Original research
Solar battery statistics: what share of American solar homes actually has storage
Paired-battery rates across the whole installed residential fleet, computed from the mandatory federal census of US electric utilities.
Written by HyreSolar Research team Research and analysis
Audited by HyreSolar Research team Data audit and fact check
The headline
Two numbers, one name, and a factor of several between them
Before any figure on this page is useful, one distinction has to be settled, because almost every published battery statistic gets it wrong.
What this page measures
The installed-fleet rate. Take every residential net-metered solar system standing in the United States at the end of 2024 (5,077,543 of them, some connected last month, some connected in 2011) and ask how many have a battery beside them. The answer is 297,550, or 5.9%.
That denominator includes a decade and a half of systems sold when a battery cost more than the array and exporting to the grid paid full retail value. Nobody bought storage in that market, and those systems are still in the count.
It is the right number for questions about the existing housing stock: how much behind-the-meter storage is actually on the distribution system, how many households can ride through an outage, how far the flexible-load resource extends today.
What this page does not measure
The attachment rate on new sales: the share of systems sold this quarter that include a battery. That figure is far higher, it moves quickly, and it is produced by installer and market-research surveys rather than by the federal filing this page uses.
The two are not competing estimates of the same thing. They have different denominators. A market can plausibly sell most new systems with storage while the fleet rate sits in single digits, and that is roughly what is happening.
If you have arrived here after reading that "most solar homes now have a battery", you have met the sales rate wearing the fleet rate's clothes. Both numbers are real. Neither substitutes for the other.
The five figures worth carrying away
Exactly what is being counted
- Paired battery
- A battery reported by the utility as installed alongside a residential net-metered photovoltaic system, on the same customer account. Counted as an installation, not as kilowatt-hours or kilowatts.
- Standalone battery
- A home battery with no solar array behind the same meter. Reported on a different line of the same schedule and excluded from every figure here, because including it would put a numerator in a denominator it does not belong to.
- Attachment rate
- PV-paired residential battery installations ÷ residential net-metered photovoltaic installations, same state, same filing year, same utilities. A ratio EIA does not itself publish.
- Qualifying state
- A jurisdiction that filed a battery figure and reports at least 10,000 residential net-metered solar systems. 38 of 51 jurisdictions qualify.
- Fleet
- Every net-metered residential system standing at year end, regardless of when it was installed. The stock, not the year's flow.
Paired-battery rate by state, 2024
Every jurisdiction that filed a battery figure and carries at least 10,000 net-metered residential solar systems, ranked. The inclusion floor is doing real work here and the section below it explains why.
| # | State | Battery rate | Paired batteries | Solar systems | Households with solar | Residential price |
|---|---|---|---|---|---|---|
| 1 | Michigan | 18.4% | 4,316 | 23,456 | 0.52% | 19.30¢ |
| 2 | Hawaii | 18.2% | 17,953 | 98,418 | 22.04% | 42.86¢ |
| 3 | Utah | 10.7% | 8,642 | 81,115 | 6.37% | 12.22¢ |
| 4 | California | 10.6% | 221,123 | 2,090,983 | 14.71% | 31.97¢ |
| 5 | Rhode Island | 6.7% | 1,155 | 17,321 | 3.79% | 28.65¢ |
| 6 | Ohio | 6.0% | 1,647 | 27,520 | 0.54% | 15.99¢ |
| 7 | Indiana | 5.7% | 629 | 11,045 | 0.36% | 14.77¢ |
| 8 | Kentucky | 4.7% | 476 | 10,069 | 0.48% | 12.79¢ |
| 9 | Washington | 4.4% | 2,579 | 59,235 | 1.76% | 11.90¢ |
| 10 | Arizona | 3.3% | 10,345 | 317,033 | 10.11% | 14.91¢ |
| 11 | New Hampshire | 3.2% | 718 | 22,503 | 3.45% | 23.40¢ |
| 12 | Texas | 2.7% | 3,634 | 133,501 | 1.06% | 14.94¢ |
| 13 | Missouri | 2.5% | 760 | 30,022 | 1.02% | 12.91¢ |
| 14 | Nevada | 2.3% | 3,143 | 134,257 | 10.23% | 15.00¢ |
| 15 | Georgia | 2.1% | 395 | 18,612 | 0.39% | 14.08¢ |
| 16 | Idaho | 2.0% | 465 | 22,899 | 2.64% | 11.52¢ |
| 17 | Oklahoma | 1.9% | 318 | 16,755 | 0.89% | 12.24¢ |
| 18 | South Carolina | 1.9% | 753 | 40,055 | 1.55% | 14.23¢ |
| 19 | New York | 1.8% | 4,006 | 226,979 | 3.06% | 24.43¢ |
| 20 | Oregon | 1.6% | 862 | 52,600 | 2.82% | 14.70¢ |
| 21 | Pennsylvania | 1.6% | 1,298 | 81,375 | 1.47% | 17.77¢ |
| 22 | Wisconsin | 1.5% | 287 | 18,581 | 0.65% | 17.18¢ |
| 23 | Massachusetts | 1.3% | 2,528 | 189,611 | 6.48% | 29.35¢ |
| 24 | Florida | 1.1% | 3,120 | 289,774 | 2.77% | 14.14¢ |
| 25 | Virginia | 1.0% | 697 | 69,553 | 1.90% | 14.41¢ |
| 26 | New Jersey | 0.8% | 1,605 | 199,822 | 5.35% | 19.34¢ |
| 27 | Colorado | 0.7% | 1,298 | 179,751 | 7.03% | 14.92¢ |
| 28 | Connecticut | 0.7% | 677 | 103,824 | 6.69% | 28.75¢ |
| 29 | Maryland | 0.5% | 534 | 108,428 | 4.42% | 17.86¢ |
| 30 | North Carolina | 0.4% | 248 | 57,393 | 1.13% | 14.13¢ |
| 31 | Minnesota | 0.4% | 88 | 22,621 | 0.88% | 15.45¢ |
| 32 | Maine | 0.4% | 55 | 14,867 | 2.00% | 24.29¢ |
| 33 | Arkansas | 0.2% | 45 | 19,146 | 1.30% | 12.32¢ |
| 34 | New Mexico | 0.2% | 112 | 61,209 | 6.54% | 14.20¢ |
| 35 | Iowa | 0.1% | 22 | 16,392 | 1.13% | 13.40¢ |
| 36 | Illinois | 0.0% | 28 | 103,145 | 1.91% | 15.87¢ |
| 37 | Delaware | 0.0% | 3 | 14,087 | 2.98% | 16.57¢ |
| 38 | Louisiana | 0.0% | 2 | 34,417 | 1.60% | 11.73¢ |
HyreSolar analysis of EIA-861 2024. National rate 5.9%. 6 of 38 qualifying states are above it.
Figures labelled HyreSolar analysis are computed by us from the EIA source files named below. EIA publishes the inputs; it does not publish these ratios or rankings.
Why the leaders are the leaders: three tariff closures
Three of the top 4 states share a history that none of the states at the bottom of the table shares. In each of them, a regulator stopped paying retail value for exported electricity. When an exported kilowatt-hour is worth less than an avoided one, the arithmetic of a battery changes overnight: the machine stops being an insurance policy against outages and starts being an arbitrage device against your own evening load.
-
12 October 2015
Hawaii closes net energy metering to new applicants, today at 18.2%The Hawaii Public Utilities Commission capped the net energy metering programme at existing levels and closed it to new participants, replacing it with self-supply and grid-supply tariffs that credit exports below the retail rate. Hawaii had the highest household solar penetration in the country then, 22.0% now, and the highest residential electricity price in the country at 42.86¢/kWh. It has had ten years to accumulate a battery-first fleet, and it shows: 17,953 paired batteries on 98,418 systems.
-
15 November 2017
Utah closes Rocky Mountain Power Schedule 135, today at 10.7%Schedule 135, the net metering tariff, was closed to new applicants. New customers take Schedule 137, net billing, which credits exports at 4.855¢/kWh in summer and 4.033¢/kWh in winter, against a Utah residential retail price of 12.22¢/kWh. That is an export credit worth between a third and two fifths of the retail rate, in a state whose electricity is otherwise among the cheapest in America. Utah has 8,642 paired batteries and sits 3rd in the ranking below, while sitting 46th of 51 on residential price, where first is the most expensive. High storage does not require expensive electricity. It requires a poor export credit.
-
15 April 2023
California's net billing tariff takes effect, today at 10.6%Adopted by the California Public Utilities Commission in December 2022 and effective for interconnection applications from 15 April 2023, the net billing tariff replaced retail-rate export credit with hourly avoided-cost values. California is the only one of the three where the change is recent enough to be visible in the flow rather than the stock: it holds 74% of the national paired-battery count, and the same tariff change is the largest single component of the 30.6% national fall in new solar connections that year. What net metering now covers →
The shape of the distribution: a short head and a very long tail
A national average of 5.9% implies a market where roughly one solar home in seventeen has storage. No state looks like that. The distribution is bimodal in practice: a handful of states between 6.7% and 18.4%, then a cliff, then the great majority clustered under 2%.
Of the 38 qualifying states, 32 sit below the national rate and 22 sit below 2%. The national figure is not a central tendency at all: it is California's rate diluted by forty-odd states that have barely started. Remove California from both the numerator and the denominator and the rest of the country runs at 2.6%, less than half the headline.
That is worth saying plainly because a national attachment rate is routinely quoted as though it described a typical state. In this dataset it describes almost none of them. New Jersey has 199,822 solar systems and 1,605 batteries. Colorado has 179,751 and 1,298. Illinois has 103,145 and 28. These are large, mature solar markets with essentially no storage in them.
The chart below plots every qualifying state as its distance from the national rate, which makes the asymmetry legible in a way the ranked table does not: the bars above the line are few and long, the bars below are many and short, and the line itself sits much closer to the bottom of the range than to the middle.
Michigan is at the top of the table and we cannot explain it
Michigan reports the highest paired-battery rate of any qualifying state: 18.4%, 4,316 batteries on 23,456 solar systems. It is not a small-sample artefact: the base is well above our 10,000-system floor.
Every other state at the top of this table has an obvious tariff story. Michigan does not. We have not been able to verify a dated closure of retail net metering that we would be willing to set alongside Hawaii's, Utah's or California's, and we are not going to imply one. Its residential price of 19.30¢/kWh is above the national 16.48¢ but nowhere near the leaders. Its household solar penetration is 0.52%, 41st of 51, near the bottom.
There are plausible candidate explanations. Michigan loses 482 minutes of power per customer per year against a national median of 297, which would push households towards backup. It also has an unusually high 38.9% of residential customers on a time-varying tariff, against 11.1% nationally, which would push them towards arbitrage. A small solar fleet is also easier to move: 23,456 systems is one 14th of Arizona's 317,033.
We are not asserting any of that. None of it is tested here, and a plausible mechanism offered without a test is how bad statistics get made. What we can say is that Michigan is a genuine outlier at the top of the ranking, that its base is large enough to take seriously, and that we do not know why. If you know, we would like to be told.
Storage does not follow the sun, and it does not follow the solar
The intuitive model is that batteries follow solar: the more rooftop photovoltaics a state has, the more of them will have storage. Across the 38 qualifying states the correlation between household solar penetration and battery attachment is r = 0.47, positive, but weak enough that it explains under a quarter of the variation, and it is carried almost entirely by the same two or three states that carry every other solar correlation.
The scatter makes the failure visible. Nevada has 10.2% household solar penetration, more than three times Michigan's 0.52%, and a battery rate of 2.3% against Michigan's 18.4%. New Jersey, with 199,822 systems, has fewer paired batteries than Michigan does with 23,456.
Price does slightly better: r = 0.48 against residential electricity price. But Utah breaks that too, 3rd on batteries and 46th of 51 on price, first being dearest, which is to say it has among the cheapest electricity in the country and among the most storage. Average system size explains still less, r = -0.37.
What the leaders have in common is not a price level, a solar level or a sunshine level. It is a tariff structure in which the marginal exported kilowatt-hour is worth materially less than the marginal consumed one. That gap is the entire economic case for a residential battery, and no column in this dataset measures it directly, which is precisely why the correlations here are weak and the chronology in the section above is strong.
Three states hold most of the machines
Rates describe intensity. Counts describe where the hardware physically is, and on counts the picture is more concentrated still. California, Hawaii and Arizona between them hold 249,421 paired residential batteries, 83.8% of the national total of 297,550.
California's 221,123 is roughly three quarters of the country's on its own. That has a consequence for anyone reading national storage figures: a change in one state's tariff, one state's incentive programme or one state's interconnection queue moves the national number. The national number is not an average of fifty independent markets. It is California, plus noise.
Arizona is the instructive counter-case. It has 317,033 solar systems, the 2nd largest fleet in the country, and 10,345 paired batteries: a rate of 3.3%, below the national figure. Volume of solar is not volume of storage.
Battery attachment by type of utility
The same figure computed against the ownership class of the utility whose meter the system sits behind, rather than against the state line. Investor-owned utilities are where both the solar and the storage are, and the gap is wider than the gap in solar alone.
| Utility type | Utilities | Solar systems | Battery rate | Household solar share | Share of US solar |
|---|---|---|---|---|---|
| State | 18 | 90,324 | 2.2% | 7.2% | 1.9% |
| Investor Owned | 163 | 3,898,686 | 6.6% | 5.5% | 83.5% |
| Political Subdivision | 88 | 166,919 | 4.5% | 4.5% | 3.6% |
| Municipal | 458 | 270,294 | 2.6% | 2.0% | 5.8% |
| Cooperative | 602 | 244,668 | 3.5% | 1.4% | 5.2% |
| Federal | 8 | 90 | 0.0% | 0.3% | 0.0% |
HyreSolar analysis of EIA-861 2024 Utility Data joined to the Net Metering schedule. Behind-the-meter aggregators, community choice aggregators and retail power marketers are excluded because they do not own distribution and their filings duplicate the incumbent's.
A solar system on an investor-owned utility is 2.5× more likely to have a battery than one on a municipal utility, and 1.9× more likely than one on a cooperative. Adoption by utility ownership type →
The 13 jurisdictions the ranking leaves out
A ratio computed on a tiny denominator is not a rate, it is an anecdote with a percent sign. Tennessee would top this ranking at 39.5%, on 86 solar systems and 34 batteries statewide. Two more batteries anywhere in the state would move that rate by 2.3 percentage points. That is why the 10,000-system floor exists, and this table shows exactly what it removes so the decision is auditable rather than hidden.
| Jurisdiction | Solar systems | Paired batteries | Rate as filed | Why it is out |
|---|---|---|---|---|
| Tennessee | 86 | 34 | 39.5% | Below the 10,000-system floor |
| West Virginia | 3,535 | 603 | 17.1% | Below the 10,000-system floor |
| Wyoming | 2,913 | 119 | 4.1% | Below the 10,000-system floor |
| Montana | 9,151 | 205 | 2.2% | Below the 10,000-system floor |
| Mississippi | 1,789 | 11 | 0.6% | Below the 10,000-system floor |
| Nebraska | 2,807 | 10 | 0.4% | Below the 10,000-system floor |
| Kansas | 8,583 | 2 | 0.0% | Below the 10,000-system floor |
| North Dakota | 81 | 0 | 0.0% | Below the 10,000-system floor |
| South Dakota | 469 | 0 | 0.0% | Below the 10,000-system floor |
| Alaska | 2,706 | — | — | Filed no battery figure |
| Alabama | 76 | — | — | Filed no battery figure |
| District of Columbia | 17,780 | — | — | Filed no battery figure |
| Vermont | 9,193 | — | — | Filed no battery figure |
HyreSolar analysis of EIA-861 2024. Rates shown as filed, and not used in any ranking, chart or national figure on this page other than the map, where the tile is shaded but carries its own label.
District of Columbia is the notable absence among the non-filers: 17,780 solar systems and no battery figure at all. West Virginia and Tennessee are the notable exclusions among small-base states, both would rank in the top three if the floor were removed, and neither has enough systems for that to mean anything.
How to read any battery attachment figure you meet elsewhere
Four questions settle almost every apparent contradiction between two published battery statistics.
- Ask what the denominator is
Systems sold this quarter, or systems standing in total? The two answers to "what share of solar homes has a battery" differ by a large multiple and both are honest. Our answer, 5.9%, is the standing fleet.
- Ask whether standalone batteries are in the numerator
Batteries installed without solar are a real and growing category, and folding them into a solar attachment rate inflates it against a denominator they were never part of. We exclude them.
- Ask which year, and whether the series is long enough to have a direction
EIA has published this column for two data years, 2023 and 2024. Two points define a line but they do not establish a trend, and we will not describe the change as a trajectory until there is a third.
- Treat any per-system kWh figure from this dataset as unreliable
The energy-capacity column on the same schedule fails a physical plausibility check in the two largest states. Anyone publishing an average home battery size derived from EIA-861 has not run that check. See the methodology below.
Methodology
Source and vintage
US Energy Information Administration, Form EIA-861, Net Metering schedule, 2024 final release, downloaded as the published ZIP archive on 2 September 2026. EIA-861 is a mandatory annual census of US electric utilities rather than a survey of a sample, so there is no sampling error in these counts, only filing error and non-response, both of which are visible in the exclusions table above.
The battery columns were added to this schedule for the 2023 data year. Nothing before 2023 exists, which is why this page carries no eleven-year chart while the rest of the library does. A 2025 early release exists and is excluded: it covers a partial set of filers and would show a spurious collapse in every count.
How each figure is derived
National rate, total PV-paired residential battery installations ÷ total residential net-metered PV installations, summed across all filers. 297,550 ÷ 5,077,543 = 5.9%.
State rate: the same ratio computed within each state from the same two columns, never by apportioning a national figure.
California's share, California's paired battery count ÷ the national paired battery count: 221,123 ÷ 297,550 = 74.3%.
Ex-California rate, batteries and systems both net of California, which is the honest way to ask what the rest of the country looks like: 2.6%.
Correlations, Pearson product-moment, computed across the 38 qualifying states only, on the raw state values with no weighting by fleet size.
The inclusion floor, and why it is set where it is
Any state with a battery figure and at least 10,000 residential net-metered solar systems enters the ranking; 38 of 51 jurisdictions do. The floor is applied to the denominator, not to the battery count, so a state with a great deal of solar and almost no storage is still ranked, Louisiana at 0.0% on 34,417 systems is in the table, at the bottom.
10,000 is a judgement, not a derivation. It was chosen so that a single reporting utility changing its filing practice cannot move a state's rate by more than a fraction of a point, and it is applied identically on every page in this library so two of our pages can never disagree about who is third. The exclusions table publishes everything it removes.
A column we can compute and refuse to publish
The same schedule carries a PV-paired battery energy capacity field, and dividing it by the installation count would produce an average home battery size in kilowatt-hours. That figure would get cited. We are not publishing it, and this is the diagnostic that stopped us.
Combined with the reported power rating, the 2024 energy-capacity column implies an average battery duration (the hours a full battery could run at its own rated output) of 0.83 hours in California and 0.38 hours in Texas. No residential battery product sold in the United States has a duration below roughly two hours; the common products sit between two and four. A Texas fleet average of about twenty-three minutes is not a market outcome, it is a units or scope problem in the filing, most likely some filers reporting kW where the form asks for kWh, or reporting inverter capacity rather than usable energy.
We report the diagnostic rather than the metric because the metric is in active use. If you have seen an "average US home battery size" sourced to EIA-861, it was computed from this column, and the person computing it did not divide by the power rating to see what came out. No per-system kWh figure appears anywhere on this page.
Two points are not a trend
The national rate moved from 4.2% in 2023 to 5.9% in 2024, an increase of 107,326 paired batteries. We report that as a change between two observations and nothing more.
Two points cannot distinguish growth from a level shift caused by better reporting in the second year of a new column, and the second year of a new EIA column is exactly where reporting improvements land. Extrapolating a rate of change from this pair ("storage attachment is growing X% a year") is unsupported by the data, and we will not do it until a third year exists.
Limitations
Net-metered systems only. 2,031 MW of residential PV sits outside net metering on a separate schedule, with no installation count and no battery column attached. In Texas that is most of the fleet, so the Texas rate of 2.7% describes a minority of Texan solar homes. The coverage gap →
Installations, not capacity. Every figure here counts machines. A 5 kWh battery and a 40 kWh battery each count once, and for the reason set out above we cannot responsibly weight them.
Filing gaps are silent. 4 jurisdictions filed no battery figure. They are shown as blank rather than zero, because a utility that did not answer the question is not the same as a utility that answered it with none, and treating one as the other is how a state gets reported at 0%.
No causal claim is made anywhere on this page. The tariff chronology is a chronology. It is consistent with the ranking and it is the mechanism practitioners describe, but this dataset contains no counterfactual and cannot establish that the tariff change caused the storage.
Reproducing this
Every number above is generated from the source workbooks by script and read into the page from a single JSON file; none is typed into the prose. The inputs are a public federal download, the arithmetic is stated in this section, and the exclusions are published rather than described. If you find an error, tell us and we will correct it here with a dated note.
Questions
What percentage of solar homes have a battery?
Which state has the highest solar battery attachment rate?
Why does California dominate the battery numbers?
Do batteries follow high electricity prices?
How big is the average home solar battery?
Is solar battery adoption growing?
How does this compare to demand response as a source of household flexibility?
Written and audited by
HyreSolar Research
Primary-source research, data analysis and fact checking
We are a research desk, not a sales floor. We read the statute, the tariff, the code section, the federal filing or the manufacturer data sheet 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. That rule has cost us whole sections, and it is the reason the rest can be trusted.
- 160
- primary sources read and cited
- 220
- figures with a retrieval date
- 115
- federal and state government sources
- 66
- researched pages published
How this desk works
- Primary sources only. Statutes from the legislature’s own publishing system, federal data from the agency that collects it, code text from the adopted edition, manufacturer claims from the data sheet. We do not cite an article that cites a source; we go and read the source.
- Every figure carries its provenance. A named document and the date we retrieved it, so you can check it and so you know how old it is. Retrieval dates are not decoration: an EIA rate from May is a different fact from an EIA rate from August.
- We publish what we could not verify. Every research page carries a section naming the things we tried to establish and could not, and why. A paywalled standard, a state website that refused the request, a manufacturer that publishes no figure at all.
- We separate measurement from modelling from our own reasoning, and label which is which on the page. A laboratory measurement, an assumption inside a modelling tool and our own inference are three different kinds of claim and they are never presented as one.
- We do not sell solar, and we take no payment for placement, ranking or a favourable mention. Nobody buys a position on this site.
Data as of EIA-861 2024 final release. 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 EIA, Form EIA-861 — Net Metering (annual files, 2014–2024) — Utility-level net-metered installations, capacity and PV-paired batteries by state and customer sector. Retrieved 2 September 2026.
- US EIA, Form EIA-861 — Sales to Ultimate Customers (annual files, 2014–2024) — Residential revenue, sales and customer counts by utility and state. The basis for price, bill and consumption figures. Retrieved 2 September 2026.
- US EIA, Average Price of Electricity to Ultimate Customers by End-Use Sector — EIA’s published price series, used to validate our derivation. Agreement across all 357 overlapping state-years is within 0.005¢/kWh. Retrieved 2 September 2026.
- Hawaii Public Utilities Commission, Decision and Order No. 33258, closing net energy metering to new applicants — Issued 12 October 2015. Capped net energy metering at existing levels and introduced the customer self-supply and customer grid-supply tariffs in its place. Retrieved 2 September 2026.
- Utah Public Service Commission, Rocky Mountain Power Electric Service Schedule No. 135 and No. 137 — Schedule 135 (net metering) closed to new applicants 15 November 2017. Schedule 137 (net billing) export credit of 4.855¢/kWh summer and 4.033¢/kWh winter, updated annually on 1 March. Retrieved 2 September 2026.
- Utah Clean Energy, Connecting solar to the grid and export credit rates — Current Schedule 137 export credit rates and the annual 1 March reset. Retrieved 2 September 2026.
- California Public Utilities Commission, Decision D.22-12-056, Net Billing Tariff — Adopted 15 December 2022; applies to interconnection applications submitted on or after 15 April 2023. Retrieved 2 September 2026.
Would a battery pay for itself on your tariff?
The answer turns on one number this dataset does not carry: what your utility pays you for an exported kilowatt-hour against what it charges for a consumed one. The calculator asks for both.
HyreSolar is an independent analysis and matching service. We are not an installer, lender or utility. When a reader asks to be introduced, installers may pay us a referral fee. That fee never buys ranking, scores or placement in research. Our editorial policy sets out the rules.