Energy · Deep dive

Sunrun

America's largest residential solar and battery company — 1M+ subscribers on 25-year power contracts, financed by a $14B+ tower of tax equity and non-recourse debt — now the last scaled survivor of a sector in which nearly every major rival has gone bankrupt, and the chief beneficiary of a tax-code loophole (48E third-party ownership) that Congress has already scheduled for demolition.

at risk

Sunrun is the last scaled survivor of residential solar's bankruptcy wave and the prime beneficiary of the 48E third-party-ownership loophole, but that advantage is a tax-code artifact on a statutory timer — set against $14.8B of debt, a market shrinking ~20% in 2026, originations down 25%, and a sales machine that keeps drawing attorneys general, the position erodes rather than compounds.

My take ★★ Very interesting ★ Interesting − Maybe ✗ Not interesting

Snapshot

Sunrun is the largest US residential solar and battery company: 1,014,945 subscribers as of March 31, 2026, roughly 237,000 home batteries under management, and $2.96B of 2025 revenue. Its model is solar-as-a-service — homeowners pay a monthly rate for 25 years while Sunrun and its financing partners own the hardware and harvest the tax credits. That structure made it the accidental winner of Washington’s 2025 tax rewrite: the OBBBA killed the 25D homeowner credit on December 31, 2025, but third-party-owned systems still qualify for the commercial 48E credit, funneling the surviving market toward Sunrun’s subscription. It is at once the sector’s consolidation winner — SunPower, Sunnova, Mosaic and Freedom Forever all went bankrupt between August 2024 and April 2026 — and a $17B-enterprise-value bet, carried on ~$14.85B of debt, that a subsidy with a legislated sunset can be outrun.

Founding story

Sunrun began in January 2007 in the attic of Ed Fenster’s San Francisco house, founded by Lynn Jurich — then 27, fresh out of Stanford’s business school — with classmate Fenster and Nat Kreamer, a veteran who ran installations and billing. The insight was financial, not technical: panel prices were falling, but a $30,000+ upfront cost froze out ordinary homeowners, so Sunrun sold the electricity instead of the equipment — a residential PPA, with investors owning the system and monetizing the commercial tax credit homeowners couldn’t touch. Fenster, a Blackstone-trained deal structurer, took the first CEO turn and built the tax-equity machine; Jurich built the pipeline, signing early customers at county fairs. Roughly $300M of venture money preceded the August 2015 IPO at $14 a share. The defining deal came in July 2020: the all-stock acquisition of Blackstone-controlled Vivint Solar, which merged the #1 and #2 TPO players and added Vivint’s door-to-door sales army. In August 2021 Jurich handed the CEO job to board member Mary Powell — the former Green Mountain Power chief who spent 2008-2019 turning a Vermont utility into a distributed-energy showcase — who refocused Sunrun on batteries, grid services and cash over growth.

How it works

A Sunrun subscription starts with a door-knock, a web lead, or an affiliate dealer. Sunrun designs, permits and installs the system (70%+ now include a battery), and the homeowner signs a 20-25-year lease or PPA: no money down, a monthly or per-kWh rate, typically with an annual escalator of 0-2.9%. Residential PPA rates averaged roughly $0.15/kWh in 2025, from ~$0.08 in Texas to ~$0.25 in high-cost states (SolarTech, 2025) — the pitch is undercutting the utility on day one.

The financial machinery is the real product. Each tranche of systems goes into a fund with a tax-equity partner that, under a “partnership flip,” takes up to ~99% of the tax benefits — the 48E ITC averaged 42.6% of system value in Q2 2025 with domestic-content adders — until it hits a target return, after which economics flip back to Sunrun. The contracted payments are then levered through warehouse facilities and securitizations: $2.4B of senior ABS across five deals in the first nine months of 2025 alone, all non-recourse to the parent. The headline metric, net subscriber value — present value of 25 years of payments plus tax credits, minus creation cost — hit $11,892 in Q1 2026, up 14% y/y. Since late 2025, Sunrun also sells a big share of newly built systems (50% in Q4 2025, up from 10% in Q3) outright to a third-party owner at installation, booking upfront revenue while keeping the service relationship and grid-services rights — the main reason 2025 revenue jumped 45%.

The batteries add a second business: Sunrun networks its fleet — 4 GWh across 18 programs by early 2026 — into “distributed power plants” selling dispatched capacity to grid operators, like CalReady in California (~75,000 batteries, up to 375MW peak in 2025).

Product and business overview

Four components. Subscriptions — the core: ~90% of customers are on long-dated leases/PPAs. System and product sales — outright homeowner sales (marginal post-25D) plus the wholesale channel of selling installed systems to third-party owners. Storage — battery attachment hit a record 73% of new installs in Q1 2026, up from near zero in 2019. Grid services / distributed power plants — 18 programs monetizing the battery fleet, headlined by the June 2026 agreement with Tesla and Renew Home to aggregate 16+ GW of batteries and smart devices into the country’s largest distributed power plant, aimed at data-center demand, with 300+ MW ready in Virginia; Sunrun targets 10+ GWh dispatchable by end-2028.

Business model and pricing

Revenue books in two streams: customer agreements (the subscription annuity, recognized ratably over 25-year terms) and system/product sales (upfront, lumpy, +114% in 2025 on the third-party sale structure). The cash economics per customer: originate, collect the ITC via tax equity, lever the contracted payments with non-recourse ABS, keep the spread. Pricing is a monthly payment or ~$0.15/kWh-average PPA rate with escalators up to ~2.9%/yr over 25 years, with renewal, purchase, or removal options at term. The uncomfortable math: GAAP profitability is chronically negative or noisy, so the company steers investors to “cash generation” — $377M in 2025, guided to $250-450M for 2026 — a non-GAAP figure skeptics note is sensitive to management’s discount-rate and renewal assumptions.

Traction over time

The pattern: customer growth decelerated sharply even as revenue and unit margins rose — a deliberate trade of volume for value, helped by shrinking the affiliate channel (planned down 40%+ in 2026) and selling systems wholesale. Networked solar capacity passed 8 GW in 2025; storage reached 4 GWh by early 2026. Headcount moved the other way, with layoffs from 2023 through 2025.

Market analysis

US distributed solar installed 8.4 GWdc in 2025, down 5%, with residential roughly flat (SEIA/Wood Mackenzie, Q4 2025). The forecast turns ugly: Wood Mackenzie and SEIA project residential installs contracting roughly 18-21% in 2026 after 25D’s expiry. Inside that shrinking market, mix shifts violently toward TPO — the only channel still monetizing a federal credit (48E, for systems placed in service through 2027, or later if construction began before July 4, 2026 under safe-harbor rules); Sunrun is spending $50-100M in 2026 on safe-harbor equipment to stretch ITC eligibility to about 2030. The tailwinds are real: climbing retail rates, AI data centers straining the grid, aggregated batteries becoming sellable peak capacity. But the subsidy scaffolding beneath residential solar is being dismantled on a published schedule, and no one in the sector has yet shown attractive unsubsidized unit economics.

Competitive intel

The competitive story of 2024-2026 is elimination: SunPower filed Chapter 11 in August 2024; Sunnova — the #2 TPO player, 500,000+ customers — and loan financier Mosaic followed in June 2025; Freedom Forever, the #2 installer of 2025 with 6.1% share (Wood Mackenzie), filed in April 2026 owing $500M-$1B. Sunrun absorbed dealers and share from each. What remains: Tesla Energy, the strongest survivor, with brand-driven acquisition, cheap Powerwalls, and its own 535+ MW California VPP — though its June 2026 decision to build the 16 GW distributed power plant with Sunrun suggests partnership beats war in grid services. Palmetto attacks from below with LightReach, a software platform letting independent installers sell TPO without Sunrun’s overhead. EnFin/Qcells brings manufacturer money downstream but took losses in the Freedom Forever collapse. GoodLeap and the loan fintechs lost their core product with 25D and now contest TPO financing. The utilities remain the meta-competitor: California’s NEM 3.0 (April 2023) cut export values ~75%, and hostile rate design can kill a market overnight. Sunrun’s edge is scale in exactly what survived: TPO structuring, tax-equity relationships, a million-customer service book, and the largest dispatchable residential battery fleet.

History and evolution

• Jan 2007 — Founded by Jurich, Fenster, Kreamer; pioneers the residential solar PPA.
• 2014 — Buys Mainstream Energy (REC Solar residential, AEE Solar, SnapNrack), moving into direct installation.
• Aug 2015 — IPO at $14/share; raises $251M.
• Jul-Oct 2020 — Acquires Vivint Solar all-stock ($3.2B EV); becomes the undisputed #1.
• Jan 2021 — Stock peaks near $100 (~$20B market cap) in the clean-energy mania — never approached since.
• Aug 2021 — Mary Powell replaces Jurich as CEO; storage-first pivot begins.
• Jul 2022 / Oct 2023 — Muddy Waters short reports allege inflated subscriber counts and excess ITC claims; Sunrun rebuts.
• Apr 2023 — California NEM 3.0 slashes export compensation; the largest state market contracts.
• 2023-2024 — ~495 WARN-tracked layoffs; office closures; rate hikes crush growth-at-any-cost solar.
• Jul 2024 — Connecticut AG sues Sunrun and dealers over deceptive sales practices.
• Aug 2024 — SunPower files Chapter 11; Sunrun recruits its dealers.
• Jun 2025 — Sunnova (500k+ customers) and Mosaic file Chapter 11.
• Jul 4, 2025 — OBBBA signed: 25D dies Dec 31, 2025; 48E sunsets after 2027 with a construction-start safe harbor.
• Feb 2026 — FY2025 results: revenue +45% to $2.96B, $377M cash generation; affiliate volumes to be cut 40%+.
• Apr 2026 — Freedom Forever (#2 installer) files Chapter 11.
• May 2026 — Q1 2026: adds -25% y/y but EPS $0.62 beats; stock rallies ~10%.
• Jun 2026 — Announces 16 GW distributed power plant with Tesla and Renew Home targeting data-center demand.

What people say

The case for. Bulls read Sunrun as the sector’s structural winner: every scaled competitor’s bankruptcy delivered it dealers, customers and pricing power; 48E-plus-safe-harbor gives it a subsidized product into 2030 while cash-and-loan rivals lost theirs entirely; and the margins show it — net subscriber value up 14% y/y, $377M of 2025 cash generation above guidance midpoint. Grid services are the kicker: the 16 GW Tesla/Renew Home program positions home batteries as merchant capacity for the AI-load era. Happy Trustpilot customers describe smooth installs and honest reps; Glassdoor sits at 3.2/5, 46% willing to recommend.

The complaints. They are serious and recurring. The Connecticut AG’s July 2024 suit alleges forged signatures, backdated contracts, and an employee impersonating a homeowner on a verification call; GBH reported in April 2025 that Sunrun had filed 420+ lawsuits against its own Massachusetts customers since 2023, while homeowners counter-sued alleging fraud; BBB and Trustpilot complaints center on undisclosed escalators, systems never activated, and 25-year contracts complicating home sales. Muddy Waters argued in 2022 and 2023 that subscriber counts were overstated by up to five quarters of deployments and that excess tax credits — ~$200M for 2022 alone, on its math — were claimed on inflated system values; Sunrun disputes all of it, but IRS-audit tail risk is not zero. Employees describe the 2023-2025 layoff waves as a “bloodbath” on layoff boards. And the structural critique — the SolarCity echo: a door-to-door sales machine, negative GAAP earnings, a debt tower, and a valuation resting on 25-year renewal and default assumptions management itself sets.

Outlook: well positioned or at risk?

At-risk. The bull case is real but has an expiration date printed on it. Sunrun’s advantage — the only scaled player whose product still carries a federal tax credit — exists because Congress killed the homeowner credit and left the commercial one briefly alive; the same law sunsets 48E after 2027, and the $50-100M safe-harbor program stretches eligibility only to about 2030. An ITC averaging 42.6% of system value is not a moat; it is most of the margin. No US residential solar company has shown healthy unsubsidized unit economics, and Sunrun enters that test with ~$14.85B of debt against a $4B market cap, in a market shrinking ~19-21% in 2026, with originations down 25% y/y. The consolidation “win” cuts both ways: Sunnova was the business-model twin at half scale, and it is in Chapter 11 — proof the leverage stack fails when growth stops and capital tightens. Add the consumer-trust overhang, and the origination engine faces rising friction just as it must learn to sell on pure economics.

What would flip this call: the distributed-power-plant business becoming a real second P&L. If the 16 GW Tesla/Renew Home program turns the battery fleet into contracted merchant capacity priced against data-center scarcity, Sunrun becomes a power producer with a million captive nodes and the tax-credit cliff matters far less. That is a genuine asset, and Powell is the right CEO to build it. But today grid services are a rounding error next to the subsidy-dependent core, whose economics degrade on a statutory schedule the company does not control. A well-positioned incumbent compounds its advantage; Sunrun is racing to build a second business before its first loses the tax code that makes it work. That is the definition of at-risk.

How a challenger would attack it

Build for the unsubsidized market Sunrun is postponing. Sunrun’s entire margin structure assumes an ITC averaging 42.6% of system value, safe-harbored to about 2030; a challenger engineers for the day after — radically lower creation cost through standardized hardware, self-serve digital origination, and no door-knockers — so it wins by default when the subsidy math dies and Sunrun’s high-overhead machine has to sell on pure economics. Palmetto’s LightReach shows the shape: an asset-light software platform letting independent installers sell TPO without Sunrun’s cost stack, attacking origination margin directly. Second vector: trust. Sunrun’s sales record — a Connecticut AG suit alleging forged signatures and impersonation, 420+ lawsuits against its own Massachusetts customers, BBB complaints about undisclosed escalators and 25-year contracts snarling home sales — is a standing invitation for a challenger selling plain-language contracts, no escalators, and portable agreements, marketed against the incumbent’s court docket. Third: skip solar entirely. The valuable asset in the AI-load era is the dispatchable battery, and a storage-first aggregator — batteries retrofitted onto the millions of existing arrays, monetized in grid-services markets — captures the second P&L Sunrun is racing to build without carrying $14.85B of solar-annuity debt or a shrinking install business.

Same playbook, new buyer

Fenster’s real invention was the financing structure, and it ports to any credit-worthy home asset with a tax or savings stream. The nearest translation: heat pumps, whole-home electrification and backup power sold as-a-service on the same tax-equity-plus-ABS rails — categories with utility-bill savings, no NEM exposure, and no legislated subsidy cliff aimed at them, where no scaled TPO player exists. Second: small commercial and multifamily rooftops, structurally underserved because they’re too small for utility-scale developers and too complex for residential sales machines — yet they qualify for the same 48E credit Sunrun is safe-harboring, on better roofs with stronger credits. Third: geography — markets like Texas, where PPA rates run ~$0.08/kWh against rising retail prices, reward a low-cost regional specialist, while Sunrun’s national overhead prices it for ~$0.25 states. Sunrun won’t chase these fast: its million-customer service book, dealer channel and securitization templates are all shaped around residential solar subscriptions, its capital is committed to safe-harbor equipment and the Tesla/Renew Home buildout, and its balance sheet has no room for a second origination machine. The playbook outlives the tax code; the company running it may not follow it to the next asset.

Sources and further reading

• Sunrun Reports Fourth Quarter and Full Year 2025 Financial Results — Sunrun IR, February 2026. FY2025 revenue $2.96B, $377M cash generation, Q4 subscriber adds -17%.
• Sunrun touts record storage attachment rate, positive cash flow in 2025 — pv magazine USA, February 2026. Third-party sale structure, affiliate cuts, 4 GWh storage fleet, safe-harbor plans.
• Sunrun Reports First Quarter 2026 Financial Results — Sunrun IR, May 2026. Adds -25% y/y, subscribers 1,014,945, 73% storage attach, NSV $11,892.
• Sunrun saw steep sales drop in Q1 with end of solar tax credit, tariffs — Utility Dive, May 2026. OBBBA demand shock and Q1 2026 context.
• Attorney General Tong Sues SunRun — Connecticut AG, July 2024. Deceptive-sales allegations including forged signatures and impersonation.
• Solar panel company accused of shady business in Massachusetts — GBH News, April 2025. 420+ Sunrun suits against customers; homeowner fraud claims.
• Here’s why Muddy Waters is short solar company Sunrun — CNBC, August 2022 (and Muddy Waters follow-up, October 2023). The phantom-subscriber and excess-ITC thesis; Sunrun’s rebuttals on its IR site.
• Sunrun Announces Definitive Agreement to Acquire Vivint Solar — Sunrun IR, July 2020. $3.2B EV all-stock merger terms.
• Residential solar installer Freedom Forever files bankruptcy — Solar Power World, April 2026. #2 installer’s Chapter 11; with Sunnova/Mosaic (June 2025) and SunPower (Aug 2024) coverage, the sector’s bankruptcy wave.
• Tesla, Sunrun, Renew Home team up on massive 16GW virtual power plant — Canary Media, June 2026. The distributed-power-plant program targeting data-center demand.
• Outlook for US solar worsens under the OBBBA — Wood Mackenzie, 2025. 25D/48E changes and the ~19-21% 2026 residential contraction forecast.
• Sunrun’s Lynn Jurich: Finding Her Place in the Sun — Sequoia Capital. The founding story; with Forbes (November 2016) on Jurich and Fenster.