Amentum Holdings, Inc. (AMTM) stock intrinsic value fair value estimate & owner earnings analysis

Amentum is a large government services contractor. It runs nuclear sites, maintains military aircraft and equipment, provides engineering, logistics, digital and intelligence support, and supports NASA programs, mostly under multi-year contracts. Today's company was formed on September 27, 2024, when the old Amentum (a private-equity-owned contractor built from DynCorp and PAE) merged with Jacobs' Critical Mission Solutions and Cyber & Intelligence businesses. In June 2025 it sold its hardware unit, Rapid Solutions, to Lockheed Martin for $360 million. Of fiscal 2025 revenue of $14.4 billion, 53% came from the Department of War and the intelligence community, 28% from other U.S. agencies (the Department of Energy, NASA and others) and 19% from international and commercial customers, mainly in the United Kingdom and Australia. 63% of revenue is cost-plus-fee, 24% fixed-price and the rest time-and-materials. Contracts are re-competed every few years, so the business lives on winning and re-winning work. It pays no dividend. How we measure earnings. We start from cash from operations, subtract stock compensation and capital spending (small, about $25 million a year), and add back cash interest so we can value the business before its lenders and subtract the debt separately. We do not count the non-cash charge for amortizing contracts acquired in the merger ($479 million in fiscal 2025, about $375 million this year) as a cost, because the cost of winning replacement work is already in the cash numbers. Whether that work is actually being replaced is therefore the central question, and the cases below carry the answer: none of them assumes the contract base grows on its own. Amentum keeps putting money into its joint ventures ($56 million in fiscal 2025, $53 million in the first nine months of fiscal 2026, partly offset by returns of capital), so we subtract about $35 million a year for that. Working capital has consumed cash in each of the last four years, including $61 million in fiscal 2025, much of it from paying down long-term liabilities, and we treat about $50 million a year of that as a lasting cost. Only fiscal 2025 (a 53-week year) describes the combined company; earlier years were the old Amentum alone, so the last completed year gives no like-for-like growth rate. In fiscal 2025 the business produced about $862 million of cash before interest, or about $811 million after removing the extra week and the joint-venture funding. The first nine months of fiscal 2026 were about 17% ahead of the same period a year earlier ($668 million vs $570 million) despite the sale of Rapid Solutions. Three things drove that: merger integration costs fell from $62 million to $36 million, joint-venture payouts rose from $57 million to $87 million, and favorable changes in estimated contract profits doubled from $40 million to $83 million. The last of these is unlikely to repeat at that pace, so we set aside about $45 million of this year's figure. That gives a current level of about $870 million, or $820 million ($3.35 a share) after the working-capital cost, and we start the base case at no growth, below this year's pace. After interest of about $240-250 million, owners keep roughly $570 million (about $2.35 a share), in line with the company's own expected free cash flow for the year. What the contracts say. Management reports a $48.2 billion backlog, but almost all of its growth is management's estimate of unexercised options and future task orders. Work actually under contract, as the audited accounts measure it, fell from $12.9 billion in September 2024 to $9.9 billion in October 2025 and $8.9 billion in July 2026. Part of that drop is the Rapid Solutions sale and contracts moving into joint ventures, whose earnings still reach Amentum as payouts. Funded backlog rose 10% over the year to $6.2 billion, and third-quarter revenue fell 2%, all of it from the sale and the joint-venture transfers; the work that stayed grew about 1%. Full-year revenue is expected at $13.8-13.95 billion, about 3.5% below fiscal 2025. We give no credit to the unfunded backlog, to management's cost-saving targets, or to the Sellafield joint venture in the United Kingdom, where Amentum is preferred bidder but has no signed contract. Base case ($40 a share). Cash before interest stays at $820 million a year with no growth. Flat in dollars means the business shrinks a little each year after inflation, which matches contracted work that has stopped falling but is not growing. We assume it continues after year 10 because this kind of outsourced government work has been carried out for decades by the predecessor companies and is re-won in recompetitions. The business is worth about $13.7 billion ($55.8 a share). It owes about $3.9 billion ($16.0 a share): $3.875 billion of debt, $94 million belonging to outside partners in its subsidiaries and about $61 million for the extra cost of its 7.25% notes, less about $117 million of surplus cash. Of the $459 million cash on hand at July 3, 2026, $142 million had been collected on receivables already sold to a bank and is owed to it, and we keep about $200 million to run the business. That leaves about $40 a share. Bear case ($16 a share). This case combines today's visible erosion with a full government-spending downturn of the kind this industry has actually had. First, the run-off of contracted work seen in fiscal 2025 continues through older program wind-downs and recompetition losses, taking cash before interest to $780 million and then $740 million in years 1 and 2. Then a downturn on the scale of the 2011-2015 Budget Control Act cuts and the Iraq and Afghanistan drawdown arrives. In that episode Defense Department contract spending fell 31% from fiscal 2011 to 2015 while civilian agency spending held roughly steady, and Amentum's own predecessor DynCorp saw revenue fall from more than $3.2 billion in 2012 to about $1.9 billion in 2015 as its war-related work ended. Amentum is more diversified today, so we apply a 20% decline to its Department of War and intelligence revenue, 5% to other U.S. agencies and none to international work, a 12% fall in total. With overhead spread over less revenue and price pressure on recompetes, the cash margin slips about one point, from about 5.7% to 4.75%. That takes cash before interest to about $540 million by year 5. Spending recovered after Congress raised the caps from 2013 onward, but contractors did not win all their work back, so we let a third of the lost ground return, to $600 million by year 10, and then shrink 3% a year as lost contracts are not fully replaced. The trough lands on the 2031 term-loan and 2032 note maturities: after using its cash to pay down about $1.5 billion of debt, the company refinances the remaining $2.4 billion about 3.5 points more expensively for seven years, which adds about $0.4 billion to what it owes. The old Amentum's goodwill write-offs of $108 million in fiscal 2022 and $186 million in fiscal 2023 show that acquisitions in this industry do not always pay off. The business is worth about $8.2 billion ($33.6 a share) against about $4.3 billion owed ($17.6 a share), leaving about $16 a share. Even at the trough, cash flow covers the interest bill a little over twice. Bull case ($51 a share). Cash before interest runs at this year's pace of about $900 million, with contract profit estimates staying favorable, and rises to $945 million as the remaining merger integration costs (about $45 million a year) end. It then grows 1.5% a year through year 5, to about $990 million, as the 10% rise in funded backlog converts to revenue and cost-plus contracts in force pass wage increases through to fees, and stays flat after that because those contracts will be re-competed. The business is worth about $16.3 billion ($66.6 a share), less $3.9 billion owed ($16.0 a share), leaving about $51 a share. There have been no new filings since our last review, so the business facts are unchanged. The previous bear, bull and margin-of-safety figures were set on a different basis and are not comparable. Margin-of-safety price ($19). We are moderately sure of the earnings level. It rests on one full year of the combined company plus nine months that point the same way, and it matches the company's own cash-flow outlook, but two of our judgements could be wrong: if the working-capital drain is $100 million a year rather than $50 million, the shares are worth about $3.40 less, and if contracted work resumes its fiscal 2025 rate of decline, the base case drifts toward the bear. The worst plausible outcome is a spending downturn arriving just as the term loans come due in 2031, combined with the loss of one or two large recompetes, or management using the balance sheet for another large debt-funded combination, which is how this company was built. The business would survive the first: net debt of $3.9 billion is a little under five years of cash before interest, interest is covered about twice even at the trough, and the customer is the U.S. government. But its assets are people and contracts, not hard assets, so there is nothing to fall back on if the earnings go, and shareholders could wait many years. Over the holding period the company pays no dividend, but the roughly $2.35 a share of cash it keeps after interest goes to paying down debt (it prepaid $125 million of its term loan in the last quarter), and that moves value from lenders to shareholders each year. If the base case holds, that alone lifts the shares' value to about $47 in three years and $53 in five. From $19, the base case implies about 35% a year over three years or 23% a year over five, and still about 20% a year over five if the working-capital drain proves twice our estimate. In the bear case the debt paydown before the downturn leaves the shares worth about $22 after five years, so from $19 the bear case implies roughly the starting value. At $22 the base case over five years would fall just short of 20% a year, so $19 is the highest price that clears the target with room for our most likely mistake. What would make this wrong: contracted work falling again rather than holding near $9 billion, a sustained cut in U.S. defense, NASA or Energy Department spending, several large recompetition losses in a row, a large fixed-price loss, or another debt-funded acquisition. On the upside, if funded backlog keeps rising and revenue returns to growth after the joint-venture transfers, the base case is too low. Debt is the main amplifier. At July 3, 2026 the company owed a $1.4 billion term loan due April 2031 and a $1.59 billion term loan due September 2031, both secured on substantially all assets and together about $2.9 billion after a $125 million voluntary prepayment, plus $1 billion of 7.25% notes due August 2032. About $1.3 billion of the floating-rate loans is fixed through interest-rate swaps. The company was in compliance with its covenants and had $917 million available on its $1 billion revolving credit line. It also sells up to $400 million of receivables to a bank. We treat that as part of operations because its cost is already in the earnings, but losing the facility would tie up about $300 million of cash. Shareholders are paid only after the lenders.