FSLR 10-K Analysis: The $1.6 Billion Subsidy Behind First Solar's Margins

First Solar, the only US-headquartered manufacturer of utility-scale solar modules, reported $5.2 billion in revenue and 40.6% gross margins in FY2025. Free cash flow swung from negative $308 million to positive $1.2 billion. Production of cadmium telluride thin-film modules — a technology that bypasses Chinese polysilicon supply chains entirely — reached 17.5 GW in shipments. By every headline metric, First Solar looked like a manufacturing success story firing on all cylinders.

The 10-K tells a different story. Buried in the cost of sales bridge is a disclosure that reframes the entire profitability narrative: $1,600 million in Section 45X advanced manufacturing production credits — classified as a reduction to cost of sales — constitute 75.5% of reported gross profit. Strip them out, and the 40.6% gross margin drops to approximately 10%. The credits don't generate excess profit; they almost exactly offset the cost premium of manufacturing in the United States. And as of July 4, 2025, the One Big Beautiful Bill Act has "severely limited" these credits by enacted law — not risk factor, but statute.

The question for investors isn't whether First Solar makes good solar panels. It's whether a three-layer competitive barrier — AD/CVD tariff duties up to 3,404%, a TOPCon patent offensive against the ten largest crystalline silicon importers, and FEOC restrictions that disqualify Chinese-origin modules from tax credit eligibility — can convert subsidy dependence into monopoly pricing power before the credits phase down. The filing provides the data to test this thesis. The answer is more complicated than either bulls or bears expect.

The Subsidy Anatomy: Where 75% of Gross Profit Actually Comes From

First Solar's reported gross profit of $2,120 million on $5,219 million in revenue looks like the output of a highly efficient manufacturer. But the filing's cost of sales decomposition reveals that $1,600 million of that gross profit — roughly three out of every four dollars — comes from Section 45X manufacturing credits classified as a reduction to cost of sales, not from the manufacturing process itself.

The per-watt economics make the dependence concrete. First Solar sold 17.5 GW for $5,219 million, yielding an average selling price of $0.298 per watt. Reported cost of goods sold was $0.177 per watt — a number that already includes the 45X credit offset. Add back the $1,600 million in credits, and true manufacturing cost rises to $0.268 per watt. The organic gross margin — what First Solar actually earns from making solar modules — is $0.030 per watt, or approximately 10%.

Put differently: if First Solar sold 1 GW for $298 million, the filing shows $177 million in reported COGS — leaving $121 million in gross profit. Of that $121 million, approximately $91 million comes from Section 45X credits. The true manufacturing profit is about $30 million. The other 30.6 percentage points of reported gross margin? Subsidy.

The COGS bridge in the MD&A confirms this isn't coincidence — it's structural. Non-volume US production cost increases totaled $614.7 million: a $216.5 million production premium from higher US manufacturing mix, $173.1 million in logistics including detention and demurrage charges, $130.7 million in warehousing, and $94.4 million in tariffs on raw materials. The incremental 45X credit offset was $601.8 million — a 97.9% match.

"Cost of sales increased $750.6 million, or 32%...driven by (i) higher costs of $651.6 million due to an increase in the volume of modules sold; (ii) higher production costs of $216.5 million, largely due to a higher sales mix of U.S.-produced modules and tariffs on raw materials; (iii) higher logistics costs of $173.1 million, which included detention and demurrage charges; (iv) higher warehousing costs of $130.7 million; and (v) tariffs on international modules imported into the United States of $94.4 million. These increases were partially offset by (vi) a higher sales mix of modules qualifying for the advanced manufacturing production credit under Section 45X of the IRC, which decreased cost of sales by $601.8 million."

This matters because the credits are no longer secure. The One Big Beautiful Bill Act, signed into law on July 4, 2025, fundamentally altered the landscape.

"On July 4, 2025, the U.S. President signed H.R.1 into law, commonly referred to as the 'One Big Beautiful Bill,' which significantly curtails the availability of certain energy tax credits. H.R.1 also severely limits Section 45X tax credits."

The investment implication is direct: any phase-down of 45X credits erodes gross margin dollar-for-dollar unless First Solar can raise ASPs by $0.06 to $0.09 per watt through its competitive barriers. At $0.030 per watt of organic margin, the manufacturing business generates roughly $525 million in gross profit on 17.5 GW — enough to cover operating expenses, but leaving minimal operating income. First Solar's $1,600 million in Section 45X manufacturing credits constituted 75.5% of reported gross profit in FY2025, with a per-watt decomposition revealing the credits almost exactly offset the $614.7 million US production cost premium — leaving an organic gross margin of approximately 10%.

The Three-Layer Moat: Can Competitive Barriers Replace Subsidies?

If Section 45X credits are a cost equalizer rather than a margin enhancer, First Solar's investment case pivots entirely to whether competitive barriers can convert subsidy dependence into protected pricing power. The filing reveals the most comprehensive competitive barrier structure in US solar manufacturing — three simultaneous layers that, if all hold, could make First Solar the de facto sole supplier of fully domestic, subsidy-eligible utility-scale modules.

Layer 1: Trade Barriers. Anti-dumping and countervailing duty final determinations have imposed extraordinary rates on Southeast Asian crystalline silicon imports — the channel through which Chinese manufacturers access the US market. The rates are not modest adjustments; they are effective import bans.

"Following final affirmative determinations by the USDOC and USITC that identified final subsidy rates of 534.67% to 3,403.96% for Cambodia, 14.64% to 168.8% for Malaysia, 263.74% to 799.55% for Thailand, and 68.15% to 542.64% for Vietnam."

At 534% to 3,404% on Cambodian-origin modules alone, these duties make Chinese-manufactured c-Si panels economically unviable in the US market. No pricing strategy can absorb a 5x to 34x tariff multiplier.

Layer 2: IP Enforcement. On February 24, 2026, First Solar filed an ITC Section 337 petition against 10 of the largest crystalline silicon module importers — Axitec, Canadian Solar, JA Solar, JinkoSolar, Mundra, Philadelphia Solar, Hanwha QCells, Runergy, Trina Solar, and VSUN — seeking to block TOPCon technology imports. Three additional district court patent suits were already filed against JinkoSolar, Canadian Solar, and Mundra/Adani. If the ITC grants a technology exclusion order, it would block the dominant next-generation c-Si technology from the US market entirely.

Layer 3: Regulatory Restrictions. The OBBBA's FEOC provisions restrict tax credit eligibility for products with Foreign Entity of Concern supply chain ties. Since virtually all crystalline silicon manufacturing has some Chinese supply chain connection — from polysilicon to cell fabrication — this effectively disqualifies most imported modules from US incentive programs.

The OBBBA creates a fascinating dual impact: it simultaneously limits First Solar's 45X credits and restricts competitors' access to the US market. Management acknowledges this in the filing, noting that US module pricing "has remained stable due, in part, to...energy tax credit eligibility restrictions (including foreign-entity-related limitations) as amended by the OBBBA." The competitive barriers may be enabling price stability even as the subsidy foundation erodes.

The critical vulnerability is that each layer faces challenge. AD/CVD duties could be reduced in future reviews. The ITC Section 337 petition is a binary outcome with no guaranteed timeline — historically, such decisions take 12 to 18 months, suggesting a ruling by Q3 to Q4 2027 at the earliest. And the FEOC restrictions depend on enforcement rigor that could shift with political priorities. There's also a counter-vulnerability: China announced tightened export controls on tellurium-containing products in February 2025, and First Solar's CdTe technology requires tellurium as a critical raw material. The three-layer moat has its own supply chain crack.

First Solar's ITC Section 337 petition names 10 of the largest crystalline silicon importers as respondents, while AD/CVD duties of 534% to 3,404% on Cambodian-origin modules effectively function as import bans on Chinese-manufactured solar panels.

The Cash Flow Illusion: Headline Growth Reverses Sign

First Solar's cash flow narrative in FY2025 was compelling on the surface: operating cash flow surged 69% to $2,057 million, and free cash flow swung from negative $308 million to positive $1,187 million — a $1,495 million improvement. For investors focused on cash generation, this looked like a company hitting its stride.

But the 10-K reveals that the headline numbers include $1,241 million in Section 45X credit monetization proceeds. First Solar sold $1,401.6 million in 45X tax credits in two tranches — $701.9 million sold for $668.1 million in June/July 2025, and $699.7 million sold for $668.2 million in October 2025 — receiving $1,241 million in cash during the fiscal year (with $95.2 million remaining as a receivable in Q1 2026). The credits monetized at 95.3 cents on the dollar, validating their quality. But these proceeds flow through operating cash flow, inflating the headline number.

Strip out the 45X credit sale proceeds, and organic operating cash flow was approximately $816 million — a 33% decline from FY2024's $1,218 million. The cash flow growth narrative doesn't just moderate; it reverses sign entirely.

Two additional one-time factors inflated the FY2025 cash flow picture. First, a non-repeatable inventory drawdown.

"During 2025, we produced 16.1 GW and sold 17.5 GW of solar modules."

Selling 1.4 GW more than produced means FY2025 revenue included prior-period inventory at today's prices — a boost that cannot repeat at the same magnitude. Second, capital expenditures dropped 43% from $1,526 million in FY2024 to $870 million, driving 44% of the total FCF swing. This was an investment cycle transition (capex-to-depreciation ratio collapsed from 3.6x to 1.6x), not a permanent improvement in capital efficiency.

The implication for 2026 is direct: organic cash flow is the metric that matters. If OCF excluding 45X credit monetization can't exceed $1.0 billion, the market will reprice the cash generation narrative. FY2026 guidance of $4.9 to $5.2 billion in revenue — flat to down versus FY2025 — combined with capex guidance of $800 million to $1.0 billion, suggests organic FCF will be tested. Investors anchoring on the headline $1.2 billion FCF are pricing in a subsidy-inflated number that the 10-K shows is not representative of underlying cash generation.

First Solar's organic operating cash flow — excluding $1,241 million received from Section 45X credit sales — declined 33% from $1,218 million in FY2024 to approximately $816 million in FY2025, reversing the headline narrative of 69% cash flow growth.

The Backlog Paradox: $15 Billion in Contracts, Flat Revenue Guidance

First Solar's contracted backlog of 50.1 GW with an aggregate transaction price of $15.0 billion — extending through 2030 — represents approximately 2.9 years of FY2025 revenue. For any manufacturer, that kind of visibility would typically anchor a growth narrative. But the filing reveals structural complications that undermine the backlog's reliability as a revenue predictor.

The most significant disclosure is buried in the MD&A: nearly half the backlog has variable pricing.

"As of December 31, 2025, we had entered into contracts with customers for the future sale of 50.1 GW of solar modules for an aggregate transaction price of $15.0 billion...This volume includes contracts for the sale of 23.2 GW of solar modules with anticipated price adjustments for future module technology improvements."

Those 23.2 GW — 46.3% of the total backlog volume — have price adjustments contingent on First Solar delivering CuRe technology improvements that promise better bifaciality, temperature coefficients, and warranted degradation rates. CuRe is just beginning its production ramp: limited commercial production started in late 2024, first customer sales occurred in H1 2025, and one Ohio facility will permanently convert to CuRe production starting Q1 2026. The technology has not yet been proven at scale.

If CuRe achieves only 80% of expected ASP lift — a conservative technology disappointment scenario — the variable contracts would contribute approximately $6.4 billion instead of an estimated $6.9 billion, reducing effective backlog to roughly $14.5 billion. This isn't catastrophic, but combined with flat 2026 guidance, it implies revenue doesn't recover above $5.2 billion until 2027 or 2028.

The backlog fragility extends beyond technology risk. First Solar terminated contracts with BP Solar/Lightsource for breach, triggering $384.6 million in termination provisions and filing a $323.6 million complaint on September 30, 2025. The company itself acknowledges that contract terms "have in the past and may in the future be breached by our customers or subject to renegotiation." Additionally, glass substrate supply agreements carry termination penalties up to $300 million, locking First Solar into production commitments regardless of demand conditions.

The most telling signal is the guidance itself. If $15 billion in backlog truly represented 2.9 years of firm revenue, FY2026 guidance should reflect growth — not a potential decline. The midpoint of $5.05 billion implies an ASP of approximately $0.287 per watt at estimated volumes, a 3.7% compression from FY2025's $0.298 per watt. The backlog is not converting to revenue at the rate the headline number suggests.

First Solar's 50.1 GW contracted backlog includes 23.2 GW — 46.3% of total volume — with anticipated price adjustments tied to undelivered CuRe technology improvements, creating a gap between the $15 billion headline value and the economically committed revenue base.

The Valuation Verdict: A $21 Billion Policy Risk Premium

At $261.23, First Solar trades at 18.3x trailing earnings — roughly half the cross-sector peer median of approximately 32x. This discount is the market's price for policy risk. The question is whether it's justified, excessive, or insufficient.

At peer-median P/E of approximately 32x, First Solar's FY2025 EPS of $14.21 would imply a stock price of $455 — a 74% premium to the current price. That $193 per share gap, multiplied across 107.2 million shares outstanding, quantifies the policy risk premium: approximately $21 billion in lost market cap relative to what the market would pay for identical financial performance without the subsidy overhang.

But framing this as a simple discount-to-peers misses the binary nature of the investment. First Solar is effectively two different companies depending on the outcome of its competitive barriers and subsidy trajectory.

In the bull scenario — where 45X credits persist at meaningful levels through 2029-2030 and the three-layer competitive barrier holds — First Solar is a $14 per share earnings company with a protected domestic monopoly, a fortress balance sheet ($2.3 billion net cash, zero US debt, 2.67x current ratio), and improving returns as CuRe technology converts variable backlog at premium pricing. At that earnings power, even 18x is cheap.

In the bear scenario — where 45X phase-down accelerates to 2027, the ITC Section 337 petition is denied, and AD/CVD rates are reduced — First Solar becomes a 10% gross margin manufacturer with approximately breakeven operating income. Adjusted gross profit at FY2026 guidance midpoint would be roughly $505 million, against operating expenses of approximately $520 million. Earnings would collapse to $2 to $4 per share, and the current price would represent a premium, not a discount.

The binary catalyst with a defined timeline is the ITC Section 337 ruling on the TOPCon technology exclusion order. Based on historical timelines for similar patent-based exclusion proceedings, a decision is expected by Q3 to Q4 2027 — approximately 12 to 18 months from the February 2026 filing. If granted against all 10 respondents, it would block the dominant next-generation c-Si technology from the US market, giving First Solar monopoly pricing power on domestic utility-scale solar for years. If denied, the competitive barrier loses its strongest pillar.

The $21 billion policy risk premium embedded in First Solar's 41% valuation discount will compress or expand based on two catalysts with 12 to 18-month horizons: the ITC Section 337 decision on TOPCon technology exclusion, and the OBBBA phase-down implementation trajectory on residual 45X credit availability. Investors have a defined window before the verdict.

At 18.3 times trailing earnings, First Solar trades at a 41% discount to its cross-sector peer median of approximately 32 times, implying the market assigns roughly $21 billion in policy risk premium — a gap that collapses if the three-layer competitive barrier enables margin preservation through pricing power rather than subsidies.

What to Watch

The filing data points to five tracking metrics that will determine which version of First Solar emerges over the next 12 to 18 months:

• Q1 2026 ASP trajectory: FY2025 ASP was $0.298/W. Guidance implies compression to approximately $0.287/W. If ASP holds above $0.295/W despite 45X uncertainty, pricing power exists. If it falls below $0.280/W, competitive barriers aren't compensating for subsidy erosion.

• Organic OCF (ex-45X monetization): FY2025 organic OCF was approximately $816 million. Quarterly organic OCF above $250 million confirms sustainable cash generation. Below $150 million signals the cash flow narrative was entirely subsidy-driven.

• CuRe production ramp: One Ohio facility is permanently converting in Q1 2026. Monitor for production disruption disclosures, quality metrics, and whether variable backlog contracts begin converting at anticipated prices.

• ITC Section 337 timeline: Track procedural milestones — investigation institution, hearing dates, initial determination. Any acceleration or delay changes the binary catalyst window.

• Backlog conversion rate: The 50.1 GW / $15.0B backlog implies $0.299/W blended ASP. If quarterly bookings-to-billing ratios drop below 1.0x, the backlog is eroding rather than converting. Watch for additional contract terminations following the BP Solar precedent.

At $261, the market implies that First Solar's three-layer competitive barrier has a meaningful probability of failure — pricing roughly even odds between the $14 EPS and $2-4 EPS scenarios. The filing supports both possibilities but complicates the simplistic narrative in either direction. What makes this stock genuinely binary is that the answer arrives on a defined schedule: the ITC ruling and OBBBA phase-down trajectory will resolve the key uncertainties within 12 to 18 months. That is a rare analytical luxury.

Frequently Asked Questions

What are Section 45X manufacturing production credits, and how dependent is First Solar on them?

Section 45X credits are advanced manufacturing production credits under the Inflation Reduction Act that pay approximately $0.17 per watt for solar modules produced in the United States and sold to third parties. In FY2025, First Solar recognized approximately $1,600 million in 45X credits as a reduction to cost of sales — representing 75.5% of the company's $2,120 million in reported gross profit. Without these credits, First Solar's gross margin would have been approximately 10.0% instead of the reported 40.6%. The credits are now "severely limited" by the One Big Beautiful Bill Act (OBBBA), signed into law on July 4, 2025.

What is First Solar's actual manufacturing margin without government subsidies?

Using FY2025 data, First Solar's average selling price was $0.298 per watt (17.5 GW sold for $5,219.4 million). Adjusted cost of goods sold — adding back $1,600 million in 45X credits — was $0.268 per watt. This yields an organic gross margin of approximately $0.030 per watt, or about 10.0%. The filing's COGS bridge confirms that $614.7 million in non-volume US production cost increases were almost exactly offset by $601.8 million in incremental 45X credits.

Did First Solar's cash flow really improve as much as the headline suggests?

No. Headline operating cash flow grew 69% to approximately $2,057 million. However, $1,241 million came from Section 45X credit monetization proceeds (two sale tranches: $668.1 million and $573.0 million). Excluding these proceeds, organic operating cash flow was approximately $816 million — a 33% decline from FY2024's $1,218 million. Additionally, First Solar sold 17.5 GW but produced only 16.1 GW, drawing down 1.4 GW of inventory — a non-repeatable boost to FY2025 revenue.

How reliable is First Solar's $15 billion backlog?

The 50.1 GW / $15.0 billion contracted backlog extends through 2030 and represents approximately 2.9 years of FY2025 revenue. However, 23.2 GW (46.3% of total volume) includes anticipated price adjustments for future module technology improvements — effectively variable pricing tied to undelivered CuRe technology. FY2026 guidance of $4.9-5.2 billion (flat to down) despite the massive backlog suggests conversion timing and contract modification risks. The BP Solar/Lightsource termination ($384.6 million provision triggered) demonstrates that backlog contracts can be breached.

What competitive barriers protect First Solar from Chinese solar imports?

The FY2025 10-K reveals three simultaneous barriers: (1) AD/CVD duties of 534% to 3,404% on SE Asian c-Si imports effectively functioning as import bans. (2) ITC Section 337 petition filed February 24, 2026 against 10 major c-Si manufacturers seeking to block TOPCon technology imports. (3) OBBBA restricting tax credits for products from Foreign Entity of Concern-linked suppliers. Combined, these barriers could make First Solar the de facto sole supplier of fully domestic, subsidy-eligible utility-scale solar modules in the US.

What is the OBBBA, and how does it affect First Solar?

The One Big Beautiful Bill Act (H.R. 1), signed July 4, 2025, has a dual impact on First Solar. It "severely limits Section 45X tax credits" — First Solar's primary profit driver at $1,600 million in FY2025. Simultaneously, it restricts tax credit eligibility for products with Foreign Entity of Concern supply chain ties, disqualifying most Chinese-origin c-Si modules from US incentive programs. The net impact depends on whether First Solar can raise selling prices to offset lost credits — which the FEOC restrictions and AD/CVD duties may enable by reducing competitive supply in the US market.

How does First Solar's valuation compare to peers?

First Solar trades at 18.3x trailing P/E and 12.3x EV/EBITDA — roughly half the peer group averages of approximately 31x P/E and 26x EV/EBITDA (based on cross-sector peers ABNB, MRVL, MAR, and EMR). This discount implies the market assigns approximately $21 billion in policy risk premium. The discount is justified if 45X credits are substantially reduced and competitive barriers fail to compensate; excessive if the three-layer moat enables a transition to protected domestic pricing power.

What is CuRe technology and why does it matter for the backlog?

CuRe (Copper Replacement) is First Solar's next-generation CdTe module technology promising improved bifaciality, temperature coefficient, and degradation rates. Beginning Q1 2026, one Ohio facility will permanently convert to CuRe production. This matters because 23.2 GW (46.3%) of the $15 billion backlog has price adjustments contingent on CuRe improvements. If CuRe delivers, these contracts realize at higher prices. If it underperforms, the effective backlog value is lower.

What happened with BP Solar/Lightsource, and how does it affect First Solar?

First Solar terminated contracts with BP Solar/Lightsource for breach, triggering $384.6 million in termination provisions. First Solar filed a $323.6 million complaint on September 30, 2025. The company recognized $61.0 million from prior advance payments as revenue. This removed near-term volume from the delivery pipeline, the $323.6 million claim is an uncertain receivable, and it demonstrates that multi-billion-dollar backlog contracts are not ironclad.

Does First Solar have supply chain vulnerability despite avoiding Chinese polysilicon?

Yes. While CdTe technology avoids Chinese polysilicon dependence, it requires tellurium — and China is a major global producer. In February 2025, China announced tightened export controls on tellurium-containing products. Additionally, First Solar has glass substrate supply agreements with termination penalties up to $300 million. The CdTe differentiation has its own supply chain vulnerability that partially offsets the polysilicon independence advantage.

Why is 95.7% of First Solar's revenue from the United States?

International revenue collapsed in FY2025: France fell 82% and Other fell 65%. Only India held stable at $196 million. The concentration reflects IRA/45X incentives creating domestic demand advantages, European markets captured by below-cost Chinese modules, and India's own regulatory barriers — a draft MNRE proposal from November 2025 could raise minimum efficiency requirements above CdTe's current capability. The company has $1.44 billion in international manufacturing assets operating at reduced utilization.

What would make the bear case wrong?

Three conditions would invalidate the thesis that First Solar is primarily a subsidy-dependent manufacturer: (1) ASP holds at or above $0.298 per watt in 2026 without proportional 45X support, proving pricing power exists independent of subsidies; (2) ITC Section 337 petition granted, converting the competitive barrier into enforced domestic monopoly pricing; (3) CuRe technology achieves production targets and the 46.3% variable-price backlog converts at or above expected ASPs. Any two of three would demonstrate that the three-layer moat is converting subsidy dependence into genuine competitive advantage.

Methodology

Data Sources

This analysis draws on three data layers. MetricDuck Pipeline data — revenue, margins, returns, multiples, and cash flow metrics — is extracted from XBRL filings via automated processing and provides the financial foundation. Filing-specific data comes from First Solar's FY2025 10-K filed February 24, 2026, including MD&A decompositions, risk factor disclosures, and verbatim management characterizations. Derived calculations — including per-watt economics, organic cash flow adjustments, and the 45X offset ratio — are documented with explicit formulas and can be verified from the source data. Peer financial data for ABNB, MRVL, MAR, and EMR is sourced from MetricDuck pipeline data for the most recent fiscal year available.

Limitations

• 45X phase-down timeline unknown. The OBBBA "severely limits" 45X credits but the specific phase-down schedule is not disclosed in the 10-K. All post-subsidy margin scenarios are directional, not precisely timed.
• Cross-sector peer limitations. ABNB, MRVL, MAR, and EMR are not solar industry peers. Comparisons are valid for financial benchmarking (margins, returns, multiples) but not for industry-specific dynamics.
• ITC Section 337 probability not modeled. The patent offensive against 10 c-Si importers is described qualitatively. Historical ITC grant rates for technology exclusion orders vary widely.
• CuRe technology performance unverified. The backlog variable pricing analysis (46.3%) depends on CuRe delivering specified improvements. At-scale performance data is not yet available.
• 45X credit classification is GAAP-compliant. The "adjusted" margins presented in this analysis are an analytical construct designed to isolate manufacturing economics from subsidy economics — they are not management's view.
• 2026 GW guidance not in 10-K. Revenue guidance ($4.9-5.2B) is provided but production/shipment volume for 2026 is not disclosed, requiring estimated ASP calculations for forward projections.

Disclaimer:

This analysis is for informational purposes only and does not constitute investment advice. The author does not hold positions in FSLR, MRVL, MAR, ABNB, or EMR. Past performance and current metrics do not guarantee future results. All data is derived from public SEC filings and may contain errors or omissions from the automated extraction process.