The 2025 tax law didn't kill solar-plus-storage. It moved where the deal dies.

Start with the claim, because this one is worth stating before the evidence: the 2025 rewrite of federal clean-energy tax policy changed who captures the value in a solar-plus-storage deal, but it did not change whether the deal pencils, and it did nothing at all to the two constraints that actually decide whether a project gets built. Those constraints are the interconnection queue and the county commission. Neither one got any easier in the same window Washington was rewriting the tax code, and that timing is the whole story.

What did the tax law actually do?

Less than the headlines suggested, and more unevenly than most coverage captured. The One Big Beautiful Bill Act terminated the residential solar credit, Section 25D, outright on December 31, 2025, with no phase-down and no transition window, according to legal summaries tracking the law's clean-energy provisions (see EnergySage's tax credit explainer and the Latham & Watkins client alert). A homeowner who signs a purchase contract in 2026 gets nothing from Washington directly; the credit simply does not exist for that transaction anymore. Utility-scale and commercial solar under Section 48E fared better but not by much: a project must begin construction before July 5, 2026, or be placed in service by December 31, 2027, to qualify at all, a materially tighter window than the pre-OBBBA rule that gave projects starting in 2025 or 2026 until 2029 or 2030 to reach commercial operation.

Battery storage, by contrast, got the softer landing almost nobody expected. Standalone storage credits do not begin phasing down until projects that start construction after 2033, per Latham & Watkins' summary of the bill's technology-neutral credit structure, with the first real haircut, 75%, hitting projects starting in 2034. The bill's other lever, a set of escalating Foreign Entity of Concern sourcing thresholds, requires qualified facilities to source 40% of "material assistance" cost domestically or from allied supply chains in 2026, rising to 60% by the early 2030s, with storage held to an even tighter 55%-to-75% band over the same stretch. That threshold is a real supply-chain redesign problem for developers who have spent a decade building Chinese-battery-dependent procurement, but it is a sourcing problem, not a returns problem, and sourcing problems get solved with lead time and diversified vendors.

None of that touches the underlying physics of what a battery costs to build, and that number kept falling through the entire legislative fight.

Why did battery costs keep falling regardless of what Congress did?

Because the cost curve is a manufacturing story, not a subsidy story. BloombergNEF's December 2025 global cost survey put the average turnkey battery energy storage system at $117 per kilowatt-hour in 2025, a 31% decline from an inflation-adjusted 2024 baseline of $169/kWh, with four-hour-duration systems running cheaper still at roughly $110/kWh (per Energy-Storage.News' coverage of the BNEF and Ember data). The regional spread is wide, China at $73/kWh against the United States at $219/kWh, which tells you the FEOC sourcing rules are fighting the exact price gap that made Chinese cells attractive in the first place. Ember's independent October 2025 assessment, stripping out China and the US, put ex-those-two-markets capex around $125/kWh with a levelized cost of storage near $65 per megawatt-hour. Every one of those numbers moved in the direction of cheaper storage throughout 2025, tax bill or no tax bill.

The federal government's own forecasters expect the build-out to keep accelerating on the back of that cost curve. The Energy Information Administration's February 20, 2026 preliminary generator inventory projects 86 gigawatts of new utility-scale capacity coming online in 2026, a record if it holds, split 51% solar (43.4 GW), 28% battery storage (24 GW), and 14% wind. That would top 2025's already-record 53 GW, the largest single-year addition since 2002. Solar and storage together, not wind, not gas, are carrying the American grid's growth right now, and roughly 48% of storage already on the grid sits co-located with a solar array specifically to smooth output and shift it into the evening peak.

Cheaper batteries and a friendlier phase-down schedule than solar itself received. On paper, 2026 should be storage's best year yet.

So why is the queue five years long?

Because the tax code was never the bottleneck, the grid was. The Lawrence Berkeley National Laboratory's most recent interconnection queue accounting, Queued Up: 2025 Edition, published in December 2025 and covering data through the end of 2024, puts roughly 1,400 gigawatts of generation and another 890 gigawatts of storage capacity in active queues nationwide, with solar and storage together making up more than 80% of that combined backlog (see the American Public Power Association's summary of the report). Total queue volume actually fell 12% year over year, the first real contraction in years, and that is not good news: more than 700 gigawatts of capacity withdrew from queues in 2024 alone, outpacing roughly 500 gigawatts of new submissions, as developers gave up on stalled projects rather than waited them out. Wait times for the projects that do survive kept lengthening anyway. The same "Queued Up" series has tracked median request-to-operation timelines stretching from under two years for projects built in the 2000s to more than four years for projects completed between 2018 and 2024, with 2023-vintage completions running a median of five years, building on a trend Utility Dive's coverage of the lab's earlier findings had already flagged hardening years ago. Congested markets stretch that further still, to six or eight years even for projects with clean site control and financing already lined up. Of everything that entered a queue nationally between 2000 and 2019, only 13% had actually reached commercial operation by the end of 2024; 77% had been withdrawn along the way. Most of what shows up in a queue count is a number on a spreadsheet, not a project.

That is the real gate on 2026's record forecast. The EIA's 86-gigawatt number describes what developers plan to bring online, not what the queue will actually let through on schedule, and those two figures have diverged before.

And the county commission is getting less friendly, not more

The other constraint, local siting authority, moved the same direction as the interconnection queue: no easier for a developer trying to get a project built. By the end of 2025, roughly 24% of U.S. counties carried some impediment to new utility-scale wind or solar, up from as few as 15% only two years earlier, per a USA Today analysis republished through RealClearEnergy. Columbia University's Sabin Center put a harder number on the same trend in its fifth annual report, published July 1, 2025: at least 459 counties and municipalities across 44 states had adopted severe local restrictions on renewable siting by the end of 2024, a 16% increase in a single year, alongside 498 separately contested projects across 49 states and 20 state-level restrictions in 16 states (see the Sabin Center's report summary). The same report cites outside research finding at least 30% of utility-scale wind and solar projects were cancelled during siting between 2018 and 2023, mostly to community opposition and local zoning rather than economics. Ohio's Richland County is the sharpest recent example: after commissioners imposed an initial solar and wind ban in 2025 citing agricultural land preservation, voters used the citizen-referendum mechanism built into the state's 2021 Senate Bill 52 to make it stick at the ballot box on May 5, 2026, passing the ban in 11 of the county's 18 townships with just under 53% in favor, per Ideastream Public Media's coverage of the vote, a pattern we have tracked county by county. Battery storage carries its own separate opposition vector on top of that: our Escondido case file documents a 320-megawatt, 1,280-megawatt-hour AES battery project killed outright after two nearby fires triggered evacuations, a city moratorium, and an extended moratorium even after AES cut proposed capacity 20% and widened its safety setbacks.

None of that has anything to do with the FEOC sourcing table or the Section 48E construction deadline. Voters banning a solar field in a farm township are not reading the tax code. They are reading a fire report from a neighboring city, or simply protecting agricultural land the way their county always has. The tax law and the county ballot measure are two entirely separate risk systems, and 2025-2026 happened to move both of them against developers at the same time, for reasons that have nothing to do with each other.

What does a developer actually do with two unrelated headwinds hitting at once?

Underwrite them separately, because they resolve on different timelines and respond to different levers. The FEOC and construction-deadline risk is a procurement and scheduling problem: lock construction starts ahead of the July 2026 cliff, and diversify cell suppliers now rather than betting the sourcing thresholds get relaxed. The interconnection and county-siting risk is not solvable with better paperwork on the same timeline; it requires screening jurisdictions for the specific local posture, referendum mechanisms like Ohio's SB 52, BESS-specific fire-safety ordinances, agricultural land preservation politics, before spending real money on interconnection studies that take years to resolve either way. Our own solar coverage reads county by county for exactly this reason, because a national capacity forecast like the EIA's 86 gigawatts tells a developer nothing about whether the specific township next to a proposed array already has a referendum petition circulating.

My call: by 2028, more counties will have adopted a version of Ohio's citizen-referendum lock-in than will have repealed an existing solar or storage restriction, meaning the 24% impediment share keeps climbing even as battery costs keep falling and the tax credit fight fades from the news. If that share flattens or reverses by then, I underestimated how much of 2025's local opposition was itself temporary noise. Either way, price the county before you price the battery. The battery is only getting cheaper. The county is not getting friendlier.

This analysis is a source-cited research summary drawn from public records and industry reporting, not legal or financial advice. It can contain errors and should be verified independently before any investment decision.