FIRST SOLAR, INC. (FSLR) stock intrinsic value fair value estimate & owner earnings analysis

First Solar is the largest US-headquartered maker of solar panels. It uses its own thin-film cadmium telluride technology rather than the crystalline silicon used by Chinese producers, and sells mainly to developers of large utility-scale solar farms. In 2025 it produced 16.1 GW and sold 17.5 GW of modules for $5.2 billion of revenue (about 30 cents per watt), with 96% of sales in the United States. It runs five US factories (Ohio, Alabama, and a Louisiana plant that started in 2025), plus factories in Malaysia, Vietnam and India. A sixth US site in Gaffney, South Carolina, due in the second half of 2026, will finish modules started overseas. What drives the value. Most of First Solar's current profit comes from one federal tax credit. The Section 45X advanced manufacturing credit pays about 17 cents for every watt of modules made in the US and sold. In 2025 the company recorded $1.6 billion of these credits, against pre-tax income of $1.58 billion. Put simply, without the credit First Solar roughly broke even in 2025. It earned about $230 million pre-tax without the credit in 2023 and about $410 million in 2024. The credit is written into law through 2032. It steps down to 75% in 2030, 50% in 2031 and 25% in 2032, and then ends. The July 2025 budget law kept this schedule but added foreign-ownership restrictions, which a US-owned producer with a non-Chinese supply chain is placed to meet. The company sells the credits for cash at roughly 95 cents on the dollar. That means the profit shows up as real cash, and we take off a 5% discount for it. Because the credit is large but has a fixed end date, we did not value the company on a steady growing earnings stream. We valued the credit stream year by year to its legal end, and valued the unsubsidized business separately. How we measured earnings. We start from operating cash flow and take out swings in working capital (such as customer prepayments), stock compensation, and the capital spending needed to keep the existing business running. On that basis, cash earnings were $13.98 per share in 2025 and $3.37 in 2024, and negative in 2023. In 2023 and 2024 the company spent about $1.4 to $1.5 billion a year building new factories before those factories produced sales, so our standard formula counts much of that building spend as upkeep. Those years therefore understate what the business earns. If upkeep spending is set equal to depreciation instead, the figures are about $7.30 (2023), $11.60 (2024) and $14.90 (2025) per share. The years before 2022 describe a different company, one that also developed and sold solar power plants, and they came before the credit existed. We therefore treat 2023 to 2025 as the relevant record. Recent results continue the trend: first-half 2026 net income was $769 million, up 39% from $551 million a year earlier, as more US-made (credit-earning) modules were sold. We set the next year's growth at or below the roughly 28% to 39% growth actually seen, backed by the Louisiana plant ramping up. Base case, about $90 per share. US-made modules sold rise from roughly 9.4 GW in 2025 (implied by the $1.6 billion credit) to 12.5 GW in 2027 and 13.5 GW from 2028, as the Louisiana plant runs at full rate. The credit brings in about $19 to $20 per share a year through 2029, then falls with the legal step-down to zero after 2032. On top of that, the unsubsidized business earns about $1 per share a year. That is close to its 2023-2025 average without the credit, and we assume it lasts through 2042, roughly the useful life of the new factories. Support for these volumes: a contracted backlog of 50.1 GW worth $15.0 billion (about 30 cents per watt) running through 2030, and long-standing US trade barriers against imported crystalline silicon panels. Everything is discounted at 6%. Bear case, about $62 per share. US volumes stall at 10 to 11 GW. Customers who can no longer claim federal solar project tax credits cut back: under the 2025 law, utility-scale projects generally had to begin construction by early July 2026 to qualify. Contracts also break down, as happened with BP and Lightsource, whose defaults cut 6.6 GW and $1.9 billion from the backlog in 2025. The unsubsidized business loses about 50 cents per share a year through 2032. Causes here are ones already disclosed: tariffs on raw materials and on imported modules, detention and warehousing costs, a $50 million Series 7 warranty reserve, possible underuse of the Malaysian and Vietnamese plants, and India's domestic-cell rule from 2027. Nothing is counted after 2032. Even in this case, the credits already written into law give the shares most of their value. Bull case, about $129 per share. US volumes reach 14.5 GW in 2027 and 16 to 16.5 GW from 2028, as the existing plants run full and the South Carolina finishing line adds some qualifying output. The unsubsidized business returns to about $3 per share a year, near its 2024 level once 2025's one-off logistics and warehousing costs fall away, and holds there through 2045 behind the US tariff wall. What would make this wrong. Nearly all of the value is a government subsidy that ends in 2032. A change in law, or a ruling that First Solar fails the new foreign-entity tests, would hit every case. The key long-run question is whether US-made thin-film panels can earn a profit without the credit. The evidence so far is thin: the business made little or nothing without it in 2025, and the decade before the credit shows little owner cash profit after capital spending. We did not subtract debt. The company held about $2.8 billion of cash at the end of 2025 against modest borrowings, and that cash is not added to the values above. The margin-of-safety price of $55 sits below even the bear case. From that price, the base case implies the credit payments of 2027-2032 alone return roughly 20% a year, the margin required for a business whose earnings are this dependent on policy.