July 5, 2026 — less than two weeks away — marks the expiration of the Investment Tax Credit (ITC) "safe harbor" rule under the Inflation Reduction Act (IRA), beginning a nine-year subsidy phaseout schedule. Simultaneously, LONGi announced a certified 35.0% efficiency perovskite-silicon tandem solar cell (1 cm²), reaching 33.0% on large-area (260.9 cm²) modules; US solar electricity generation surpassed coal for the first time in May 2026 (12.8% vs 12.2%); and solar-plus-storage accounted for 91% of all new US generating capacity in Q1 2026. These seemingly independent events collectively trace the outline of a structural inflection point for the solar industry — under the pressure of subsidy withdrawal, technological innovation and market demand are converging to push solar into genuine unsubsidized competition.
The Policy Clock: Structural Impact of the ITC Phaseout
The US ITC has been the core policy driver behind the explosive growth of US solar installations over the past decade. Under the IRA 2022 timeline, the 30% ITC was scheduled to begin phasing down in 2032, reaching zero by 2035. However, the OBBBA (Orbital Budget Baseline Alignment Act) accelerates this timeline, beginning ITC reduction on July 5, 2026 — six years earlier than the original plan:
| Period | ITC Rate (Original IRA) | ITC Rate (OBBBA) | Difference |
|---|---|---|---|
| Before July 2026 | 30% | 30% | — |
| Jul 2026 - Jun 2027 | 30% | 26% | -4% |
| Jul 2027 - Jun 2028 | 30% | 22% | -8% |
| Jul 2028 - Jun 2029 | 30% | 18% | -12% |
| Jul 2029 - Jun 2030 | 30% | 14% | -16% |
| After Jul 2030 | Phase to 0% | 10% → 0% | Accelerated exit |
For a typical 200 MW utility-scale solar plant (approximately $200 million investment), a 4% ITC reduction represents $8 million in lost tax benefits. This loss will be absorbed through multiple mechanisms: partial manufacturer price reductions, partial developer margin compression, and partial compensation from improved system efficiency.
LONGi 35%: The Perovskite Tandem Efficiency Revolution
In June 2026, LONGi announced its perovskite-silicon tandem solar cell had reached 35.0% certified efficiency — the highest official certification ever achieved for this technology pathway. Even more notably, LONGi also reported 33.0% efficiency on large-area (260.9 cm²) modules.
The Lab-to-Fab Gap is Narrowing: Perovskite-silicon tandems have long been considered "the next technology always in the future" — lab efficiency records falling frequently, but mass production consistently plagued by stability issues. LONGi's large-area module efficiency being only 2 percentage points below its small-area cell suggests the company has achieved critical breakthroughs in uniform deposition, defect passivation, and encapsulation technology.
| Metric | LONGi (2026) | Oxford PV (2025) | Conventional Silicon |
|---|---|---|---|
| Small-area certified efficiency | 35.0% | 28.6% | 27.3% (LONGi HBC) |
| Large-area module efficiency | 33.0% | 26.0% (commercial) | 24.5% (commercial) |
| Stability (T80) | Undisclosed | >1,000 hours | >25 years |
| Estimated mass production timeline | 2-4 years | Early commercial shipments | Mass production |
| Cost premium | +30-50% estimated | +50-80% | Baseline |
Oxford PV began small-volume commercial shipments in 2025-2026, becoming the first company to bring perovskite-silicon tandem products to market. LONGi's higher efficiency suggests that 2027-2028 could see multiple manufacturers' tandem products entering the market simultaneously — similar to the PERC technology diffusion curve from lab to fab in the 2010s.
Industry Milestone: Solar Surpasses Coal
In May 2026, US solar electricity generation surpassed coal for the first time — 12.8% vs 12.2%. This figure represents the convergence of multiple forces:
Exponential Capacity Growth: US utility-scale solar capacity grew 20% year-over-year compared to May 2025. In Q1 2026, solar-plus-storage accounted for 91% of all new generating capacity, with pure solar at 60%. For every 10 MW of new capacity added, 6 MW came from solar.
Coal's Irreversible Decline: Coal's share of the US electricity mix has fallen from approximately 45% in 2010 to 12.2% in 2026. This is not a cyclical fluctuation — since 2015, over 40% of US coal plants have retired or announced retirement plans. Remaining coal plants serve two primary purposes: (1) backup capacity during extreme weather events; and (2) baseload power during winter mornings and evenings when solar generation is minimal.
The Structural Significance of Surpassing Coal: Solar overtaking coal is more than an energy statistics update. It signifies that solar has transformed from an "alternative energy" to a "mainstream energy source" — not because of policy preference, but because of economic reality. In the sun-rich southwestern United States, solar LCOE has fallen to $20-30/MWh, far below coal at $40-60/MWh and combined-cycle natural gas at $30-50/MWh.
Post-Subsidy Demand Structure
The impact of ITC phaseout on solar demand may be buffered by two structural demand waves arriving simultaneously:
Hyperscaler Clean Energy Procurement: Google, Meta, and Microsoft announced approximately 940 MW of new solar procurement agreements in Q2 2026. The total hyperscaler/AI clean energy procurement market reaches tens of GW, with solar representing roughly half. AI data center electricity demand growth — projected to reach 9-10% of total US electricity consumption by 2030 — provides a structural demand driver for solar. Unlike subsidy-driven residential solar markets, hyperscaler PPAs are based on pure economic calculation: large-scale solar LCOE is already below grid electricity prices.
Storage Pairing Becomes Mandatory: The fact that solar-plus-storage accounted for 91% of Q1 2026 new capacity reveals another structural shift — solar plants are no longer deployed in isolation. Four-hour battery storage costs have fallen to approximately $50/kWh, bringing the LCOE of "solar-plus-storage" as a 24-hour dispatchable power source to approximately $40-60/MWh — competitive with natural gas peaker plants.
Observatory Analysis: The Post-Subsidy Solar Race
The July 2026 ITC phaseout initiation is a stress test — not testing whether solar technology works (proven over the past decade), but whether solar's economic model stands without policy support.
Technology Progress is Catching the Subsidy Baton: LONGi's 35% tandem efficiency and Oxford PV's commercial shipments demonstrate that solar module efficiency improvement curves have not plateaued. From 15% commercial module efficiency in 2010 to 24-25% in 2025, and potentially 30-35% by 2030 — efficiency gains directly translate into BOS (balance of system) cost reductions. Each 1% of policy loss offset by efficiency-driven cost reduction buys the industry one more year of adjustment time.
"Solar Oversupply" is a Benign Structural Feature: Solar's seasonal and diurnal variability has always been considered its greatest weakness. But 2026 data reveals a different picture — "oversupply" is itself a resource. In California and Texas, midday solar overgeneration has driven wholesale prices negative, which in turn drives the economics of storage deployment, electrolytic hydrogen production, and flexible loads. Solar is valuable not because it is "perfect," but because it is "abundant."
The US-China Technology Race's Renewable Energy Dimension: LONGi's 35% efficiency record reminds us that despite escalating tariffs and policy barriers, Chinese solar manufacturers remain at the technological frontier. While US IRA subsidies have driven domestic capacity construction, they cannot alter China's accumulated advantages in perovskite research and manufacturing engineering. The global solar market may settle into a dual-track pattern — China dominating high-efficiency module supply, while the US maintains a degree of capacity autonomy through tariffs and local production requirements.
Disclaimer: The information in this article is for reference purposes only and does not constitute investment advice or commercial decision-making basis. Data and time-sensitive information are current as of the publication date and may change with subsequent developments. Neither the author nor POC.HK assumes any liability for losses arising from the use of this information.