Commerce’s July 6 publication closes the countervailing-duty side of the anti-dumping and countervailing-duty case on crystalline silicon photovoltaic cells from India, Indonesia, and Laos, and the finals track the April preliminary determinations within a narrow band on early readings from the parties. The AD companion finals are on the July 13 calendar. The pair of publications resolves a trade-remedy question that has been sitting on top of every utility-scale solar procurement decision in the United States since the petition was filed in September 2025, and it does so 48 hours after the July 4 begin-construction deadline that fixed the safe-harbored pool at Wood Mackenzie’s 216 to 240 GW DC tally. The interaction between the tariff stack and the safe-harbored pool is the story worth carrying forward.
What the finals actually say
The countervailing-duty finals published July 6 apply to crystalline silicon photovoltaic cells, whether or not assembled into modules, imported from India, Indonesia, and Laos. Preliminary CVD margins from the April 2026 determination ran 15.3 to 41.7 percent across the top three lanes, with country-wide “all others” rates set at 24.6 percent for India, 33.8 percent for Indonesia, and 41.7 percent for Laos. Individual-respondent rates published on named producers ran below the country-wide numbers on India and Indonesia and at or near the country-wide on Laos. Early readings from counsel involved in the case suggest the finals will land within a 3 to 6 percentage-point band of the prelims, with the direction of adjustment mixed by respondent and program.
The AD side of the case, publishing July 13, carries preliminary dumping margins in a wider range: 21 to 87 percent on India, 35 to 219 percent on Indonesia, and 76 to 271 percent on Laos, with the top of each range reflecting adverse-facts-available applied to non-cooperative respondents. The AD finals are the larger number in the stack and are the operative constraint for most importers because the countervailing side does not stack additively with the AD margin in the delivered-cost math the way non-specialists sometimes read it. Both duties apply, cash deposits are collected at the sum of the two margins, and the combined stack is what shows up in the landed-cost calculation on any post-July-4 shipment.
The July 6 CVD finals are the smaller of the two publications by margin size but are the first one to become a final Commerce determination that parties can appeal to the Court of International Trade. The appeal window on the CVD finals opens July 6 and runs 30 days for the summons and 60 days for the complaint. That timing matters because it fixes the trajectory of the litigation calendar for the rest of 2026, and because settlement discussions between US petitioners and named foreign producers historically move most sharply in the 30 to 90 day window after finals.
What the three-country lane math does to the delivered cost curve
The safe-harbored solar pool of 216 to 240 GW DC that closed on July 4 is not a pool of built modules. It is a pool of projects that have established begin-construction status through physical-work or 5 percent cost-incurred pathways, with a four-year continuity-of-construction window that runs to July 2030. The modules that will actually go on those projects have not been fully sourced. Roughly 60 to 70 percent of the pool has firm module-supply agreements in place. The residual 30 to 40 percent sits on letters of intent, framework agreements, or open procurement.
The lane redirection question is where the residual 30 to 40 percent will source. India, Indonesia, and Laos together carried roughly 42 percent of US solar cell imports through Q1 2026, up from a combined 8 percent in Q1 2023 when the case was still on the Cambodia, Malaysia, Thailand, and Vietnam quartet. The three-country pool absorbed the redirect from the earlier case. The July 6 and July 13 finals now close the second redirect.
Three lanes remain available in size, and each carries its own constraint.
Domestic US cell production sits at roughly 14 GW annualized capacity as of Q2 2026, up from 2.5 GW in Q1 2024, with First Solar’s Alabama and Louisiana ramps carrying most of the increment. That number is set to reach 22 to 26 GW by end-2027 on announced buildouts. Domestic supply is the FEOC-clean option under IRS Notice 2026-15 and carries the Section 45X advanced-manufacturing credit stack on the sell side, which flows through to lower delivered price by 4 to 6 cents per watt. Domestic supply is also fully allocated through Q4 2027 on current bookings. Post-July-4 procurement into the 30 to 40 percent unfilled tranche will not clear against domestic capacity alone.
Korea and Turkey together offer roughly 11 GW of cell capacity that is uncovered by the current AD/CVD stack and that clears the FEOC test under Notice 2026-15 on ownership and license-agreement grounds. Korean cell production has been re-oriented toward US-bound flow since Q4 2025, with Hanwha Qcells’ Georgia integrated line ramping in parallel. Turkey has emerged as a smaller but growing lane on the back of European module demand. Both are priced 8 to 14 cents per watt above the pre-tariff Southeast Asia benchmark and 3 to 5 cents above the AD/CVD-adjusted price on the surviving Southeast Asia flow, once the July 6 and July 13 finals are stacked.
Residual Southeast Asia flow at the AD/CVD-adjusted price is the third lane, and it is the mathematically largest option for the unfilled 30 to 40 percent. India, Indonesia, and Laos production will not disappear at the July 13 finals. The AD/CVD stack raises the landed cost by 30 to 90 cents per watt depending on lane and respondent, and it removes the pricing advantage the Southeast Asia lane had carried against Korean and Turkish cells, but it does not remove the physical supply. Where the FEOC test under Notice 2026-15 clears (which requires walking the ownership and license-agreement chain on the producer, not just the country of manufacture), the AD/CVD-adjusted Southeast Asia lane remains a viable procurement option on price parity with Korea and Turkey.
Where the FEOC bright line and the trade case intersect
IRS Notice 2026-15 hardened its FEOC bright line on July 4 as part of the same-day OBBBA compliance package. The notice moves any storage license, module license, or cell license with a specified foreign entity into the effective-control test for Section 48E, and it applies to placed-in-service events after July 4, 2026 rather than to begin-construction status. The Southeast Asia solar cell pool carries a mix of FEOC exposures on the license side. Roughly 55 to 65 percent of Indonesia and Laos cell production and 40 to 50 percent of India cell production sits on module-design licenses or process-technology licenses that trace back to specified foreign entities. The remainder is on independent process technology or on licenses from non-FEOC jurisdictions.
The intersection with the July 6 and July 13 trade case matters because a developer sourcing post-July-4 residual Southeast Asia cells has to clear two independent tests. The AD/CVD stack sets the price. The FEOC bright line under Notice 2026-15 sets the credit eligibility on the placed-in-service event. A cell that clears the AD/CVD math at 30 cents per watt of stacked duty but that fails the Notice 2026-15 effective-control test will disqualify the 48E credit on the project it is deployed into. The two tests are being read together in Q3 2026 procurement decisions, and the projects moving fastest are the ones that carry the physical-work BOC on the safe-harbored pool and the FEOC-clean license documentation on the cell source.
What Q3 2026 procurement will actually clear
The 30 to 40 percent unfilled tranche of the 216 to 240 GW safe-harbored pool, which is roughly 65 to 96 GW DC of module supply that needs to be sourced against a July 2030 continuity-of-construction end date, will clear across the three lanes in a mix that the July 6 and July 13 finals will begin to price this week.
The observable early signal is on the Korea and Turkey lane. Q3 2026 procurement letters of intent on Korean and Turkish cell capacity have accelerated sharply since the April preliminary determinations, with named US developers (NextEra, AES, Recurrent, Enlight, and Doral) taking option positions on 8 to 12 GW of Korean cells for 2027 and 2028 delivery. The Korean lane is filling faster than domestic capacity, and the pricing on the option positions is closing the gap to domestic cells as domestic capacity remains sold out through 2027.
The Southeast Asia residual lane is quieter but is not going away. The delivered-cost math on the AD/CVD-adjusted Southeast Asia cells sits roughly in parity with Korea and Turkey after the July 13 stack, and the physical supply is materially larger. The projects that will pull on the residual lane are the ones that carry clean Notice 2026-15 FEOC documentation on the license side, and those tend to be developers that spent 2024 and 2025 mapping their cell-supply licenses independently of the country-of-origin question. The mapping effort was the underappreciated diligence of the pre-July-4 cycle, and it is where the residual Southeast Asia procurement window opens.
What to watch through July and August
Three datapoints will calibrate the delivered-cost curve on the safe-harbored pool over the next 60 days.
The July 13 AD finals publication and the appeal-window filings that follow. If the AD finals track the preliminary determinations within a 5 to 10 percentage-point band and no major respondent files a Court of International Trade appeal in the first 30 days, the delivered-cost math is fixed for the rest of 2026 and Q3 2026 procurement will move at pace. If a large named respondent files a fast appeal seeking preliminary injunction relief on part of the case, the delivered-cost math holds but the litigation overhang widens the risk premium on Southeast Asia sourcing.
The Q3 2026 domestic cell capacity ramp reporting from First Solar, Hanwha Qcells Georgia, and Silfab Utah. If domestic capacity comes online at or ahead of the guided ramp schedule, the 22 to 26 GW end-2027 number moves inside a delivery window that matters for the safe-harbored pool. If domestic ramps slip by two to three quarters, the residual Southeast Asia lane absorbs more of the 65 to 96 GW unfilled tranche than the current base case assumes.
The Q3 2026 Notice 2026-15 guidance thread from IRS and Treasury. The FEOC effective-control test on cell licenses is being interpreted case-by-case in early July as developers sort their license documentation. A Q3 clarification or a first published ruling on a representative license structure will compress the interpretive risk on the residual Southeast Asia lane. Silence through the end of Q3 keeps the interpretive risk wide and shifts more of the unfilled procurement toward Korea and Turkey.
The shape of the question
The July 6 CVD finals close one half of the trade-remedy case that has been sitting on top of US utility-scale solar procurement for nine months. The July 13 AD finals close the other half. Together they redirect the residual 30 to 40 percent of the safe-harbored pool onto a three-lane sourcing mix (domestic, Korea and Turkey, and residual Southeast Asia at AD/CVD-adjusted prices) that will price out over Q3 and Q4 2026. The FEOC bright line under Notice 2026-15 sits alongside the trade case as a second independent test, and the interaction is where the 2027 and 2028 delivered-cost curve is being written this month.