First Solar, Inc. (NASDAQ: FSLR) — A Subsidized Cash Machine Priced as a Cyclical

Report date: 2026-06-13 An independent analyst’s note — skeptical, evidence-driven fundamental research.

⚡ Claude’s Take

This block is the author’s own subjective opinion and general information only. It is not investment advice and is not a recommendation to buy or sell any security. The analysis that follows takes no position and sets no price target; only this clearly-labeled block expresses a view.

Verdict: HOLD / accumulate only on a policy-panic selloff (sub-~$190, roughly ≤12× clean earnings / ~1.5× book). Not a short here despite the subsidy dependence. Conviction: medium.

Tag: “The best house in a subsidized neighborhood — own the house, never forget who pays the mortgage.”

First Solar is the rare solar name that is genuinely, structurally profitable — $1.5B of net income, $2.0B of operating cash flow, a $2.4B net-cash balance sheet, and a 50.1 GW contracted backlog stretching past 2030. It is the only at-scale, vertically integrated, non-Chinese module manufacturer on earth, and the entire US policy edifice — the §45X manufacturing credit, AD/CVD tariff walls on Southeast-Asian crystalline silicon, the new FEOC (foreign-entity-of-concern) restrictions, and domestic-content ITC adders — is purpose-built to advantage exactly that profile. That is a real, durable, multi-year earnings stream that the market, with the stock at ~17× trailing earnings and a Wall Street price target (~$244) below the current price, is treating with more skepticism than its near-term cash flows deserve. On the next 2–3 years of visible numbers, this is cheap.

But I cannot call it a high-conviction buy, because the quality of those earnings is the whole debate. In FY2025, First Solar recognized $1.6 billion of §45X tax credits as a reduction of cost of sales — against $2.1 billion of total gross profit and $1.5 billion of net income. Strip the subsidy and the underlying module business runs at roughly a 10% gross margin — a thin, cyclical, commoditized manufacturer whose product is more expensive per watt than Chinese silicon outside the US tariff wall (witness Malaysia/Vietnam plants idling and India modules clearing at $0.20/W vs $0.35/W domestically). The bull thesis and the bear thesis are the same fact viewed across different time horizons: the §45X credit is law through 2032 (phasing down from 2030), FEOC rules are tightening in FSLR’s favor today, and net bookings actually turned negative in 2025 (8.3 GW of debookings against 7.4 GW of gross bookings) while 2026 revenue is guided flat and management pointedly declined to give an EPS number. You are buying a government-policy annuity with a known expiry and a demand picture that is quietly softening underneath the subsidy. That is a HOLD you accumulate when the market periodically panics about Washington — not a compounder you pay up for. The single piece of evidence that flips me bullish: durable, positive net bookings at stable US ASPs proving end-demand (not just the subsidy) is carrying the business. The single piece that flips me bearish: any credible legislative move to repeal or accelerate the phase-out of §45X, or a US ASP break below ~$0.28/W signaling the trade wall is leaking.

1. Executive Summary

First Solar designs and manufactures cadmium-telluride (CdTe) thin-film photovoltaic (PV) solar modules and sells them, primarily to US utility-scale developers, independent power producers, and large corporate energy buyers. It is the largest thin-film PV manufacturer in the world (>93 GW shipped cumulatively) and the only vertically integrated module manufacturer of meaningful scale headquartered outside China. Its technology is differentiated — a continuous “glass-to-module” process with no polysilicon and no Chinese supply-chain exposure, and superior real-world energy yield in hot climates — but it is not a structurally lower-cost producer than the Chinese crystalline-silicon (c-Si) industry on a global, unsubsidized basis.

The investment reality is straightforward and uncomfortable in equal measure. FSLR is highly profitable, but the bulk of that profit is a US government manufacturing subsidy. In FY2025, the company earned $5.2 billion of revenue, $2.1 billion of gross profit, and $1.5 billion of net income ($14.21 diluted EPS) — and recognized $1.6 billion of Section 45X advanced-manufacturing tax credits as a reduction of cost of sales. Without that credit, gross profit would have been roughly $0.5 billion (~10% gross margin) and the business would sit near operating breakeven. The §45X credit is also genuinely cash-generative: FSLR sells the credits to third parties for near-face cash (2024: $857M of credits sold for $819M).

The bull and bear cases derive from the same fact set and differ mainly on horizon:

• The case for owning it: The subsidy is law through 2032 (phasing from 2030), the FY2025 One Big Beautiful Bill Act (OBBBA) preserved §45X while tightening FEOC rules that exclude Chinese-linked competitors — an unambiguous relative tailwind for FSLR. The balance sheet is a fortress ($2.4B net cash, $0.6B debt), the heavy build-out capex cycle has peaked (capex fell from $1.5B in 2024 to $0.9B in 2025; FCF flipped from −$0.3B to +$1.2B), and a 50.1 GW contracted backlog with US ASPs locked near $0.35/W through 2028 gives multi-year revenue visibility. At ~17× trailing earnings with a net-cash balance sheet, the next several years are not expensively priced.
• The case against paying up: Essentially all of the profit is a subsidy with a legislated phase-down and live political risk; underlying module economics are thin and commoditized; net bookings were negative in 2025; 2026 revenue is guided flat and management declined to provide EPS guidance; and international (Malaysia/Vietnam) capacity is idling because FSLR’s product cannot compete outside the US policy wall.

This memo takes no position and sets no price target. It frames FSLR as a high-quality manufacturer whose reported economics are inseparable from a time-limited policy regime, and whose central analytical question is not “is the business good?” but “how much of this earnings stream survives the subsidy, and over what horizon?”

2. Business Overview

What the company does. First Solar manufactures and sells rigid, glass-on-glass CdTe thin-film PV solar modules. Unlike ~97% of the global industry, which uses crystalline-silicon (c-Si) wafers, First Solar deposits a thin layer of cadmium-telluride semiconductor onto glass in a vertically integrated, continuous manufacturing line — described by the company as a “glass-to-module” process that converts a sheet of glass to a finished module in a matter of hours, with no reliance on the polysilicon → ingot → wafer → cell value chain that defines the Chinese-dominated c-Si industry. The company shipped over 93 GW cumulatively as of year-end 2025 and sold a record 17.5 GW in FY2025 alone.

Who it sells to. First Solar is almost exclusively a module merchant to the utility-scale market — it sells modules to system developers, independent power producers (IPPs), utilities, and large corporate/data-center energy buyers, who then build solar power plants. It is not (since divesting its systems/EPC and O&M businesses years ago) a meaningful project developer. Its primary market is the United States; secondary markets include India, and historically Europe, Chile, and elsewhere. Customer contracts are typically multi-year, take-or-pay-style volume commitments with pricing set per watt, frequently with “technology adjusters” that allow ASP to rise as module efficiency/wattage improves over the delivery window (management cited ~$0.9B of additional backlog revenue from technology adjusters, mostly landing 2027–2028).

How it makes money — and the §45X overlay. Revenue is module sales (price-per-watt × watts shipped). But the economics are dominated by the Section 45X Advanced Manufacturing Production Credit under the Inflation Reduction Act of 2022 (as amended by OBBBA 2025). For a vertically integrated US producer, §45X pays roughly 7¢/watt for a finished module plus 4¢/watt for the PV cell (and $12/m² for a wafer) — credits that flow through First Solar’s income statement as a reduction of cost of sales, not as “other income.” This is why FSLR’s gross margin looks like a premium technology company’s (41–47%) while the underlying module business is a thin-margin commodity manufacturer. The credit is also transferable/saleable for cash, and is available 2023–2032 with a phase-down beginning 2030.

Revenue segmentation. First Solar reports essentially as a single segment — module manufacturing and sales. The meaningful disaggregation is geographic / by-factory and by-policy:

• US-manufactured modules (Ohio ×3, Alabama, Louisiana, and the scaling South Carolina facility): sold into the US utility-scale market at ASPs of ~$0.35/W, substantially committed through 2028, and fully §45X-eligible — the profit engine.
• International modules (Malaysia, Vietnam): historically exported to the US, now squeezed by tariffs and global oversupply; these plants are running at “significantly reduced” utilization.
• India (3.2 GW Series 7 facility): a domestic “book-and-bill” market clearing at ~$0.20/W — far below US pricing, illustrating what FSLR’s product fetches without the US policy umbrella.

Recurring vs. non-recurring. Revenue is contractual and backlog-driven (50.1 GW contracted) but is fundamentally transactional hardware sales — there is no installed-base/subscription recurring revenue. The closest thing to recurrence is the multi-year backlog and the technology-adjuster mechanism. One-time items recur with some regularity: contract-termination payments (a $384.6M claim against a former customer in 2025, $61M recognized), warranty/manufacturing-defect charges (recorded as reductions of revenue), and §45X credit-sale gains/losses.

Verdict: A focused, vertically integrated module manufacturer with genuine technological differentiation and exceptional revenue visibility — but a business model whose reported profitability is structurally fused to a US manufacturing subsidy and a protected domestic market. Understanding FSLR requires reading every margin line through the §45X lens.

3. Industry Dynamics

Structure: a brutally oversupplied, commoditized global industry, partitioned by trade policy. The global PV module industry is one of the least attractive manufacturing industries in the world on a standalone basis. It is dominated by Chinese c-Si manufacturers (LONGi, JinkoSolar, Trina, JA Solar, Canadian Solar) who have built vast, vertically integrated polysilicon-to-module capacity. First Solar’s own 10-K estimates that ~105 GW of new module capacity was added in 2025 alone, primarily in China — against global annual demand on the order of ~500–600 GW, but with total nameplate capacity running far ahead of it. The result is chronic structural overcapacity, collapsing global ASPs (many Chinese players have been selling below cash cost), and a multi-year shakeout. This is a textbook Marathon “capital-cycle” bust on the supply side: years of subsidized capital flooding in, returns driven to (and below) zero, with eventual capacity rationalization the only cure. For a generic module maker, this is a value-destroying industry.

The US carve-out is the entire story. What makes First Solar investable is that the United States has, through layered policy, effectively partitioned itself out of the global commodity market:

1. §45X manufacturing credit — pays US producers ~11¢/W (cell+module) for domestic manufacturing, a direct per-unit subsidy.
2. Trade remedies — antidumping/countervailing duties (AD/CVD) on c-Si cells/modules from China and, increasingly, from Southeast Asia (Malaysia, Vietnam, Thailand, Cambodia), plus Section 201/301 tariffs and Uyghur Forced Labor Prevention Act (UFLPA) enforcement. These raise the landed cost of imported c-Si into the US.
3. FEOC restrictions (OBBBA 2025) — newly tightened rules that disqualify products with “material assistance” from foreign entities of concern (i.e., Chinese-linked supply chains) from key credits, and condition downstream project credits on non-FEOC content.
4. Domestic-content ITC/PTC adders — bonus investment/production tax credits for projects that use US-made modules, creating pull demand for FSLR’s output from developers.

The combined effect: US module ASPs (~$0.35/W for FSLR) have remained stable even as global ASPs collapsed, because demand for domestically manufactured, FEOC-compliant modules outstrips the still-nascent US manufacturing base. First Solar is the largest, most-established beneficiary of this wall.

Demand backdrop is genuinely strong — for now. US utility-scale solar installations exceeded 30 GW in 2025 (total US installed solar ~266 GW), driven by data-center/AI electricity demand, broad electrification, and the economics of solar-plus-storage as the cheapest incremental generation in many regions. This is a real secular tailwind. The risk is on the policy side of demand, not the physics: OBBBA accelerated the termination of the §48E/45Y clean-electricity project credits (ITC/PTC) for wind and solar — meaning the downstream demand subsidy that pulls solar projects forward is being curtailed faster than previously legislated, even as the upstream §45X manufacturing subsidy was preserved. A faster ITC/PTC wind-down could soften project economics and, with a lag, module demand and pricing.

Regulatory landscape = the dominant variable. No industry analysis matters more here than the policy stack. The key tension: the manufacturing subsidy (§45X) that drives FSLR’s margins survives through 2032; the deployment subsidies (ITC/PTC) that drive its customers’ demand are being curtailed sooner. The durability of both — under a US administration whose posture toward renewables incentives is, at best, ambivalent — is the single largest source of fundamental uncertainty.

Verdict: structurally bad global industry, artificially good US sub-market. First Solar does not operate in a good industry; it operates inside a policy-engineered fortress within a terrible one. That fortress is real and currently widening (FEOC), but it is a policy construct, not an economic moat — and policy constructs can be amended by the same legislatures that built them.

4. Competitive Position

Name the moat honestly. First Solar’s durable competitive advantage, in Greenwald’s taxonomy, is not a classic supply-side cost advantage, not demand-side customer captivity (modules are largely fungible at the project level), and not a network effect. It is best described as a regulatory/policy moat layered on genuine but narrow technological differentiation. Let me pressure-test each candidate:

• Cost advantage (supply-side)? Partial and policy-contingent. First Solar’s vertically integrated CdTe process avoids polysilicon and the Chinese c-Si supply chain, and it has a genuinely low-cost, capital-efficient continuous line. But the hard evidence says it is not the global low-cost producer: its international plants (Malaysia, Vietnam) are idled because they cannot compete with Chinese c-Si on price outside the US wall, and its India modules clear at ~$0.20/W. If FSLR had a true structural cost advantage, those assets would run flat-out and win share globally. They don’t. The cost “advantage” is really §45X + tariff protection inside the US.
• Technological differentiation? Real, but a premium-niche, not a dominance, advantage. CdTe modules produce more real-world energy per nameplate watt in hot, humid climates (lower temperature coefficient, better spectral response, lower degradation), improving project-level LCOE. First Solar also owns valuable IP — and is now enforcing it: in March 2026 the US International Trade Commission instituted First Solar’s Section 337 action alleging TOPCon (c-Si) patent infringement against competitors, seeking exclusion orders. That IP position is a genuine, if speculative, asset (potential royalties/exclusions). But CdTe remains <5% of the global market; the world has standardized on c-Si, and First Solar’s efficiency roadmap (CuRe, bifacial, perovskite R&D via the Evolar acquisition) is a race to keep pace, not a runaway lead.
• Scale + the FEOC/domestic-content advantage? This is the real, current moat. First Solar is the only US module manufacturer with the scale (14.9 GW US nameplate in 2026, →17.1 GW 2027), track record (>93 GW shipped), bankability, and fully non-Chinese supply chain to satisfy FEOC and domestic-content requirements at volume. New entrants face years of ramp, qualification, and bankability hurdles. As FEOC rules tighten, the set of qualifying suppliers narrows toward First Solar — a widening relative advantage, but one entirely defined by the regulatory regime.

Direct comparison vs. competitors. Against Chinese c-Si giants (Jinko, LONGi, Trina): First Solar is far smaller globally and higher-cost per watt on an unsubsidized basis, but is structurally advantaged inside the US by policy. Against US-based or US-expanding c-Si assemblers (Qcells/Hanwha, and the wave of announced-but-nascent US c-Si lines): First Solar is more vertically integrated (most US c-Si “manufacturing” is cell/module assembly still dependent on imported wafers/polysilicon with FEOC exposure), more established, and more clearly FEOC-compliant end-to-end. Against other thin-film: First Solar is effectively the only at-scale CdTe producer — it is the thin-film industry.

The Greenwald test — does the moat show up in financials that would deteriorate without it? Yes, devastatingly: remove §45X and gross margin collapses from ~41% to ~10%. That is precisely the point — the “moat” is the subsidy. The technological and FEOC advantages are real and would leave First Solar as a viable, differentiated, marginally profitable niche manufacturer without §45X; the exceptional profitability is the policy. A moat that evaporates with a change in the tax code is a conditional moat, and must be underwritten as such.

Verdict: a genuine but conditional advantage. First Solar has the strongest competitive position in its addressable (US-policy-protected) market that it has ever had, and the FEOC tailwind is currently widening it. But the advantage is durable only as long as the policy regime is, and the underlying technology is differentiated-niche rather than dominant. This is not a Coca-Cola moat; it is a regulated-utility-style moat without the regulated-return guarantee.

5. Growth History and Forward Opportunities

Historical growth — lumpy, policy-inflected, now inflecting up. First Solar’s revenue history is anything but a smooth compounder: $2.9B (2021) → $2.6B (2022) → $3.3B (2023) → $4.2B (2024) → $5.2B (2025). The 2022 trough (and the −$44M net loss that year) reflected pre-IRA pricing pressure, logistics/freight inflation, and ramp costs; the post-2023 acceleration is the IRA era — §45X kicking in, US capacity ramping (Ohio expansion, Alabama, Louisiana), and US ASPs firming under the tariff wall. Net income tells the §45X story even more starkly: −$44M (2022) → $831M (2023) → $1,292M (2024) → $1,528M (2025). Volumes sold reached a record 17.5 GW in 2025.

The growth is real but its quality is policy-dependent. This is high-magnitude growth, but it is not the organic, pricing-power-driven growth of a structural share-gainer. It is capacity-ramp growth subsidized per unit: First Solar built US factories, the modules qualified for §45X, US ASPs held up because of trade protection, and revenue/profit scaled accordingly. Unit growth (GW shipped) is the cleaner signal than dollar growth, and it has been strong — but see the bookings caveat below.

Forward opportunities:

1. US capacity ramp — the clearest, most contracted growth lever. US nameplate goes 14.9 GW (2026) → 17.1 GW (2027) as Louisiana and South Carolina scale. With a 50.1 GW backlog and US output committed through 2028, near-term volume growth is largely sold.
2. Technology adjusters — ~$0.9B of backlog revenue tied to efficiency/wattage improvements, mostly 2027–2028, allowing ASP uplift on already-contracted volume. A genuine, contracted margin-expansion lever.
3. Mix shift to US (§45X-rich) volume — as international plants idle and US capacity grows, the §45X-eligible share of shipments rises, structurally lifting reported gross margin (the FY2026 guide to ~50–52% GM reflects this).
4. IP monetization — the TOPCon Section 337 litigation and broader patent portfolio could yield royalty streams or competitive exclusions (speculative, binary, but real optionality).
5. India — 3.2 GW serving a growing domestic market, though at structurally lower ASPs (~$0.20/W).
6. Next-gen technology — CuRe, bifacial, and perovskite (Evolar) R&D to extend the efficiency roadmap.

The warning signs in the growth profile. Three facts cut against extrapolating the recent growth:

• FY2026 revenue is guided flat at $4.9–5.2B vs. 2025’s $5.2B — an explicit growth pause even as capacity expands.
• Management declined to provide FY2026 EPS guidance, and analysts pointedly asked why; the likely answer is uncertainty around §45X recognition timing, tariffs, and tax — an uncomfortable signal for a company whose earnings are the credit.
• Net bookings were negative in 2025 — 7.4 GW of gross bookings against 8.3 GW of debookings (driven by First Solar terminating contracts for customer breaches, including BP affiliates, removing 6.6 GW / $1.9B of backlog). A backlog that shrinks on net, in a year of “record” shipments, says demand at First Solar’s required pricing is not as deep as the headline backlog implies.

Verdict: high-magnitude but lower-quality, policy-dependent growth that is decelerating. The contracted backlog provides excellent near-term visibility, but flat 2026 revenue, withheld EPS guidance, and negative net bookings indicate the demand-at-price picture is softening beneath the subsidy-driven earnings. This is not a self-sustaining growth flywheel; it is a subsidized capacity build approaching its near-term ceiling.

6. Financial Quality

Headline financials look like a high-quality compounder; the §45X overlay is the entire interpretive key.

(§45X figures: 2024 credits generated ~$857M; 2025 recognized $1.6B in COGS. Gross-margin/operating lines for 2022–2023 not separately reconciled here.)

The central quality-of-earnings finding. In FY2025, the $1.6 billion of §45X credits recognized as a reduction of cost of sales exceeded reported net income ($1.5 billion) and represented ~75% of gross profit ($2.1 billion). Backing the credit out, the underlying module business generated roughly $0.5 billion of gross profit on $5.2 billion of revenue — a ~10% gross margin — and would sit close to operating breakeven after ~$0.5 billion of operating expenses. An analyst on the Q4 call explicitly noted the FY2026 guidance implies “roughly a 10% component gross margin… even if I factor out underutilization and the §45X credit.” This is not a hidden fact — management discloses the credit clearly — but its magnitude reframes the entire P&L: First Solar’s GAAP profitability is, to a first approximation, the §45X subsidy.

Is the subsidy “real” earnings? Importantly, yes — in the near term. §45X is not an accounting accrual that may never convert to cash: First Solar sells the credits to third parties for cash (2024: $857M of credits sold for $819M cash, a ~4.5% haircut). The credit is legislated, refundable/transferable, and available through 2032. So this is cash-generative, bankable, recurring-while-the-law-stands income. The quality problem is not realizability; it is durability and source — it is a government transfer with a phase-down (beginning 2030) and live legislative risk, not a market-earned return.

Cash flow and the capex inflection — the genuinely positive structural story. First Solar spent 2021–2024 in a heavy build-out: cumulative negative free cash flow as it funded the US factory expansion (capex peaked at $1.5B in 2024). That cycle has crested — 2025 capex fell to $0.9B, operating cash flow jumped to $2.1B, and free cash flow flipped sharply positive to +$1.2B. With the US footprint now largely built (Louisiana online, South Carolina the last major scale-up), capex intensity declines from here, and the §45X cash receipts plus working-capital normalization should drive several years of strong free cash flow. This is the most underappreciated positive in the financial profile.

Balance sheet — a fortress. Year-end 2025: $2.9B gross cash / $2.4B net cash, against only ~$0.6B of debt (primarily an India DFC project loan). Stockholders’ equity of $9.5B, total assets $13.3B, book value ~$91.8/share. There is essentially no financial-leverage risk, no refinancing wall, and ample liquidity to self-fund the remaining expansion, weather a downturn, and return capital. Management explicitly targets a $1.5–2.0B working-capital reserve “to account for industry cyclicality.” For a company in a brutally cyclical industry, this conservatism is appropriate and a clear positive.

Returns on capital. ROE ran ~18% (TTM) and ROIC is healthy as reported — but the same caveat applies: those returns are subsidy-inclusive. On a pre-§45X basis, returns on the heavily capital-intensive manufacturing asset base would be mediocre. The “do economics improve with scale?” test cuts both ways: §45X is a per-watt credit, so it scales linearly with volume (more US GW → more credit), which is why margins expand as US mix rises; but the underlying (pre-credit) unit economics show only modest operating leverage in a commoditized product.

Accounting red/amber flags. (1) Warranty and manufacturing-defect charges recorded as reductions of revenue (a 2025 issue with certain Series-7 modules), which can obscure underlying ASP trends. (2) Contract-termination payments recognized as revenue ($61M in 2025) — lumpy, non-operating in character. (3) §45X credit-sale timing can shift income between periods. (4) The gap between GAAP net income and the cash value of credits sold requires care in any normalized-earnings exercise. None of these are aggressive in the GE/Visa-gain sense, but they require normalization before drawing run-rate conclusions.

Verdict: high reported financial quality, conditional underlying quality. The balance sheet is genuinely excellent and the FCF inflection is real and durable for several years. But the income statement’s quality is entirely contingent on §45X: this is a company that converts a government subsidy into cash very efficiently, sitting atop a thin-margin commodity manufacturer. Economics improve with scale because the subsidy scales with volume, not primarily because the business gets structurally better.

7. Capital Allocation

The record is conservative, disciplined, and — by the low standard of the solar industry — excellent. First Solar has navigated multiple brutal solar downturns (2011–2013, 2016–2019, 2022) without ever impairing its balance sheet, diluting shareholders destructively, or making value-destroying acquisitions. That survival-through-cyclicality track record is itself the strongest evidence of capital-allocation competence in this sector.

Use of cash — the stated framework. Management articulates a clear, sensible hierarchy: (1) maintain a $1.5–2.0B working-capital reserve as a cyclicality buffer; (2) fund organic growth and “replicate technology improvements across the fleet” (i.e., reinvest in US capacity and the efficiency roadmap); (3) consider returns of capital thereafter. In practice, the 2021–2025 period was dominated by step (2) — the multi-billion-dollar US factory build-out funded entirely from internal cash and the balance sheet, with no equity raises and minimal debt. That self-funded, non-dilutive growth is a strong positive.

M&A — minimal and bolt-on. First Solar is not a serial acquirer. The notable recent deal was the 2023 acquisition of Evolar (perovskite thin-film technology, a few hundred million dollars) — a small, R&D-oriented, technology-option purchase, not a scale or diversification play. The company has historically divested non-core businesses (its systems/EPC and O&M/operating businesses, and earlier its stakes in development assets), sharpening focus on module manufacturing. This is disciplined, focus-enhancing capital allocation — the opposite of empire-building.

Buybacks and dividends — essentially none. First Solar pays no dividend and has not run a material, sustained buyback program. Capital has been retained and reinvested in capacity. Given (a) the heavy growth capex cycle now ending, (b) the strong FCF inflection, and © a $2.4B net-cash position, the forward capital-allocation question becomes pointed: what will management do with the rising free cash flow? A buyback initiated near a policy-driven valuation trough would be highly accretive; one initiated near a cyclical/subsidy peak would not. There is, as yet, no committed return-of-capital program — an open question for the thesis.

Dilution / share count. Share count is stable (~107M diluted), with no meaningful equity issuance and modest SBC. This is a clean, non-dilutive equity story — a notable contrast to most of the solar/clean-energy complex, where serial dilution is the norm.

Insider behavior and incentives. Insider ownership is low (~5.4%) — First Solar is a professionally managed company, not a founder/owner-operator, so alignment rests on the comp structure rather than large personal stakes. Executive compensation is restrained: CEO Mark Widmar’s FY2025 total compensation was ~$8.1M (≈$1.0M salary, $5.5M equity, $1.6M cash incentive), CFO Alexander Bradley ~$2.9M — modest for a $28B-market-cap company, and the proxy states comp is targeted slightly below peer-group median. Incentive metrics emphasize production, operating margin, and ROIC plus relative TSR — operationally sensible metrics, though “operating margin” and “ROIC” are, again, §45X-inflated, so the comp plan implicitly rewards harvesting the subsidy. There is no evidence of egregious pay, option repricing, or insider self-dealing; Form 4 activity is routine (grants/vesting and planned sales), with no notable open-market insider buying signal.

Verdict: management has allocated capital intelligently and conservatively — self-funded growth, no destructive M&A or dilution, fortress balance sheet, restrained pay, and survival through multiple downturns. The one forward question mark is the deployment of the coming FCF wave: a disciplined, opportunistic buyback would be value-additive, but no such program is yet committed. On the historical record, this is a clear positive for the thesis.

8. Changes and Headwinds — Last Two Years

Policy — the dominant change. The single most important development is the One Big Beautiful Bill Act (OBBBA) of 2025, which amended the IRA. The net effect for First Solar is mixed but on balance favorable: it preserved the §45X manufacturing credit (the margin engine) and tightened FEOC restrictions (narrowing the qualifying-supplier set toward First Solar) — both positives — while accelerating the termination of the §48E/45Y clean-electricity project credits (ITC/PTC) for wind and solar — a negative for downstream demand. The market’s recurring anxiety is that the next legislative or executive action could reach the §45X manufacturing credit itself; so far it has not, but the 2024 election outcome and the administration’s renewables posture keep this risk live and have driven much of the stock’s volatility (52-week range $135.50–$320.95).

Demand/bookings deterioration. As noted, 2025 net bookings were negative (7.4 GW gross vs. 8.3 GW debookings). Some debookings were First Solar-initiated terminations for customer breach (BP affiliates; 6.6 GW / $1.9B removed, with a $384.6M payment claim now in litigation in the NY Supreme Court), which is arguably a quality-of-backlog improvement (culling non-performing counterparties). But the low gross-bookings number in a strong demand year is a yellow flag on demand-at-price.

Tariff cost pressure. FY2025 gross margin declined (44% → 41%) primarily due to tariff costs on First Solar’s own international (Malaysia/Vietnam) modules imported into the US, plus associated warehousing/logistics expense — an irony: the trade walls that protect FSLR’s US pricing also tax its own international output. In response, First Solar has reduced international production and is building a US Series-6 finishing facility to optimize freight/tariff/domestic-content outcomes and capture §45X module-assembly credits.

Capacity ramp and the capex peak. First Solar completed its major US expansion — Louisiana came online in 2025 (its fifth US facility), South Carolina scaling into 2027 — and the capex cycle peaked and turned down. This is a structural positive (FCF inflection) and a change in the company’s financial character from “cash-consuming builder” to “cash-generating harvester.”

Technology and IP offense. First Solar moved from defending to enforcing its IP, with the March 2026 ITC Section 337 action over TOPCon patents — a notable strategic shift that could yield royalty/exclusion upside. It continues the CuRe/bifacial efficiency roadmap and perovskite R&D.

Manufacturing-quality issues. First Solar disclosed manufacturing issues on certain Series-7 modules, recording warranty-related charges (as reductions of revenue) and contracted-volume adjustments — a reminder that even a differentiated manufacturer carries execution/quality risk that directly hits revenue.

Verdict: the changes are a mixed bag that, on net, modestly weaken the forward thesis at the margin even as they strengthen the near-term competitive position. FEOC tightening and the capex/FCF inflection are real positives; negative net bookings, flat 2026 revenue, withheld EPS guidance, tariff-driven margin pressure on international output, and the OBBBA curtailment of downstream demand subsidies are real cautions. The center of gravity has shifted from “growth story” to “harvest-the-subsidy, defend-the-wall” story.

9. Risk Analysis (Risk Matrix)

Overall risk posture: The balance sheet eliminates financial/solvency risk. The dominant risks are policy (single-variable exposure to §45X and the trade wall) and demand-at-price. This is a low-balance-sheet-risk, high-policy-risk equity — an unusual and important profile to underwrite correctly.

10. Valuation Discussion (Embedded Expectations)

No price target and no recommendation in this section. The purpose is to characterize what the current price embeds.

Where the stock trades. At ~$267/share, First Solar carries a market capitalization of ~$28.7B and, net of ~$2.4B net cash, an enterprise value of ~$27B. On trailing FY2025/TTM figures: P/E ~17–18× (TTM EPS $15.47), forward P/E ~16×, EV/EBITDA ~12×, EV/Revenue ~5×, P/B ~2.9×. Against its own 10-year history, the stock screens cheap on earnings (≈30th percentile P/E) but rich on book and sales (≈84th / 77th percentile) — a composite around the 64th percentile. Tellingly, the Wall Street consensus price target (~$244) sits below the current price, even as the analyst rating skews bullish — a sell-side that likes the business but finds the price full, consistent with subsidy-durability concerns.

The embedded-expectations question: what must be true to justify ~$28.7B of equity value?

The cleanest way to frame this is to separate the subsidy annuity from the underlying business:

• The §45X annuity. First Solar is generating on the order of $1.6B+/year of §45X credits, scaling with US volume toward ~$2B+ as US capacity ramps to 17 GW. If one capitalizes the after-haircut cash value of the credit stream from 2026 through its 2032 expiry (with the 2030–2032 phase-down), discounted, the §45X cash flows alone are worth a very large fraction of the current enterprise value — plausibly the majority of it. In other words, at ~$27B EV, the market is paying roughly the present value of the legislated §45X credit stream plus a modest amount for the residual, post-2032 business.
• The residual business. Strip the subsidy and First Solar is a ~$5B-revenue, ~10%-gross-margin, capital-intensive commodity manufacturer with a differentiated niche technology, a fortress balance sheet, valuable IP optionality, and a leading position in a structurally protected US market. On unsubsidized economics, such a business would not command a premium multiple. The market is not underwriting much for the post-2032 world — which is either appropriate caution or an opportunity, depending on one’s view of policy renewal.

Scenario framing (illustrative, not a forecast):

• Bear (§45X curtailed early / demand softens / ASP leaks): Earnings power resets toward the unsubsidized ~$2–4/share range; the stock would be valued as a cyclical commodity manufacturer at high-single-digit EV/EBITDA on depressed numbers. Substantial downside from current levels — the policy tail.
• Base (§45X runs to schedule, FEOC widens the moat, US ASPs hold ~$0.35/W through 2028, backlog converts): Several years of $14–20+ EPS, strong FCF, a growing net-cash pile, and potential return of capital. At ~16× a flattish-to-growing subsidized earnings stream with a net-cash balance sheet, the stock is reasonably-to-attractively priced for the visible window, with the post-2030 phase-down as the overhang.
• Bull (§45X extended/renewed, ITC/PTC stabilizes demand, technology adjusters and IP royalties add upside, buyback initiated): The “annuity” extends beyond 2032, the residual-business value re-rates, and FCF deployment compounds book value. Meaningful upside, but it requires political outcomes outside the company’s control.

What the market is pricing correctly vs. incorrectly. The market appears to be correctly pricing the near-term subsidized cash flows as real (hence the not-demanding ~16× forward multiple) while appropriately discounting the post-2032/policy-risk tail (hence the sub-current consensus target and the rich-on-book, cheap-on-earnings split). The variant-perception opportunity, if any, is in the durability of the policy regime and the FEOC-driven widening of the competitive moat — areas where the consensus is cautious and where a more constructive policy view would justify a higher multiple. The variant-perception risk is symmetric: the same policy could be curtailed.

Verdict: At ~17× trailing earnings with a net-cash balance sheet and a 50.1 GW backlog, First Solar is not expensively priced against its visible, subsidized earnings — and is arguably cheap if §45X simply runs to its legislated schedule. The valuation debate is entirely about what survives the subsidy and the phase-down, and the market is currently splitting the difference: paying roughly the PV of the legislated credit stream and very little for the post-policy business.

11. Variant Perception

Consensus belief. The sell-side and most investors view First Solar as the highest-quality, most-defensible US solar manufacturer — a clear IRA/FEOC beneficiary with a fortress balance sheet and excellent backlog visibility — but one whose earnings are policy-dependent and whose growth is decelerating, justifying a full-but-not-cheap mid-teens multiple and a price target around or slightly below the current quote. In short: “great company, policy-dependent earnings, fairly valued.”

Strongest bull case. First Solar is the single best-positioned beneficiary of a widening US policy moat. OBBBA preserved §45X (the margin engine) and tightened FEOC rules that progressively exclude Chinese-linked competitors, narrowing the qualifying-supplier universe toward First Solar precisely as US solar demand (data centers, electrification) surges. The capex cycle has peaked, FCF has inflected sharply positive, the balance sheet holds $2.4B net cash, US ASPs are locked ~$0.35/W through 2028, technology adjusters add contracted margin upside, and the TOPCon IP litigation offers free optionality. The market is over-discounting policy risk on a credit that runs to 2032 and is more entrenched after OBBBA, not less. At ~16× forward with net cash and rising FCF — and an eventual buyback — this is a mispriced, de-risked compounder hiding behind policy fear.

Strongest bear case. First Solar is a thin-margin (~10% pre-subsidy gross margin), commoditized, cyclical manufacturer whose entire reported profitability is a US government transfer with a legislated phase-down (2030–2032) and live repeal risk under an administration hostile to renewables incentives. The “moat” is the subsidy and a tariff wall, both revocable by the legislature that built them. The tell is in the company’s own numbers: net bookings turned negative in 2025, FY2026 revenue is guided flat, management withheld EPS guidance, international plants are idling because the product can’t compete unsubsidized, and US gross margin fell on the company’s own tariff costs. You are paying ~$27B EV for the present value of a credit stream that legally begins shrinking in under four years, atop a business that earns mediocre returns without it. Any negative policy headline (and there will be many) re-rates the stock sharply lower — as the $135–$321 52-week range already demonstrates.

The 3–5 assumptions that matter most:

1. §45X durability — does the manufacturing credit survive intact through 2032 (and possibly get renewed), or is it curtailed early? The single dominant variable.
2. US ASP/tariff-wall durability — do US module prices hold near $0.35/W, or does global oversupply leak through eroding trade remedies?
3. Demand-at-price — do net bookings turn positive again at stable pricing (proving real end-demand), or does the flat-revenue/negative-bookings trend continue?
4. Pre-subsidy economics trajectory — does the underlying ~10% gross margin structurally improve with scale/technology, building a post-2032 floor?
5. Capital deployment — does the coming FCF wave get returned accretively (buyback at a trough) or sit idle / chase low-return growth?

What would falsify each side:

• Falsifies the bull: credible legislative action to repeal/accelerate-phase-out §45X; a US ASP break below ~$0.28/W; continued negative net bookings into 2027.
• Falsifies the bear: durable positive net bookings at stable/ rising US ASPs; structural pre-subsidy gross-margin expansion above ~15%; a §45X extension/renewal; a large accretive buyback at a depressed valuation.

Verdict: This is a stock where the bull and bear are arguing about horizon and policy, not about the current facts (which both sides agree on). The variant-perception edge belongs to whoever has the better read on US energy/tax policy durability over 2026–2032 — an unusual, and uncomfortable, basis for an equity thesis.

12. Fact vs. Interpretation

13. Open Questions

1. Will §45X survive to 2032 intact? The dominant unknown. What is the realistic probability distribution of repeal / early phase-out / renewal under the current and next US administrations?
2. What is the true post-2032 (unsubsidized) earnings power? Can pre-subsidy gross margin structurally exceed ~10–15%, and what residual multiple does the business deserve in a no-§45X world?
3. Why was FY2026 EPS guidance withheld? Is it §45X-recognition timing, tariff uncertainty, tax complexity, or a softer demand signal management is reluctant to quantify?
4. Do net bookings recover? Is the negative 2025 figure a one-off (counterparty culling) or the start of a demand-at-price deterioration?
5. What will management do with the FCF wave? Buyback (and at what valuation), special dividend, more capacity, or M&A?
6. How exposed is FSLR to §45X indemnification claims if the IRS challenges credits sold to third parties?
7. What is the realistic value of the IP/TOPCon litigation — royalty stream, exclusion order, settlement, or nothing?
8. Does the technology roadmap (CuRe/bifacial/perovskite) keep CdTe competitive with advancing c-Si and emerging tandem cells beyond this decade?

14. What Must Be True (Bull and Bear, with Falsification Tests)

For the BULL case to be right, these must hold:

1. §45X runs to its legislated schedule (or is renewed) — the credit is not repealed or sharply accelerated before 2030. Falsification test: any enacted legislation, or credible high-probability legislative action, that repeals or materially accelerates the §45X phase-out.
2. The US trade wall holds and US ASPs stay ~$0.35/W — domestic-content/FEOC demand keeps US pricing insulated from global oversupply through at least 2028. Falsification test: a sustained US blended ASP decline below ~$0.28/W, or a material rollback of AD/CVD tariffs.
3. Demand recovers at price — net bookings turn durably positive and the backlog grows again. Falsification test: net bookings remain negative through 2026–2027 with the 50.1 GW backlog eroding.
4. FCF converts and is deployed accretively — the capex-down/FCF-up inflection persists and capital is returned at attractive valuations. Falsification test: FCF disappoints (working-capital drag, capex re-acceleration) or cash is deployed into low-return growth/M&A.

For the BEAR case to be right, these must hold:

1. The subsidy is the business — unsubsidized economics stay thin (~10% gross margin) with no structural improvement. Falsification test: pre-§45X gross margin expands durably above ~15% on scale/technology.
2. Policy risk is mispriced as benign — the market under-discounts §45X/tariff revocability. Falsification test: a §45X extension/renewal, or a durable bipartisan policy consensus, that removes the repeal tail.
3. Demand is softening beneath the subsidy — flat revenue, withheld EPS guidance, and negative bookings are the leading edge of a deterioration. Falsification test: re-acceleration of bookings, revenue, and ASP into 2026–2027.
4. Global oversupply eventually leaks into the US — ~105 GW/yr of (mostly Chinese) additions overwhelm the trade remedies over time. Falsification test: US ASPs hold firm through the oversupply peak, proving the wall is durable.

15. Source Appendix

See FSLR_source_appendix.md (Appendix B in the combined report) for the full, dated source list. Primary sources relied upon: First Solar FY2025 Form 10-K (filed 2026-02-24); Q1-2026 Form 10-Q (filed 2026-04-30); Q4-2025 (2026-02-24) and Q1-2026 (2026-04-30) earnings-call transcripts; DEF 14A proxy (filed 2026-04-02); SEC EDGAR XBRL financial data (CIK 0001274494); and public market data (third-party, reconciled to filings).

This article is independent research for general information only and is not investment advice. The analysis carries no investment recommendation and no price target; the sole exception is the clearly-labeled opinion block at the top, which is the author’s own view.

APPENDIX A — Standard Diligence Questionnaire

First Solar, Inc. (NASDAQ: FSLR) — as of 2026-06-13

Supplemental to the analysis above. Fact / Interpretation / Assumption labels applied where it matters.

General

What thoughtful questions have other investors asked about this company? The recurring institutional questions, evidenced in the earnings-call Q&A: (1) How much of earnings is the §45X subsidy vs. the underlying business? — and an analyst on the Q4-2025 call directly modeled “roughly a 10% component gross margin even factoring out underutilization and §45X” (Fact). (2) Why no FY2026 EPS guidance? — analysts pressed management on this (Fact); the implied concern is earnings-quality/visibility. (3) Is US ASP (~$0.35/W) durable, or will global oversupply leak through the tariff wall? (4) What happens to §45X under the current administration / after the 2030 phase-down? (5) What will the company do with the rising free cash flow? These map directly to the variant-perception axes in §11.

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Reported earnings are near a subsidy-driven high — FY2025 net income $1,528M is a record, but ~$1.6B of it is the §45X credit (Fact/Interpretation). On an unsubsidized basis, earnings would be near a cyclical low (thin margins, idled international capacity, oversupplied global market). So the answer is paradoxical: reported earnings high, underlying earnings low.

Driven by the external environment or internal actions? Predominantly external — US tax policy (§45X), trade policy (AD/CVD tariffs, FEOC), and the domestic-content demand pull. Internal actions (US capacity build-out, technology roadmap, balance-sheet discipline) position the company to capture the policy benefit, but the benefit itself is exogenous (Interpretation).

How stable are revenues? Near-term: highly visible (50.1 GW contracted backlog, US output committed through 2028). Medium-term: less stable than the backlog implies — 2025 net bookings were negative and FY2026 revenue is guided flat (Fact). Revenue is transactional hardware sales with no recurring/subscription component.

Outlook for products/services? US utility-scale module demand is strong (data centers, electrification; >30 GW US utility-scale installed in 2025). The product (CdTe modules) is competitive within the US policy-protected market and in hot climates; uncompetitive on unsubsidized global price (Fact — evidenced by idled Malaysia/Vietnam plants).

How big will this market be — growing, shrinking, domestic or international? The US solar market is growing (266 GW installed, 30+ GW/yr utility-scale additions). First Solar is now primarily domestic by design (international plants idling, US the profit center). The global market is enormous but commoditized and oversupplied; First Solar has effectively retreated to the protected US market.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? Globally more competitive and oversupplied (~105 GW added in 2025, mostly China). Within the US, less competitive for First Solar specifically, as FEOC rules tighten and narrow the qualifying-supplier set in its favor (Fact/Interpretation).

How profitable is the business (ROIC, ROE)? Reported ROE ~18% (TTM); ROIC healthy — but subsidy-inclusive (Fact/Interpretation). Pre-§45X returns on the capital-intensive asset base would be mediocre.

How profitable is the industry — competitors, barriers to entry? The global industry is unprofitable (many Chinese players below cash cost). Barriers to entry into the US FEOC-compliant niche are high (scale, bankability, non-Chinese supply chain, years of ramp), which is why First Solar earns supernormal reported returns there.

Can the business be easily understood? The mechanics yes; the durability no — the central variable (US tax/trade policy through 2032 and beyond) is outside the company’s control and hard to forecast.

Can it be undermined by foreign low-cost labor? It already is, globally — Chinese c-Si is cheaper per watt. First Solar is insulated only by US trade policy. This is the crux of the bear case.

Do brands matter? Modestly — “bankability”/track record and warranty creditworthiness matter to project financiers, and First Solar’s >93 GW shipped and balance-sheet strength are genuine assets. But modules compete on price-per-watt and energy yield, not consumer brand.

Nature of competition / switching costs? Competition is price-per-watt (adjusted for energy yield, degradation, domestic-content/FEOC eligibility). Switching costs at the project level are low (modules are substitutable), though qualification, warranty, and contracted-volume commitments create some stickiness within a delivery window.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? The §45X credit-generation capability and the IP portfolio (TOPCon patents now in active ITC litigation; perovskite/Evolar technology) are economically valuable but not capitalized as such (Interpretation). The contracted backlog (50.1 GW) is an off-balance-sheet revenue asset.

Off-balance-sheet liabilities? Warranty obligations (some recorded as revenue reductions), §45X credit-sale indemnification obligations to purchasers (could become material if the IRS disallows credits), and contractual purchase commitments. No unusual hidden leverage (Fact).

How conservative is the accounting? Generally conservative, with a few lumpy items requiring normalization: §45X recognized in COGS (clearly disclosed); warranty/quality charges booked as revenue reductions (can obscure ASP); contract-termination payments recognized as revenue ($61M in 2025). Management maintains a $1.5–2.0B working-capital reserve explicitly for cyclicality — a conservative posture (Fact).

How CapEx-hungry is the business? Very, historically (peaked at $1.5B in 2024 on the US build-out), but the cycle has crested — 2025 capex fell to $0.9B and declines from here as the footprint completes (Fact). This is the key FCF-inflection point.

Capital Allocation & Management

How much FCF does the business generate; how is it used; philosophy? FCF flipped sharply positive in 2025 (+$1.2B) after years of build-out-driven outflows. Stated philosophy: (1) maintain $1.5–2.0B working-capital reserve; (2) fund growth/technology replication; (3) returns thereafter (Fact). No committed buyback/dividend yet — deployment of the coming FCF wave is an open question.

Significant acquisitions recently? Minimal — Evolar (perovskite R&D, 2023, small). First Solar is a divestor, not an acquirer, having shed systems/EPC and O&M businesses to focus on modules (Fact). Disciplined.

Buying back shares? No material sustained buyback historically; no dividend. Share count stable ~107M, minimal dilution (Fact) — a clean equity story vs. the dilution-heavy clean-energy norm.

Issuing large amounts of new shares to insiders? No — modest SBC, restrained equity grants (Fact).

Compensation policy of directors/management? Restrained: CEO Widmar FY2025 total comp ~$8.1M (≈$1.0M salary / $5.5M equity / $1.6M cash incentive), CFO Bradley ~$2.9M; proxy targets comp slightly below peer median. Metrics: production, operating margin, ROIC, relative TSR (Fact). Sensible operationally, though margin/ROIC are §45X-inflated.

Motivations of management? Professionally managed (insiders ~5.4%, no founder-operator); incentives are operational-metric-driven. No evidence of self-dealing or empire-building; the track record is survival-through-cyclicality and self-funded growth (Interpretation).

Valuation & Market Data

Is the stock an ADR, MLP, or K-1 issuer? No — common stock of a US C-corporation (NASDAQ: FSLR). Standard 1099 treatment.

Dividend policy? No dividend (Fact).

How profitable is the business? Highly profitable as reported (~30% net margin TTM, ~18% ROE); the profitability is subsidy-driven (see above).

Is net income diverging from cash from operations? In 2025, OCF ($2.1B) exceeded net income ($1.5B) — a healthy sign (driven by §45X cash receipts and working-capital/collections), the opposite of a low-quality earnings divergence. The earnings-quality caveat here is source (subsidy) and durability (phase-down), not cash conversion (Fact/Interpretation).

Risks & Downside

What factors would cause the stock to decline? (1) §45X repeal/early phase-out; (2) faster ITC/PTC wind-down softening demand; (3) US ASP erosion as global oversupply leaks through tariffs; (4) continued negative net bookings / weak demand-at-price; (5) tariff-relief on imports; (6) manufacturing-quality/warranty events; (7) disappointing FCF or value-destructive capital deployment. The 52-week range ($135.50–$320.95) shows how violently the stock re-rates on policy headlines (Fact).

Risk of a catastrophic loss? Low in the near term — net-cash balance sheet, no leverage, tangible assets, and a multi-year backlog. A catastrophic (vs. merely large) loss would require simultaneous §45X repeal and tariff-wall collapse, and even then residual asset/cash value cushions the downside (Interpretation).

Chance of a total loss? Very low. No solvency risk; ~$2.4B net cash and real assets (Fact).

Recent News & Events

Has the business environment changed recently? Yes, materially: the OBBBA (2025) amended the IRA — preserving §45X and tightening FEOC (positives) while accelerating ITC/PTC project-credit termination (negative). FEOC tightening is widening First Solar’s competitive moat in real time (Fact).

Significant acquisitions? None material (Evolar 2023 the most recent of note).

Change in accounting policies? No material change; §45X recognition (as COGS reduction) is the dominant accounting feature, consistently applied.

Recent changes — new markets, facilities, management? Louisiana facility (fifth US plant) came online in 2025; South Carolina scaling into 2027; a US Series-6 finishing facility planned to optimize tariff/freight/domestic-content and capture §45X assembly credits. India at 3.2 GW. Management team stable. First Solar has shifted from defending to enforcing its IP (March 2026 ITC Section 337 TOPCon action) (Fact).

Note: where a question maps imperfectly to a module-manufacturing business, the closest applicable analog has been used. The §45X subsidy is the recurring interpretive thread across nearly every diligence dimension.

APPENDIX B — Source Appendix

First Solar, Inc. (NASDAQ: FSLR) — Research Sources, as of 2026-06-13

Primary sources prioritized. All financial figures reconciled to SEC filings / EDGAR XBRL where possible. Third-party aggregated market data is labeled and treated as a cross-check, not as primary authority.

Primary — SEC Filings (EDGAR, CIK 0001274494)

Primary — Earnings-Call & Event Transcripts

Secondary / Third-Party (labeled; cross-check only)

Key quantitative reference points (all reconciled to EDGAR XBRL unless noted)

• Revenue ($M): 2021 2,923 / 2022 2,619 / 2023 3,319 / 2024 4,206 / 2025 5,219; Q1-26 1,044
• Net income ($M): 2021 469 / 2022 −44 / 2023 831 / 2024 1,292 / 2025 1,528; Q1-26 347
• Diluted EPS: 2024 $12.02 / 2025 $14.21 / Q1-26 $3.22; TTM $15.47
• §45X credit (COGS reduction): FY2025 $1.6B; 2024 credits generated $857.2M sold for $818.6M cash
• Gross margin: 2024 44.2% / 2025 40.6% / Q1-26 46.6%; FY2026 guide ~50–52%
• OCF ($M): 2023 602 / 2024 1,218 / 2025 2,057; Capex ($M): 2023 1,387 / 2024 1,526 / 2025 870; FCF 2024 −308 / 2025 +1,187
• Cash 2025 $2,804M (gross $2.9B / net $2.4B); debt ~$587M; equity $9,538M; assets $13,321M; BVPS ~$91.8; shares ~107M
• Backlog 50.1 GW (Dec-2025); 2025 shipments 17.5 GW; 2025 net bookings negative (7.4 GW gross / 8.3 GW debookings)
• Capacity: US nameplate 14.9 GW (2026) → 17.1 GW (2027); global 19 GW (2026) → 22.1 GW (2027); India 3.2 GW; >93 GW shipped cumulatively
• Valuation (price ~$267): P/E ~17–18×, fwd ~16×, EV/EBITDA ~12×, EV/Rev ~5×, P/B ~2.9×; own-history percentiles P/E ~30th, P/B ~84th, P/S ~77th; consensus target ~$244; short interest ~10.5% of float

All URLs accessible via SEC EDGAR full-text search (https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001274494). Access date: 2026-06-13.