUs contributed 25% or 34% of each property’s income-approach appraised value. | Peer-reviewed measurement | n = 14. The 25–34 per cent is the ADU’s share of an income-approach value, not a market premium over otherwise identical houses, and not California. The paper’s argument is that sales comparison may have been undervaluing the income stream — a method claim, not a resale statistic. |
| Adomatis, The Appraisal Journal 89(4), Fall 2021 (Appraisal Institute) | Practice paper: agency guideline comparison and data-search method. No sale-price sample. | How an ADU appraisal assignment is scoped, how comparables are found, how agency rules differ | No percentage. The operational finding is that three comparable ADU sales were “extremely difficult to find,” and that MLS and public-record coding of ADUs remains unreliable. | Peer-reviewed measurement | It is the professional body’s method paper, not a measurement of value add. Its 2021 exhibit of agency rules has since been overtaken on the rental-income rows. |
| Terner Center and Center for Community Innovation, 2020–2024 | California ADU-owner survey of 752 homeowners who built; HCD Annual Progress Report data; regional cost estimates | Who builds, what it cost, what it rents for, who lives there | 51% of new ADUs generating rental income; 16% no-cost housing for a relative; 8% short-term rentals. Regional construction-cost medians. No sale-price premium. | Trade / survey estimate | Occupancy and cost research. Citing it for a resale percentage is a category error, and it is the single most common misattribution in the field. |
| Realtor and builder pages, various | Unstated | Unstated | “ADUs add 20–30%,” sometimes 25–35% in the Bay Area, often attributed to Terner or to “studies.” | Trade / survey estimate | No primary paper, table, geography or method located on retrieval. The highest published Los Angeles coefficient is 0.45 times the low end of that band. Recorded as circulation. Not shipped as a fact. |
Peer-reviewed and university measurements of ADU value, set against the untraceable percentage that dominates search results. Retrieved 2026-09-05. For Brueckner and Thomaz we quote the published abstract and the results-section figures the publisher makes available; a paywalled appendix is not reconstructed.
HyreADU analysis. Seven to nine per cent in one city, on a 2013–2019 panel, with an identification strategy the authors themselves could not make work, is a real number and a modest one. It is also the best number the category has.
A reader who wants a statewide or national percentage does not have one, and interpolating between a Los Angeles assessed-value coefficient and a Portland income-approach share would be the same sin as the blogs, dressed in citations.
Four numbers that are not the same
Almost every ADU return-on-investment page treats construction cost, appraised market value, the tax assessment and a capitalised rent as interchangeable.
They are four different objects with four different definitions, four different decision-makers and four different dates.
On the same Tuesday, on the same property, they can sit at four different levels — and that is the system working, not a paradox.
| The number | What it actually is | Who decides it | A figure we can source | Class |
|---|---|---|---|---|
| Cost to build | The cash outlay to a permitted, occupiable unit: hard costs, site work, utilities, fees, design and contingency. A denominator, not a value. | The market, through bids | Indexed medians of $153,567 (Los Angeles County), $230,351 (statewide) and $276,421 (detached) | HyreADU calculation |
| Declared permit valuation | A figure written on a building-permit application, used to compute the fee. Systematically pulled downward, and in some cities replaced by a published table. | The applicant, or a city valuation table | Median $40,000 on contractor-named City of Los Angeles ADU permits — about 26% of the indexed Los Angeles cost median | HyreADU calculation |
| Appraised market value (the ADU’s contributory value) | An appraiser’s opinion of what the whole property would sell for, with the ADU’s contribution isolated on its own line of the sales-comparison grid, for a named intended use. | An appraiser, under USPAP and an agency guide | No defensible general figure exists. No GSE publishes a factor; the closest measurement is a 7–9 per cent Los Angeles differential on 2013–2019 assessor data. | Peer-reviewed measurement |
| Assessed value (the tax base) | A new base year value enrolled for the newly constructed portion only, under Revenue and Taxation Code § 71. The rest of the property keeps its existing factored base year value. | The county assessor | Statutorily the market value added; in several counties’ own words, in practice often derived from construction cost | Agency published |
| Capitalised rent | A value inferred from rent — a gross-rent multiplier, or net operating income over a capitalisation rate. | An appraiser, where a substantial rental market exists | Permitted as support, forbidden as the sole indicator of market value (Fannie Mae B4-1.3-10). And 16 per cent of new California ADUs have no rent at all. | Agency published |
Sources as named in each row and in the source list. The cost and declared-valuation figures are HyreADU calculations from the cost study; the appraisal and assessment rows are the guides’ and the statute’s.
A worked illustration of why they refuse to collapse. A $320,000 invoice, a $180,000 contributory appraisal for want of comparable sales, a $240,000 assessed increment derived from a cost manual, and a $0 net operating income because a parent lives in the unit, is not a contradiction.
It is four instruments measuring four things. Any page that reports one of them as “what your ADU is worth” has already thrown away the other three.
The size of the question, without a coefficient
The cost-against-value table below involves no coefficient and no transfer. It divides one published figure by another: an indexed ADU construction median by the Census Bureau’s median value of owner-occupied housing units.
It cannot tell you what an ADU adds. It can tell you how large a decision this is relative to the asset it is attached to — which is the context missing from every percentage in circulation.
| Indexed construction median | Amount | As a share of the California median home value | As a share of the US median home value | Class |
|---|---|---|---|---|
| Garage conversion or expansion | $138,211 | 18.2% | 38.3% | HyreADU calculation |
| Los Angeles County, all types | $153,567 | 20.2% | 42.6% | HyreADU calculation |
| Statewide, all types | $230,351 | 30.3% | 63.9% | HyreADU calculation |
| Detached, newly built | $276,421 | 36.4% | 76.7% | HyreADU calculation |
| San Francisco Bay Area, all types | $272,582 | 35.9% | 75.6% | HyreADU calculation |
Construction medians are the 2018–19 California ADU owner survey indexed to August 2026 on the DGS California Construction Cost Index — a HyreADU indexation, not a 2026 survey and not a bid.
Home values are ACS 2024 1-year table B25077: California $759,500 (margin of error ± $3,644), United States $360,600 (± $618).
B25077 is the median value of owner-occupied housing units as reported by occupants — all housing, not a sale price, not an assessed value, and not ADU-specific.
The last column is on this page for one reason. California’s median home value is 2.11 times the national figure.
A 7–9 per cent differential in Los Angeles is a completely different dollar amount from a 7–9 per cent differential in a $200,000 market — and no measurement of the latter exists.
Anyone taking a California-derived percentage or a California-derived cost national is making the error this column is here to make visible.
The inversion: what would have to be true
Read the framing before the numbers, because the framing is the whole safeguard. This table does not predict a value-add.
It asks a conditional question inside the one city where both instruments exist: if the Brueckner and Thomaz coefficient described a property, how valuable would that property have to be for the differential to equal the indexed cost of building the unit?
Both inputs are Los Angeles — a Los Angeles coefficient and a Los Angeles County indexed construction median. Nothing is being transferred to another market, and no dollar value-add is being asserted for any property.
| Coefficient, as published | Value | Host value for the differential to equal the indexed LA County median | Host value to equal the indexed detached median | Class |
|---|---|---|---|---|
| all-property specification, column 1 | 7.9% | $1,953,779 | $3,516,807 | HyreADU calculation |
| all-property band, low | 7.0% | $2,193,814 | $3,948,871 | HyreADU calculation |
| all-property band, high | 8.0% | $1,919,588 | $3,455,263 | HyreADU calculation |
| just-sold subset | 9.0% | $1,706,300 | $3,071,344 | HyreADU calculation |
HyreADU calculation: indexed construction median ÷ published coefficient. Coefficients are Brueckner and Thomaz’s (Real Estate Economics 52(3), 2024), from City of Los Angeles assessor parcels 2013–2019.
Construction medians are the 2018–19 owner survey indexed to August 2026, from the cost study.
The detached column applies a statewide type median to a Los Angeles coefficient and is the weaker of the two; it is included as a sensitivity, not as a second finding.
What this does and does not establish. It establishes that the two best-sourced Los Angeles figures the category has are not consistent with an ADU recovering its build cost through a resale differential on an ordinary house — the break-even host value at the highest published coefficient is about $1,706,300, some 2.25 times California’s median home value.
It does not establish that any particular property would see a 7–9 per cent lift, that the coefficient applies outside Los Angeles, or that the indexed cost median is what a given unit costs.
It is an internal-consistency check on two published numbers, and it is offered as one.
Why the appraisal evidence is thin, in the guides’ own words
The clearest evidence that ADU value is hard to measure is the accommodation the agencies themselves wrote into their rules. Guides do not add flexibility for fun. Each of the following sentences exists because the comparable sales were not there.
Fannie Mae B4-1.3-05 requires the appraisal to include a description of the ADU and an analysis of any effect it has on the value or marketability of the subject — and then provides that “an aged settled sale will qualify as a comparable, and an active listing or under contract sale will qualify as a supplemental exhibit to show marketability.” For an illegal ADU it requires typicality to be demonstrated “through an analysis of at least two comparable sales with the same non-compliant zoning use,” with a minimum of three settled sales, and permits aged sales where recent ones are unavailable.
Freddie Mac is more prescriptive and, in being so, more revealing: at least one comparable sale with an ADU “when available,” and if a recent one is not available in the subject’s market area, “the appraiser can use an older sale with an ADU from the subject’s Market Area or a sale with an ADU from a competing Market Area.” Where ADU rent is used to qualify, three comparable rentals are required, at least one of them a rented ADU, and automated collateral evaluation is not acceptable.
The Appraisal Institute’s own peer-reviewed practice paper says it outright: before the 2020 guideline update, “it was next to impossible for appraisers to meet federal Agency appraisal guidelines, because three sales of comparable ADU properties were extremely difficult to find in the local market.” Its data-search section has not dated — “All MLSs do not have searchable fields to identify ADU properties. Public records also may not separate the ADU living area from the main structure” — and it lists more than twenty search terms, from granny flat to ohana unit to backyard cottage, because that is what the work actually looks like.
The one contemporaneous count we have is consistent with all of it.
Zillow figures reported in 2023 for a three-month Los Angeles window recorded 642 sold houses of about 2,000 square feet, of which 106 listings included an ADU — 16.5%, and that is a listing flag rather than a verified permitted unit.
In a market the size of Los Angeles, over three months, that is a thin set from which to isolate a contributory value.
HyreADU analysis. The instruction to the appraiser is correct: value the unit as a market feature, separately from the main house, against comparables that look like it.
The consequence is that two competent appraisers in the same ZIP code, given the same thin set, can reach different contributory values and both be inside professional standards. A percentage that collapses that variation is not a measurement of ADU value.
It is a measurement of how confident somebody was willing to sound.
What the guides actually require
The rules are worth reading directly, because they are frequently reported as if they contained a valuation factor. They do not. They contain an assignment, a set of comparable minima, and — separately, and constantly confused with value — an underwriting cap on rental income.
| Rule | Fannie Mae | Freddie Mac | FHA | Class |
|---|---|---|---|---|
| Is a contributory-value factor published? | No. The appraiser must describe the ADU and analyze its effect on value and marketability. | No. The report must conclude whether an adjustment is supported for the ADU — “supported”, not assumed. | No. The appraiser measures contributory value using one or more of the three approaches. | Agency published |
| ADU area in the primary dwelling’s gross living area | No, except where the unit is contained within the primary dwelling with interior access and above grade (B4-1.3-05) | No. Finished square footage of the ADU is reported separately (§ 5601.2) | No. ADU living area is not included in primary GLA | Agency published |
| Comparable-sale instruction | Aged settled sale qualifies; a listing is a supplemental exhibit only. Illegal ADU: at least two comps with the same non-compliant use, three settled sales in the report. | At least one comparable sale with an ADU when available; otherwise an older sale, or one from a competing market area. | Sales comparison required for the property; contributory value of the ADU from one or more of the three approaches. | Agency published |
| Cost or income as the sole indicator of value | Forbidden (B4-1.3-10), manufactured homes excepted | Sales comparison is the framework; automated collateral evaluation waived when ADU rent is used | Three approaches available for the ADU line; sales comparison still required for the property | Agency published |
| ADU rent toward qualifying income (underwriting, not value) | 30 per cent of qualifying income, one existing ADU, purchase or limited cash-out only | 30 per cent, 75 per cent of the lease, three rent comps with at least one rented ADU | 30 per cent of effective income; not on a cash-out refinance | Agency published |
| Must the ADU be legal? | Not always — an illegal use can be eligible with insurance confirmation and extra comparables | Legal, legal non-conforming or no zoning; rent from an illegal ADU cannot qualify | Yes — legal status is required for eligibility | Agency published |
Agency ADU valuation and income rules as retrieved 2026-09-05, from the current Selling Guide topics, the February 2026 Freddie Mac fact sheet and Guide § 5601.2, and HUD Handbook 4000.1. Lender overlays can be tighter than any of these.
The 30 per cent cap is not a 30 per cent value-add, and the confusion is common enough to be worth stating twice. It is a credit-risk governor on how much of an existing unit’s documented rent may be counted toward a borrower’s debt-to-income ratio.
It says nothing about what the unit is worth, and it applies only to units that already exist. The financing digest works that cap out at real rents and incomes.
Figures we will not repeat
- “ADUs add 20–30 per cent to your property value”
We searched Terner Center working papers and reports, the Center for Community Innovation ADU series, The Appraisal Journal, Real Estate Economics and the GSE selling guides for a 20–30 or 25–35 per cent sale-price premium attached to a method.
We did not find one. What we found was the figure travelling without a citation, and in at least one builder-adjacent post attributed to the Terner Center for the Bay Area and Los Angeles County — an attribution that center’s own catalog does not support.
The highest published Los Angeles coefficient is 0.45 times the low end of that band. We will not launder it by putting it in a chart, even struck through.
- The Portland 25–34 per cent figure as a resale premium
Brown and Watkins (2012) did publish 25 per cent and 34 per cent, and neither is a market premium.
They are the ADU’s share of an income-approach value, on 14 Portland properties, in a paper whose argument is that sales comparison may have been undervaluing the income stream.
Mixing that figure with a California cost survey and printing “ADUs add 25–35 per cent” is precisely how a method result becomes a sales pitch.
- The Los Angeles coefficient applied to a California or national median
The source study calls that “exactly the interpolation this page refuses,” and this digest refuses it too.
Brueckner and Thomaz measured one city, on a 2013–2019 assessor panel, with an OLS specification whose instrumental-variable alternative failed, and they note that for parcels which have not recently sold the assessed increment may rest on the assessor’s cost estimate rather than a market price.
The inversion table on this page stays inside Los Angeles for exactly that reason.
- Construction cost restated as value added
Fannie Mae B4-1.3-10 does not accept the cost approach as the sole indicator of market value, and San Diego County’s assessor names both directions in which cost and market value part — an over-improvement for the area, and an owner-builder who spent substantially less than the resulting increase.
An invoice is a denominator. It is not a valuation, and a refinance appraisal will not hand it back.
- A resale story that assumes a tenant
The income approach assumes one, and the GSE income rules are written for one.
The 2021 California owner survey found 51 per cent of new ADUs generating rental income, 16 per cent providing no-cost housing to a relative, and 8 per cent used as short-term rentals.
Capitalising a market rent an owner will not collect, and calling the result what the ADU is worth, is a category error for a large minority of units.
- An SB 1164 property-tax holiday
It is not law. The bill would have excluded ADU construction from “new construction” for up to ten years; it passed the Senate 29–6 on 22 May 2024 and died in the Assembly Revenue and Taxation Committee on 24 June 2024.
The Board of Equalization’s published list of new-construction exclusions does not include ADUs. Value copy that promises the assessment will not move is describing a bill that failed.
What a resale actually does to the four numbers
While you own it
Cost is sunk. Appraised value appears only when you refinance or take a home-equity product, and the lender will want the ADU on its own line of the grid — plus a rent schedule if you are asking them to count the income.
Assessed value already moved, on the day the unit was complete, by the § 71 increment plus a supplemental bill.
Income is whatever the lease, the family arrangement or the vacancy actually is: haircut, capped at 30 per cent of qualifying income if a GSE is counting it, and unavailable on a cash-out refinance at Fannie Mae. Four levels, one property, one Tuesday.
When you sell
There is one price for the whole property. The buyer’s lender orders another appraisal, against whatever comparable set exists on that date, under whatever guide is then in force.
The county reassesses the entire property to the sale price as a change in ownership, and the ADU’s separate base year value disappears into a single new base year for the parcel.
An unpermitted unit is a different assignment again: FHA and VA want it legal; Fannie Mae and Freddie Mac may finance it with extra comparables and will not count its rent; and an assessor who discovers it through a permit or a sale can enrol it as new construction that should have been enrolled earlier, with escape assessments. Legalizing first is a permit question, not a percentage question.
Citing these figures
Link the value study rather than this digest where you can. The full quotations from the guides, the county assessor material, and the record of what was searched for and not found live there.
Cite Brueckner and Thomaz as Brueckner and Thomaz, with the city, the panel years and the identification caveat attached.
“ADUs raise property value by 7–9 per cent” is a misquotation of a paper that says the association holds for assessed value and selling price on City of Los Angeles parcels from 2013 to 2019, in an OLS specification the authors could not instrument.
Cite the Terner Center and the Center for Community Innovation for cost, rent and occupancy — never for a resale premium.
Where a figure is labeled HyreADU calculation, attribute it to HyreADU and name the derivation: “HyreADU ratio of an indexed California ADU construction median to the ACS 2024 median value of owner-occupied housing units,” or “HyreADU inversion of the Brueckner and Thomaz coefficient against an indexed Los Angeles County construction median.” The word “inversion” is load-bearing — dropping it turns a conditional into a forecast.
Corrections are welcome and are published on the page with a dated note: hello@hyreadu.com. If a peer-reviewed sale-price study of ADU properties outside Los Angeles exists, or appears, it supersedes the central limitation of this page and we want to know about it.
Questions
Do ADUs add value to a property?
Where does the “ADUs add 20–30 per cent” figure come from?
Will an ADU pay for itself in resale value?
Is the assessed increase the same as the value added?
Why is it so hard to find ADU comparable sales?
Does the 30 per cent GSE cap mean an ADU adds 30 per cent?
How big is an ADU relative to the house it sits behind?
Which figures on this page are HyreADU’s own?
Is any of this financial advice?
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 Brueckner and Thomaz 2024; Adomatis, The Appraisal Journal Fall 2021; Brown and Watkins, The Appraisal Journal Spring 2012; Fannie Mae Selling Guide dated 2 September 2026; Freddie Mac ADU fact sheet February 2026; HUD Handbook 4000.1; ACS 2024 1-year table B25077; county assessor guidance — all retrieved 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
- Jan K. Brueckner and Sarah Thomaz, “ADUs in Los Angeles: Where are they located and by how much do they raise property value?”, Real Estate Economics 52(3), 2024, 885–907 , City of Los Angeles tax-assessor parcel data, 2013–2019. Assessed-value regressions “show that ADU presence raises a parcel’s assessed value and selling price by 7%–9%, while also accurately capturing the unusual rules for property assessments under California’s Proposition 13.” All properties: 7–8 per cent; column 1 of the all-property specification 7.9%; just-sold subset “almost 9 per cent.” The alternative instrumental-variable approach, using a dummy for assessment years after the 2016 ADU-enabling bills, was unsuccessful. For properties that have not recently sold, “the increment to assessed value may instead rely on the assessor’s estimate of the construction cost of the ADU, which may not correspond to the incremental value upon sale.” Retrieved 2026-09-05.
- Sandra K. Adomatis, SRA, “Valuation of Accessory Dwelling Units,” The Appraisal Journal 89(4), Fall 2021 , The Appraisal Institute’s peer-reviewed practice paper. “Prior to the 2020 Fannie Mae Selling Guide update, it was next to impossible for appraisers to meet federal Agency appraisal guidelines, because three sales of comparable ADU properties were extremely difficult to find in the local market.” Data-search section: “All MLSs do not have searchable fields to identify ADU properties. Public records also may not separate the ADU living area from the main structure.” Exhibit 1 compares agency rules as of 2021; its rental-income rows have since been overtaken. Retrieved 2026-09-05.
- Martin J. Brown and Taylor Watkins, The Appraisal Journal, Spring 2012 , Portland, Oregon; 14 properties with permitted ADUs. Income-capitalisation valuations compared with actual sale prices: the income approach came in 7.2% or 9.8% above actual sale prices depending on formula, and ADUs contributed 25% or 34% of each property’s income-approach appraised value. Those shares are of an income-approach value, not a market premium over otherwise identical houses. Retrieved 2026-09-05.
- Fannie Mae Selling Guide — B4-1.3-05 (Improvements Section of the Appraisal Report) and B4-1.3-10 (Cost and Income Approach to Value) , B4-1.3-05: ADU living area is reported and adjusted on a separate line in the grid, not folded into the primary dwelling’s finished above-grade square footage unless contained within it with interior access and above grade; the report “must include a description of the ADU and analysis of any effect it has on the value or marketability of the subject property”; an aged settled sale qualifies as a comparable and an active listing only as a supplemental exhibit; an illegal ADU requires at least two comparable sales with the same non-compliant zoning use within a minimum of three settled sales. B4-1.3-10: appraisals relying solely on the cost approach are not acceptable (manufactured homes excepted), and appraisals relying solely on the income approach are not acceptable. Retrieved 2026-09-05.
- Freddie Mac Guide § 5601.2 and the Accessory Dwelling Units fact sheet, February 2026 , “At least one comparable sale with an ADU, when available, is required to demonstrate the property’s conformity and marketability to its Market Area.” Where a recent one is unavailable in the subject’s market area, an older sale with an ADU or a sale from a competing market area may be used. The report must describe general condition, room count including bedrooms and bathrooms, and finished square feet, and must conclude whether an adjustment is supported for the ADU. Where ADU rent is used to qualify, a full appraisal is required, automated collateral evaluation is not acceptable, and the rental analysis needs three comparable rentals with at least one rented ADU. Retrieved 2026-09-05.
- HUD Handbook 4000.1, II.D.3 — Accessory Dwelling Unit valuation , “The Appraiser must measure the Contributory Value of the ADU by applying techniques based on one or more of the recognized three approaches to value: cost approach, income approach, and sales comparison approach.” ADU living area is not included in the primary dwelling’s gross living area. A one-unit property with a single ADU remains a one-unit property; on a property with two or more units, a separate additional dwelling unit counts as an additional unit. Mortgagee Letter 2023-17 adds the optional ADU market-rent analysis and requires at least one comparable rental that is a single-family dwelling with a rented ADU. Retrieved 2026-09-05.
- U.S. Census Bureau, American Community Survey 2024 1-year estimates, table B25077 (median value of owner-occupied housing units) , California $759,500, margin of error ± $3,644. United States $360,600, margin of error ± $618. Value as reported by occupants, across all owner-occupied housing. Not a sale price, not an assessed value, and not ADU-specific — used on this page only as a denominator so a cost figure has visible scale. Retrieved 2026-09-05.
- Karen Chapple, Dori Ganetsos and Emmanuel Lopez, Implementing the Backyard Revolution (UC Berkeley Center for Community Innovation, 22 April 2021) , Survey of 752 California homeowners who had successfully built, identified from HCD Annual Progress Report addresses for 2018–2019 permits and certificates of occupancy. About 51 per cent of new ADUs serve as income-generating rentals, 16 per cent provide no-cost housing to a relative, and 8 per cent are short-term rentals. Median construction cost $150,000. Cost, rent and occupancy research — not a sale-price premium, and citing it for one is a category error. Retrieved 2026-09-05.
- County assessor guidance — Los Angeles, San Diego, Santa Clara and San Francisco , Los Angeles County: the Assessor’s public information officer, quoted in the Los Angeles Times on 12 May 2023 — “Our staff will determine the cost to build it,” added to the current assessment and taxed at 1 per cent plus roughly 0.25 per cent in local levies; the same article reports Zillow counts of 642 sold ~2,000 square foot Los Angeles houses in a three-month window, of which 106 listings included an ADU. San Diego County ARCC: the value added “is usually the full construction cost,” with exceptions for an over-improvement for the area and for an owner-builder who spent substantially less than the resulting increase in value. Santa Clara County: “The short answer is yes, but only on the marginal value of the ADU… The existing home will not be reassessed.” Retrieved 2026-09-05.
- California Revenue and Taxation Code §§ 70 and 71, and SB 1164 (Newman, 2023–24) , Section 70 defines new construction; § 71 sets a new base year value for the newly constructed portion only and leaves the base year value of the remainder unchanged. SB 1164 would have added § 74.9 to exclude ADU construction from “new construction” for up to ten years; it passed the Senate 29–6 on 22 May 2024 and died in the Assembly Revenue and Taxation Committee on 24 June 2024. There is no ADU new-construction exclusion in the code. Retrieved 2026-09-05.
- HyreADU cost and permit-valuation work , Indexed construction medians are the 2018–19 California ADU owner survey carried to August 2026 on the DGS California Construction Cost Index (December 2019 6,924 to August 2026 10,633, factor 1.536). Declared permit valuations are from the contractor-named City of Los Angeles harvest (median $40,000) and from LADBS resource pi9x-tg5x for the per-square-foot rate of $120. Declared valuation is a fee input, not construction cost and not market value. Retrieved 2026-09-05.
If the project only works as a 20–30 per cent value-add, it does not work
On the evidence this desk could find, the measured differential is 7–9 per cent in one city on a pre-boom panel, and there is no national figure at all. Housing a person, or collecting a rent that covers operating costs in the tight scenario, are different and more defensible jobs.
HyreADU does not appraise, assess, lend, invest, build or prepare taxes. This page is informational and is not financial, tax or legal advice, a guaranteed return, or a reason to build.
No value-add percentage on this page is HyreADU’s: the peer-reviewed coefficients belong to their authors, describe the City of Los Angeles on a 2013–2019 assessor panel, rest on ordinary least squares after an unsuccessful instrumental-variable strategy, and are not shown to apply to any other market.
The break-even table is an inversion — a statement of what host value would be required for two published Los Angeles figures to be equal — and is not a forecast, a valuation or a prediction for any property.
Construction medians used as inputs are a 2018–19 owner survey indexed by the California Construction Cost Index, which the Department of General Services states does not reflect the current market bidding environment.
Census median home values are all owner-occupied housing as reported by occupants, with the margins of error shown, and are not sale prices or ADU figures.
Appraisal practice varies by appraiser and by market; two competent appraisers given the same thin comparable set can reach different contributory values and both be within professional standards. California assessment rules are not national.
The useful next step on a specific lot is a licensed appraiser, the county assessor and your own advisor.