Original research · Electrification
How Many US Homes Are Actually Fully Electric?
Three correct answers exist to "how many homes have no gas," and no published EIA table can tell them apart. Doing so means testing four fuel choices against each other on the same household, so we opened the microdata and ran the cross-tab nobody had.
Written by HyreElectrical Research Desk Primary-source research and fact checking
The finding
Read this before you use any number on this page
The all-electric homes study reports three different, correctly-calculated populations, and using the wrong one for a claim is the most likely way this page gets misquoted. "All-electric" means electricity runs every one of the four end uses measured. "No fossil fuel" is broader — it allows a small remainder of wood, solar-thermal or other non-fossil fuel — and is 5.4 million homes larger. "No natural gas" is broader still — it allows propane and fuel oil, which are common in areas without gas pipelines — and is 11.3 million homes larger again. Pick the wrong one and a real number becomes a wrong claim.
RECS is a sample survey of 18,496 households weighted to represent 123,529,025 homes, not a census, so every figure here carries sampling error (Jackknife standard errors are stated in the method section). It is also a single 2020 cross-section. It cannot show whether the all-electric share is rising, falling, or flat — only a later RECS cycle, run through this same method, can do that. Every number here is published as a baseline for that comparison, not as evidence of a trend.
Why no published table answers this
HyreElectrical has already published the individual pieces. 42.57 million homes heat with electricity. 57.04 million heat water with it. 65.56 million cook on it, counting only households whose combined range is electric. Each of those figures comes from a separate EIA published table — HC6.1, HC8.1 and HC3.1 respectively — and each table cross-tabs one fuel choice against housing characteristics like type, era and region. None of them cross-tabs against the OTHER fuel choices in the same home. EIA does not publish a table showing how many of the electric-heat homes are also electric-water-heat homes, let alone all four at once.
HYRE analysis. Multiplying or summing the four published shares would only be valid if the four fuel choices were statistically independent within a household — if knowing a home has an electric range told you nothing about whether it also has an electric water heater. That assumption is not measured anywhere and is probably false: a home built or renovated for full electrification tends to get all four appliances switched at once, not one at random. The only correct way to answer "how many homes have all four" is to test all four conditions on the same household record and then weight the result — which means opening the microdata file behind every published RECS table, not the tables themselves. That is the calculation this page publishes.
The word "all-electric" hides three different populations
Splitting the full housing stock into four groups, ranked from least to most electrified, makes the gap concrete. 74,305,815 homes use natural gas in at least one of the four end uses. A further 11,260,759 homes have no natural gas anywhere but still burn propane or fuel oil somewhere — common in rural areas without a gas main. A further 5,384,604 homes use no fossil fuel of any kind but are not strictly all-electric either, mostly running a wood stove or furnace for at least part of their heating. What remains, 32,577,847 homes, is electricity running all four end uses with nothing else in the mix.
HYRE analysis. A journalist or a code advocate quoting "no gas" without specifying which of these three they mean can overstate the all-electric population by up to 51 per cent (49.2 million against 32.6 million) without any number in the sentence being individually wrong. That is the specific confusion this page exists to remove.
Validating the method before trusting the new number
Method. Before publishing a figure nobody has published before, the same microdata pipeline was used to reproduce three figures this site already published from EIA’s own summary tables — a test of the extraction and weighting code, not of the new finding itself.
Reproducing three already-published figures from raw microdata
| Measure | This page, from microdata | Already published (rounded) | Match |
|---|---|---|---|
| Space heating electric (FUELHEAT=5) | 42,574,007 | ≈42.57M | Within 4,007 units |
| Water heating electric (FUELH2O=5) | 57,042,810 | ≈57.04M | Within 2,810 units |
| Cooking electric, range-only households (RANGEFUEL=5) | 65,564,330 | ≈65.56M | Within 4,330 units |
The published figures are this site’s own, from /research/electric-heating-transition/, /research/electric-water-heating/, /research/electric-range-adoption/.
All three reproduce to within 4,330 housing units of a total near 50-65 million — a difference under 0.01 per cent, consistent with rounding in the previously published figures rather than an error in either calculation. The all-electric and no-fossil-fuel figures on this page use the same NWEIGHT extraction and the same fuel-code definitions, applied as a four-way condition instead of one variable at a time.
Every Census division, no-fossil-fuel and all-electric share
| Census division | Housing units | No fossil fuel | % no fossil | All-electric | % all-electric |
|---|---|---|---|---|---|
| East South Central | 7,380,717 | 3,792,378 | 51.4% | 3,638,155 | 49.3% |
| South Atlantic | 24,843,261 | 14,342,312 | 57.7% | 11,786,389 | 47.4% |
| West South Central | 14,619,094 | 6,172,235 | 42.2% | 5,731,024 | 39.2% |
| Mountain South | 4,602,070 | 1,455,372 | 31.6% | 1,230,871 | 26.7% |
| West North Central | 8,495,815 | 1,770,741 | 20.8% | 1,616,591 | 19.0% |
| Pacific | 18,505,643 | 3,925,710 | 21.2% | 2,989,133 | 16.2% |
| East North Central | 18,546,912 | 2,881,271 | 15.5% | 2,498,698 | 13.5% |
| Mountain North | 4,615,844 | 708,694 | 15.4% | 595,679 | 12.9% |
| Middle Atlantic | 16,043,503 | 2,239,762 | 14.0% | 1,909,293 | 11.9% |
| New England | 5,876,166 | 673,977 | 11.5% | 582,013 | 9.9% |
EIA RECS 2020 public microdata, HyreElectrical weighted estimates, retrieved 6 September 2026.
East South Central (49.3%) and South Atlantic (47.4%) lead on all-electric share — both regions where electric resistance heat and heat pumps became the default in postwar Sun Belt construction. New England (9.9%) and Middle Atlantic (11.9%) are lowest, where gas mains and fuel-oil heating are both long-established. South Atlantic has the largest gap between its no-fossil-fuel share and its all-electric share — 10.3 points — meaning a larger share of its non-fossil homes run on wood or another fuel rather than electricity outright.
Where the two measures disagree most
Ranking divisions by all-electric share and by no-fossil-fuel share does not produce the same order. Pacific has the fifth-highest no-fossil-fuel share (21.2%) but drops to sixth on all-electric share (16.2%), swapping places with West North Central (20.8% no-fossil, 19.0% all-electric) — Pacific’s non-fossil, non-electric remainder is larger, consistent with wood heat and solar water heating both being more common in parts of the West than the Midwest.
By year the home was built
| Built | Housing units | No fossil fuel | % no fossil | All-electric | % all-electric |
|---|---|---|---|---|---|
| Before 1950 | 20,256,553 | 2,827,430 | 14.0% | 2,200,069 | 10.9% |
| 1950–1959 | 12,479,355 | 2,046,450 | 16.4% | 1,808,270 | 14.5% |
| 1960–1969 | 12,755,665 | 2,946,880 | 23.1% | 2,546,378 | 20.0% |
| 1970–1979 | 18,339,764 | 7,001,679 | 38.2% | 5,952,097 | 32.5% |
| 1980–1989 | 16,297,370 | 6,960,305 | 42.7% | 5,882,172 | 36.1% |
| 1990–1999 | 17,156,758 | 6,170,190 | 36.0% | 5,418,020 | 31.6% |
| 2000–2009 | 16,159,910 | 5,938,326 | 36.7% | 5,142,776 | 31.8% |
| 2010–2015 | 5,527,203 | 2,247,619 | 40.7% | 1,994,806 | 36.1% |
| 2016–2020 | 4,556,448 | 1,823,573 | 40.0% | 1,633,259 | 35.8% |
| 1950–1989 combined | 59,872,154 | 18,955,314 | 31.7% | 16,188,916 | 27.0% |
EIA RECS 2020 public microdata, HyreElectrical weighted estimates, retrieved 6 September 2026. The 1950–1989 row combines the four decade bands this site’s hazard-era panel study uses, shown for comparability, not because that window has separate significance for fuel choice.
All-electric share rises steadily from 10.9% in pre-1950 homes to a peak of 36.1% in homes built in the 1980s, then goes flat: 2016–2020 construction is 35.8%, no higher than 1980s stock built four decades earlier.
Why newer homes are not more electrified than 1980s homes
HYRE analysis. If full electrification were simply getting more common over time, the newest construction band would have the highest share. It does not: 2016-2020 construction (35.8%) sits fractionally below the 1980s peak (36.1%), and every band from 1990 onward is lower than 1980. The most likely explanation is regional: the 1970s-80s building boom fell heavily in the Sun Belt divisions that already lead this ranking — South Atlantic and East South Central — where mild winters make electric resistance heat and heat pumps ordinary choices. More recent decades have added significant new construction in colder divisions with established gas infrastructure, which pulls the newest bands’ national average back down even as any individual metro area may be electrifying. RECS 2020’s public tables do not cross era against division for this measure, so this explanation is offered as the most consistent reading of the other tables on this page, not as a further cross-tab we ran.
By tenure
| Tenure | Housing units | No fossil fuel | % no fossil | All-electric | % all-electric |
|---|---|---|---|---|---|
| Owner-occupied | 82,920,129 | 21,017,874 | 25.3% | 17,971,564 | 21.7% |
| Renter-occupied | 40,608,896 | 16,944,577 | 41.7% | 14,606,283 | 36.0% |
Renter-occupied combines KOWNRENT codes 2 (rent) and 3 (occupy without payment of rent), matching the convention this site’s renter-electrification-gap study established.
Renter-occupied homes run a substantially higher all-electric share (36.0%) than owner-occupied homes (21.7%) — the reverse of the age pattern in this site’s hazard-era and renter-electrification-gap studies, where renters sit in OLDER housing on average. The driver here is housing type, not age: renters are concentrated in apartments, and apartments run a far higher all-electric share than single-family houses, shown next.
By housing type
| Housing type | Housing units | No fossil fuel | % no fossil | All-electric | % all-electric |
|---|---|---|---|---|---|
| Mobile home | 6,832,499 | 3,372,312 | 49.4% | 3,133,633 | 45.9% |
| Apartment, 5+ units | 22,835,862 | 10,896,341 | 47.7% | 9,304,789 | 40.7% |
| Apartment, 2–4 units | 9,341,795 | 3,283,747 | 35.2% | 2,815,266 | 30.1% |
| Single-family attached | 7,451,177 | 2,010,659 | 27.0% | 1,697,763 | 22.8% |
| Single-family detached | 77,067,692 | 18,399,393 | 23.9% | 15,626,396 | 20.3% |
EIA RECS 2020 public microdata, HyreElectrical weighted estimates, retrieved 6 September 2026.
Mobile homes (45.9%) and large apartment buildings (40.7%) run the highest all-electric shares; single-family detached houses (20.3%), the single largest category at 77.1 million units, run the lowest. Both ends of this range plug directly into the tenure finding above: apartments are disproportionately rented, and single-family detached houses are disproportionately owner-occupied.
Among homes that are NOT all-electric, what is stopping them
| End use | Not-electric homes affected | Share of the 90,951,178 non-electric homes |
|---|---|---|
| Space heating not electric | 80,955,018 | 89.0% |
| Water heating not electric | 66,486,215 | 73.1% |
| Cooking not electric (any appliance) | 52,855,314 | 58.1% |
| Clothes drying not electric | 20,116,512 | 22.1% |
Not mutually exclusive — a home can be blocked on more than one end use at once (see the table below).
Space heating is the single most common reason a home is not all-electric, present in 89.0% of the 90,951,178 homes that fall short on at least one end use. Clothes drying is the least common reason, at 22.1% — consistent with 21.2 million homes having no dryer at all, counted here as already electric-consistent for that category (see the convention above).
How many of the four end uses each home already runs on electricity
| Non-electric end uses | Housing units | Share of all homes |
|---|---|---|
| 0 of 4 — All four electric (the all-electric count) | 32,577,847 | 26.4% |
| 1 of 4 — One end use not electric | 18,162,995 | 14.7% |
| 2 of 4 — Two end uses not electric | 30,568,997 | 24.7% |
| 3 of 4 — Three end uses not electric | 27,764,678 | 22.5% |
| 4 of 4 — All four end uses not electric | 14,454,509 | 11.7% |
Mutually exclusive; sums to 123,529,025.
The distribution is not bimodal. Most homes sit in the middle — one, two or three end uses electric — rather than clustering at "all electric" or "all fossil." Only 11.7% of homes (14,454,509) run on a fossil or non-electric fuel for every one of the four end uses measured.
Limitations
- RECS 2020 is a single cross-section, not a trend
RECS is fielded roughly every four years. This page cannot say whether the all-electric share is rising, falling or flat — that requires a second data point from a later cycle, processed through the same definitions. It is published specifically so that comparison is possible later.
- The no-dryer convention moves the headline number by 20%
Counting a home with no clothes dryer as electric-consistent puts the all-electric total at 32,577,847. Requiring a physical electric dryer instead puts it at 26,160,892 — a 6,416,955-home swing from one modelling choice. Both figures are published; the site’s headline uses the first convention, stated plainly rather than left implicit.
- Dual-fuel ranges are counted as gas-using, not electric
RECS records 5,256,847 households with a dual-fuel range — an electric oven paired with a gas cooktop. This page counts that as fossil-fuel use and as not all-electric, because the cooktop burns gas regardless of the oven. A reader counting dual-fuel as "electric" would get a higher number than this page publishes.
- This is a survey estimate, not a census
18,496 responding households are weighted to represent 123.5 million homes. Every figure carries sampling error; Jackknife standard errors on the three national figures run from about 0.4 million to 0.44 million, roughly 1.1-1.2% of each point estimate. Division- and era-level standard errors are wider and are not separately stated in the tables above; treat any single cell as an estimate, not an exact count.
- Division is the geography, not state
This page follows its topic brief in cutting by Census division rather than state. RECS 2020 does support state-level estimates for simpler measures, but a four-way household condition this specific has not been validated at state level here, and the brief did not ask for it.
- No EIA summary table exists for clothes-dryer fuel to cross-check against
The validation table above reproduces three already-published figures for heating, water heating and range cooking. EIA does not publish an equivalent national summary table for dryer fuel, so the dryer component of this calculation has no independent published figure to check against — it rests on the same microdata and weighting as the other three, which did check out, but that is inference, not direct validation.
- This counts fuel, not emissions, cost or code compliance
A home is counted here purely on which fuel its equipment burns. This page makes no claim about greenhouse-gas emissions, utility cost, electrical capacity, or any building code or incentive programme’s own definition of "all-electric," several of which differ from the one used here.
Method
Source. EIA RECS 2020 public-use microdata file, recs2020_public_v7.csv, 18,496 responding households representing 123,529,025 US primary occupied housing units, weighted with the final analysis weight NWEIGHT. Retrieved directly from eia.gov, together with the 2020 RECS Variable and Response Codebook and the 2020 RECS Methodology Report, on 6 September 2026.
Definitions. Fossil-fuel codes for FUELHEAT and FUELH2O are 1 (natural gas), 2 (propane) and 3 (fuel oil); electricity is code 5 in every fuel field used on this page. Cooking fuel is evaluated per household across every cooking appliance present: RANGEFUEL (where RANGE ≥ 1), COOKTOPFUEL (where COOKTOP ≥ 1) and OVENFUEL (where OVEN ≥ 1), with fossil codes 1 and 2 (fuel oil is not a coded option for cooking equipment) and RANGEFUEL code 13 (dual-fuel) treated as fossil-fuel-using. Clothes-drying fuel is DRYRFUEL, read only where DRYER = 1; a household with DRYER = 0 (21,204,565 nationally) is treated as electric-consistent for this category under the stated convention. A household is "all-electric" if FUELHEAT = 5, FUELH2O = 5, every present cooking appliance is code 5, and clothes drying is electric or absent. A household is "no fossil fuel" if none of FUELHEAT, FUELH2O, any present cooking appliance, or DRYRFUEL (where present) carries a fossil code. A household has "no natural gas" if none of those fields is specifically code 1, allowing propane, fuel oil, wood or solar in their place.
Geography, era, tenure, housing type. DIVISION is EIA’s nine Census divisions with the Mountain division split into North and South. YEARMADERANGE has nine bands, coded 1 (before 1950) through 9 (2016-2020); the 1950-1989 row sums codes 2 through 5. KOWNRENT combines codes 2 (rent) and 3 (occupy without payment of rent) into "renter-occupied," matching this site’s renter-electrification-gap study. TYPEHUQ has five categories, used unmodified.
Variance. Every point estimate’s standard error uses EIA’s 60 Jackknife replicate weights, NWEIGHT1 through NWEIGHT60: Var = (59/60) x the sum, across all 60 replicates, of the squared difference between the replicate’s estimate and the full-sample estimate. This is the same formula and coefficient this site’s renter-electrification-gap study verified against EIA’s own 2020 RECS Methodology Report.
Validation. Before computing the all-electric and no-fossil-fuel figures, the same extraction code was used to reproduce three already-published, single-variable figures from this site’s own prior research — electric space heating, electric water heating and electric range cooking — each of which is independently confirmable against EIA’s own summary tables (HC6.1, HC8.1, HC3.1). All three reproduced to within 4,330 housing units. See the validation table above.
Reproducibility. The public microdata file, codebook and methodology report are all downloadable from eia.gov at no cost. Anyone with those three files can rebuild every figure on this page from the variable definitions stated here.
Questions
How many US homes are all-electric?
How many homes have no gas at all?
What is the difference between a no-fossil-fuel home and an all-electric home?
Which part of the US has the most all-electric homes?
Are renters more likely to live in an all-electric home?
Why isn’t new construction more electrified than older homes?
Does this page count a home as all-electric if it has no clothes dryer?
Will the all-electric share keep rising?
Written and audited by
HyreElectrical Research Desk
Primary-source research and fact checking
We read the model code, the federal safety notice, the municipal fee sheet or the utility tariff ourselves, and we publish the figure with the document it came from and the date we retrieved it. Where a number cannot be traced to a primary source, we publish the shorter page and say what we could not verify. HyreElectrical does not perform, supervise or warrant electrical work. Authorship is organisational: this desk, not a named persona.
- 10
- long-form launch pages in this layout
- 5
- states with verified licence records
- 16,369
- electrical companies in those records
- 74%
- of the store is Florida — disclosed first, not footnoted
How this desk works
- Primary sources only. Code statements come from NFPA 70 as the model code. Safety statements come from CPSC. Permit fees come from the city or county fee sheet. Utility charges come from the filed tariff. We do not cite a blog that cites a source; we open the source.
- Our contractor store is five states, and 74% of it is Florida. Any figure built on that store is titled to those states and names the concentration in the first screen. It is not a national sample of electricians.
- Load calculations on this site are a published simplification in the shape of NEC 220.82 — first 10 kVA at 100%, remainder at 40%, HVAC at 100%, EVSE at 125%. They are labelled as not a stamped calculation. A licensed electrician using the adopted edition does that work.
- No national price for a panel upgrade is shipped. Labour rates, the adopted code edition, whether the utility is involved, and the state of the existing wiring move the invoice too far for a roundup to help. Cost intent lives on the cost page; this page explains the decision.
- We do not perform electrical work, and we take no payment for placement, ranking or a favourable mention. Nobody buys a position on this site.
Data as of 6 September 2026. Authorship on this site is organisational: 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
- US Energy Information Administration — 2020 Residential Energy Consumption Survey, public microdata (v7) and codebook , The household-level FUELHEAT, FUELH2O, RANGE/RANGEFUEL, COOKTOP/COOKTOPFUEL, OVEN/OVENFUEL and DRYER/DRYRFUEL fields, weighted with NWEIGHT and variance-estimated with the 60 Jackknife replicate weights, that this entire page is calculated from. Retrieved 6 September 2026.
- US Energy Information Administration — 2020 RECS Methodology Report (June 2022, revised March 2023) , States the Jackknife variance-estimation method and its 59/60 coefficient used throughout this page. Retrieved 6 September 2026.
- US Energy Information Administration — 2020 RECS, Table HC6.1 (Space heating in U.S. homes, by housing unit type) , Published electric-heating total (42.57 million) used to validate this page’s microdata extraction before it was applied to the new all-electric calculation. Retrieved 6 September 2026.
- US Energy Information Administration — 2020 RECS, Table HC8.1 (Water heating in U.S. homes, by housing unit type) , Published electric water-heating total (57.04 million) used for the same validation. Retrieved 6 September 2026.
- US Energy Information Administration — 2020 RECS, Table HC3.1 (Appliances in U.S. homes, by housing unit type) , Published electric-range total (65.56 million) used for the same validation. Retrieved 6 September 2026.
Thinking about closing the gap yourself?
Moving a home from "no fossil fuel" to fully electric usually means a heat pump, a heat pump water heater, and a panel with room for both. The panel question comes first.
HyreElectrical does not perform, supervise or warrant electrical work, and takes no payment for placement, ranking or favourable mention. This page counts fuel use from federal survey data; it makes no claim about the cost, emissions or code requirements of any specific home, and it does not rest on this site’s own five-state contractor store.