Amentum Holdings AMTM

Current Stats

Post-Spinoff Performance

Computed from split-adjusted closing prices. Returns are total return — (end price + dividends received − start price) ÷ start price — with no reinvestment assumed.

⚠️ Correction (September 25, 2026) — the September 22 close was $19.70, not $19.71

Every figure on this page dated September 22 was struck against a close of $19.71. The true close is $19.70. The analysis was published on September 23, when the provider had lost the September 22 session entirely; the price used was reconstructed from hourly bars, which do not contain the closing auction print. Upstream repaired the session on September 25 and the real bar replaced the reconstruction across the archive.

What this changes — two figures, by $2m each:

	As published	Corrected
Market capitalisation	$4,819m	$4,817m
Enterprise value	$8,235m	$8,233m

What it does not change — everything else. At $19.70 the multiples and returns round exactly as printed: EV/EBITDA 7.30×, P/E 8.04×, −38.9% from the $32.25 anchor, −47.5% from the $37.53 peak, 11.4% FCF yield, 3.03× net leverage. The BUY rating, the $25 target and the B (2.73) grade are untouched, and nothing in the thesis turns on one cent.

Corrections on this site are additive: the prose below is left exactly as published, because it is the record of what the analysis was written against. Read it with this correction in hand.

Post-Spin Analysis — September 23, 2026

Written against the September 22 close of US$19.71. Amentum files as a US domestic issuer on Form 10-K and 10-Q (CIK 0002011286), with a fiscal year ending in late September or early October — FY2026 ends about 2026-10-02, a week from this writing. Every figure below names its fiscal year, because this company’s reported history changes basis twice. Backlog and remaining-performance-obligation figures are taken from the filings and from SEC XBRL company facts directly. Price, volume and cohort figures are from this site’s own archive.

The one-line version

The headline loss is measured from a price that existed for one auction. Amentum’s published return of −38.9% since day one is anchored to a close of $32.25 on 2024-09-30 — a day on which 41% of all volume printed in the closing auction and the stock closed at its exact high, 25.6% up on the day. That is an index-inclusion print, not a price a shareholder could have lived with. It reverted −20.2% the next session.

Twelve weeks later the same mechanism ran in reverse. On 2024-12-20, 33.3 million shares — 88% of the entire day’s volume — printed in the closing auction, about 14% of the company in a single cross. Measured from any non-auction baseline, Amentum’s full-life return is roughly −28%, against a nine-peer median of −25.3%. Over its whole life it has performed in line with its sector.

But the last seven months are all Amentum, and that is the real story. From its February 6, 2026 peak the stock is −47.5%, while SAIC rose 31.8% and V2X rose 8.4% over the identical window. The sector news in that period was good: DOGE was disbanded in November 2025, full-year appropriations passed in February 2026 with defence $8.4bn above request, and the FY2027 request proposes a large defence increase. Amentum had better news than Parsons and fell harder.

What broke is revenue, not earnings. Adjusted EPS is +22% since the spin, EBITDA +7%, leverage down from 4.10× to 3.03×, no dilution — and the multiple went from 16.0× to 8.1×. The whole decline is a de-rating. The thing justifying it: remaining performance obligations have fallen from $12.9bn to $8.9bn, −31%, while the company reports a record $48.2bn backlog.

And on Monday the handcuffs come off. The tax-matters agreement has barred Amentum from any merger, change of control, or repurchase of any stock since the spin. That restriction ends 2026-09-28. It has been legally unable to buy a single share through a 47.5% drawdown, at an 11.4% free-cash-flow yield.

Rating: BUY · Target US$25 · Grade: B (2.73)

What the business actually is

Amentum is a US government services contractor — engineering, technical and mission support sold almost entirely to federal agencies. It was created on 2024-09-27 when Jacobs Solutions separated its Critical Mission Solutions segment and merged it with legacy Amentum, then owned by American Securities and Lindsay Goldberg, in a Reverse Morris Trust.

The economics are a labour-and-contracts business, and the mix is the margin. Amentum carries the highest cost-plus share of its peer group — 63% of revenue, falling to 56% in the first nine months of FY2026 — and, not coincidentally, the lowest margin at about 7.7%. These are the same fact stated twice: cost-plus work transfers risk to the customer and pays less for it. Do not read the margin improvement as execution. Part of it is that mix shift, which is risk moving onto Amentum’s books, and part is the transition of consolidated joint ventures to equity-method accounting, which deletes revenue while keeping the profit line.

The price history, and why the published figure misleads

This is the most important section on the page, because every other number on this site for Amentum is measured from the anchor examined here.

The first session, 2024-09-30. The stock opened at $26.25, traded between $23.58 and $26.67 for six hours, then ran from $25.94 to close at $32.25 — the exact high of the day, up 25.6%, on 8.99× normal volume. Hourly bars account for only 10.8m of the 18.4m shares traded. The missing 7.5 million shares — 41% of the day — printed in the closing auction.

The second session it fell 20.2%, to $25.75.

Then, on 2024-12-20, 37.8 million shares traded — 17.6× baseline — and 33.3 million of them, 88% of the day, printed in the closing auction. That is approximately 14% of the company changing hands in one cross, and it is the signature of an index deletion. The stock had already slid from $23.78 to $18.93 in the four sessions running into it.

What I am asserting and what I am not. The auction volumes above are measured from market data and are facts. The natural explanation is that Amentum was added to a major index at separation and removed from it about twelve weeks later, which would make the day-one close the price index funds were compelled to pay. Secondary sources describe exactly that path. I could not verify the index membership against the index provider’s own announcements, so the attribution is my inference and is labelled as one. The volume evidence stands on its own regardless of what the index was called.

The shape is a double round trip, not a decline. Down 49%, up 128%, down 47.5%. Any account of Amentum as a stock that fell on day one and kept falling is simply wrong, and the interval grid shows how much of the damage is the first session alone: D1 −20.2%, M1 −6.6%, M3 −36.5%, M6 −44.0%, Y1 −25.7% — with zero income at every horizon, because no dividend has ever been paid.

Why this page does not change the anchor. The site’s methodology sets the day-1 close as the anchor precisely because it is attainable via a market-on-close order — which is exactly what that auction was — and it tests suspect prints by asking whether the extreme ever traded again. It did, decisively: the stock traded above $32.25 on 26 later sessions and closed above it on 24. The anchor is correct by the site’s own rules. It is the interpretation that needs the warning, not the data.

The financial record — and why it comes in three bases

Read the reported growth with care; most of it is not growth.

The legal and accounting directions are inverted. Legacy Amentum is the accounting acquirer and predecessor; the Jacobs entity is the legal survivor. So FY2022–FY2024 GAAP figures are legacy Amentum alone, with no Critical Mission Solutions at all, and the merger closed on the last day of FY2024. The resulting “+72% revenue” is entirely an artefact of the basis change. FY2025 was also a 53-week year, which the company sized at roughly $650m of revenue and $32m of EBITDA.

On a like-for-like basis, US$ millions:

Two years, no revenue growth. The FY2026 guided midpoint of $13,875m sits essentially on top of FY2024 pro forma $13,858m. Everything good on that table happens below the revenue line.

The August 2026 guidance change is the single clearest signal in the file. The company cut revenue guidance for the first time in its history — from $13,950–14,300m to $13,800–13,950m — while simultaneously raising EBITDA and EPS guidance. Management’s own framing was that “near-term dynamics have impacted our revenue outlook.” Beating on earnings while missing on revenue is the textbook setup for multiple compression, and the multiple duly compressed.

Guidance has otherwise been met or beaten. FY2025 finished above the top of every guided range — revenue $14,393m against $13,975–14,175m, adjusted EBITDA $1,104m against $1,065–1,095m, EPS $2.22 against $2.05–2.20.

Three things to hold against the earnings quality:

• Integration costs are rising, not falling — $39m, then $62m, then $85m, three years in, and all added back to adjusted EBITDA.
• Receivables factoring is doing real work in the cash flow. A factoring programme contributed $120m of first-nine-month FY2026 operating cash flow against $42m the year before — $78m of a $235m total, disclosed only in a footnote.
• The “Pro Forma” comparative label was dropped in Q1 FY2026 — the first genuinely comparable quarter is also the first to print a revenue decline.

What is clean: Ernst & Young throughout, internal control over financial reporting effective, no material weakness, no restatement, no auditor change, no goodwill impairment since the merger, covenant compliant in every period, and shares outstanding up just 0.5% in two years.

The backlog contradiction

This is the fact that justifies a discount, and the company’s own headline obscures it.

These point in opposite directions, and RPO is the one that binds. Total backlog includes unexercised option years — work the customer may or may not ever order. Remaining performance obligations are, broadly, the enforceable book. The enforceable book has shrunk by nearly a third while the advertised one hit a record. (RPO figures taken directly from SEC XBRL company facts; the series is not monotonic — it recovered to $10.8bn in Q1 FY2026 before falling to $8.9bn, which is the lowest reading on record.)

Funded backlog fell to $5.59bn at FY2025 year-end before recovering to $6.24bn — the fingerprint of the appropriations disruption, and again invisible behind the headline.

Bookings strength is also less than it looks. Quarterly book-to-bill of 1.64× in Q4 FY2025 falls to roughly 0.6× if a single $4bn protest release is excluded, and Q3 FY2026’s 1.13× falls to roughly 0.71× excluding two large awards. Two quarters — FY2025 Q2 at 0.89× and Q3 at 0.96× — came in below 1.0×, and those are the two the company chose not to disclose on a quarterly basis. They bracket the April 2025 low almost exactly.

Management, and a governance gap worth naming

Stability is genuine and unusual. Two years in there has been no CFO change — Travis Johnson has held the role throughout — no COO change, and only four executive-change 8-Ks in the company’s life. The lineage reflects a real merger rather than a carve-out: the CEO came from legacy Amentum, while the COO (29 years at Jacobs) and the Executive Chair came from the Jacobs side.

Insiders have never sold a share. Across all 82 Form 4s filed in company history there is not one open-market sale. Every disposition is tax withholding or a gift. In the other direction, Executive Chair Steve Demetriou bought 100,000 shares of his own money at $20.81 on 2025-02-12 — four weeks before Jacobs’ block cleared the market at $16.50 — and a director bought 4,600 more. That is a meaningful signal in a name this cheap.

The compensation gap is the thing to flag. The annual incentive is Adjusted EBITDA 65% / days-sales-outstanding 20% / net debt reduction 15%, and it paid 113% of target for FY2025. The long-term plan is three-year cumulative free cash flow and adjusted EBITDA, 50/50, with no relative total-shareholder-return component. And searching both plans for a return-on-capital measure — ROIC, ROCE, return on invested capital — returns nothing.

For a levered, goodwill-heavy roll-up, that means management can be paid in full for buying EBITDA at any price. Nothing in the incentive structure asks what the EBITDA cost. This is the same gap noted on South Bow, and it is worth the same markdown.

One structural governance note: the Lead Independent Director is a managing director of American Securities — an 18% holder and the firm that controlled the company before the spin. Lindsay Goldberg’s nominee chairs Compensation, and sponsor designees sit on all three committees. The ownership tier entitles them to this; it is disclosed and lawful. It is still worth knowing that the independent counterweight to management is partly appointed by the selling shareholders.

Two loose threads I could not close. A named legacy matter — DOE-related claims carried at $138m asset and $125m liability in the FY2024 Form 10-K — appears nowhere in the FY2025 10-K or any FY2026 10-Q, with no explanation given. And one director drew only ~90.4% support at the February 2026 meeting against ≥97% for everyone else, with no reason disclosed.

What the separation did to the share register

The value transfer out. On 2024-09-27 the Jacobs entity borrowed $1,130m and paid $911m of it straight to Jacobs as a dividend, with the debt folding into Amentum’s $3,750m term loan the same day. This is the leveraged-dividend pattern this site has documented across Qnity, FedEx Freight, Versigent, Ralliant and South Bow: the debt and the payment are the same number, and the separated business services it.

The overhang that resolved. Jacobs retained 29,196,261 shares — 12.0% on day one, and announced on day one that it intended to dispose of the stake within twelve months. So the market traded the entire first six months knowing a seller was coming. It cleared on 2025-03-12, in one shot: Jacobs exchanged 19,464,174 shares for its own debt held by a bank, which resold them at $16.50 — a 17.5% discount to the prior close, and 49% below the day-one price. Jacobs has held nothing since.

The overhang that has not resolved, and it is larger. American Securities and Lindsay Goldberg each hold 43,893,904 shares — about 36% combined. Their lock-up expired 2025-09-27. A shelf registering 90,053,486 shares for resale was filed four days later, on 2025-10-01, and has never been used.

Our own data corroborates that independently. The nightly share capture puts free float at 62.4% of 244.5m shares outstanding — leaving about 92m shares held back, which matches the sponsors’ ~88m almost exactly. Two years after separation, roughly 36% of this company sits with two private-equity firms who have registered it for sale and are waiting.

That is the best available explanation for a seven-month, low-news grind that no peer shared. And the precedent for how it clears is Jacobs’: a block at a 17.5% discount.

Valuation

The capital structure. 244.5m shares at $19.71 is a market capitalisation of $4,819m; with net debt of about $3,416m the enterprise value is $8,235m.

The de-rating, isolated

Earnings went up and the multiple halved. Had the spin-day multiple held, the stock would be about $39. And note the middle column: at the February peak the market had restored the original multiple almost exactly, on EBITDA up just 5% in 17 months — three days before a quarter that printed a 5% revenue decline. The peak was the market re-underwriting the original growth story; the halving since is it changing its mind.

Against the peer group

Computed from primary filings at 2026-09-22 closes, each company on its own adjusted-EBITDA definition.

CACI is excluded from the median: it publishes plain EBITDA with no add-backs, so its 15.6× is not comparable. Amentum’s own definition adds back both stock compensation and amortisation, so its discount is perhaps half a turn narrower than the raw figures suggest.

Amentum has the cheapest adjusted P/E in the group and trades at a 22.8% discount to the peer median EV/EBITDA of 9.45×. It also trades 2.7 to 4.2 turns below the multiple at which it was itself created in the 2024 merger. Re-rating to the peer median alone is worth about $29.63.

Price target

$25 is about 27% above the current price, and deliberately below the peer-median re-rating. Amentum should trade at a discount to peers: its revenue is shrinking while most of the group’s is not, its enforceable backlog is down a third, and roughly 36% of its shares are registered for sale. The call is not that it deserves the group multiple — it is that an 8× multiple and an 11.4% cash yield already price a worse outcome than the company is currently reporting.

What this most resembles

Amentum is a Reverse Morris Trust, and that is the worst structure in this site’s universe.

Twelve of thirteen RMTs with a measurable excess return have lagged the index, and the exception is Wabtec at +3.8pp a year. The pattern holds even for the apparent winners: CEVA returned +553% in total and still lagged by 3.3pp a year; Comcast returned +338% and lagged by 5.2pp. The structure produces absolute returns and systematic underperformance.

The 2005 ACCO row is excluded — its price series begins more than 180 days after the spin and belongs to a different company.

Amentum at −39.8pp a year is the second-worst in that cohort, behind Magnera. It is also the only row in the universe combining an RMT with a retained parent stake, and it sits at the 25th percentile of all 349 measurable rows.

Why the cohort is informative here rather than merely gloomy. An RMT merges a carve-out into an existing shareholder base and typically leaves a parent stake plus, as here, sponsor holders — so the register is crowded with holders who did not choose the asset. Amentum’s history is that mechanism running to completion: an index add, an index delete, a pre-announced parent seller clearing at a 17.5% discount, and 36% still waiting. The structural read and the price history say the same thing.

In depth: how much of this is the sector?

Almost all of the full-life decline, and almost none of the last seven months.

The third row is the finding. Over that identical window SAIC rose 31.8% and V2X rose 8.4% — both verified from this site’s own price archive — while CACI fell 0.9%, ICF fell 4.0% and Jacobs fell 3.9%. Amentum underperformed SAIC by roughly 80 percentage points in seven months.

And the sector backdrop in that window improved. DOGE was disbanded in November 2025. Full-year FY2026 appropriations passed on 2026-02-03 with defence $8.4bn above the request. The FY2027 request proposes a large further increase. CACI raised guidance every quarter and beat; Jacobs raised three quarters running. Only Parsons cut — and Amentum, which raised its earnings guidance, fell harder than Parsons did.

The more useful frame than “civil versus defence” is reconciliation money versus annual appropriations. CACI says so explicitly: reconciliation funding “remains available… regardless of whether Congress enacts full-year appropriations, passes a CR, or enters a government shutdown.” That is why it sailed through a 43-day shutdown with guidance rising. Amentum’s larger segment sits on the annually-appropriated side, and its revenue line shows it.

One new company-specific risk appeared in the Q3 FY2026 filing and is absent from the FY2025 10-K: “NASA’s increased focus on insourcing certain activities.” It appeared in the same filing as the only revenue guidance cut, and the stock fell 8.2% the next session.

In depth: is Amentum an acquisition target?

For two years the answer was legally no. From Monday it changes.

The tax matters agreement bars Amentum, for the restricted period following the 2024-09-27 distribution, from any merger, any change-of-control transaction, asset sales above 30% — and, quoted, from “redeem or otherwise repurchase (directly or through an Affiliate) any stock.” Every waiver sits at Jacobs’ sole and absolute discretion. The restricted period ends on the first business day after the two-year anniversary: Monday, 2026-09-28.

The 10-K describes the restriction but never states the end date. You have to derive it from the distribution date.

What makes it live rather than theoretical:

• No poison pill and no classified board.
• Roughly 36% of the register is held by two private-equity firms who have registered their entire position for resale and have been unable to exit at a price they like.
• It is the cheapest adjusted P/E in its peer group, at 8.0×, with an 11.4% cash yield.
• Four separate spin-era governance arrangements — the Executive Chair’s contracted term, the sponsors’ obligation to vote for Jacobs-designated directors, the lead-independent-director split, and a standstill on the sponsors buying more — lapse in the same week.

The obvious first use of the unlock is a buyback, and there is a complication. An executive order of 2026-01-07, Prioritizing the Warfighter in Defense Contracting, limits buybacks and dividends for certain defence contractors. Booz Allen discloses it as a risk to its own capital return; Amentum does not address it anywhere. Any buyback thesis has to clear that order first, and the company has not said whether it does.

What this means for the score: Acquisition Potential is genuinely elevated — a cheap, unlevered-enough, pill-free asset with a motivated 36% holder and a legal barrier expiring in days. It is the one dimension where Amentum scores above average.

Tail scenarios

Upside

A buyback lands (2σ). At an 11.4% FCF yield, retiring 10% of the shares costs about $480m — roughly ten months of free cash flow — and would be immediately and materially accretive. The unlock is dated. Worth $3–5 a share, and it is the cheapest upside available because it requires nothing operational.

The sponsors clear in one block (2σ, and it is upside). Counter-intuitively, a single discounted block that removes 36% of the register would end the overhang that has suppressed the stock for seven months, exactly as Jacobs’ March 2025 block marked the bottom before a 128% run. The discount hurts for a week; the removal is worth more. Worth $2–4 on a twelve-month view.

A re-rating to the peer median (1–2σ). Nothing operational required: 9.45× EV/EBITDA is $29.63.

A bid (3σ, but live from Monday). No pill, no staggered board, 36% motivated, cheapest multiple in the group.

Downside

Revenue decline continues and RPO keeps falling (2σ — the central risk). The enforceable book is down 31%. If FY2027 revenue guides down again, the earnings growth that makes 8× look cheap stops, and the multiple is then correct rather than depressed. Worth −$4 to −$6.

The margin story is exposed as mix (2σ). Cost-plus fell 63% → 56% in a year, partly tracking an April 2026 executive order mandating firm-fixed-price contracting. Fixed-price work carries execution risk Amentum has not had to demonstrate. A single bad fixed-price programme in a 7.7%-margin business is worth several cents of EPS and more of the multiple.

The sponsors dribble rather than block (2σ). The worst outcome is neither holding nor a clean exit but a slow, visible selldown that caps the stock for another year.

A re-leveraging acquisition (2σ). With no return-on-capital metric in either incentive plan and a stated roll-up heritage, the unlock permits M&A as well as buybacks. Management is paid on EBITDA and free cash flow, not on what the EBITDA cost.

What would change this view

Better:

• Remaining performance obligations stabilising or rising — the single most informative number in the FY2026 10-K, due late November
• A buyback authorisation, and an explanation of how it sits with the January 2026 executive order
• The sponsors exiting in one block
• FY2027 revenue guidance at or above FY2026

Worse:

• RPO falling below $8.9bn
• A second consecutive revenue guidance cut
• Integration costs still rising in year four
• A large acquisition announced in the weeks after the restriction lapses

Dated:

• 2026-09-28 — the tax-matters restricted period ends. Buybacks, M&A and a change of control become legally possible for the first time
• Late November 2026 — FY2026 results and the 10-K, on the precedent of last year’s 24 November release. First full-year RPO and funded backlog disclosure, and FY2027 guidance
• ~2026-10-02 — FY2026 ends

Thesis-breaking:

• FY2027 revenue guided down again with RPO still falling — at that point the discount is the right price, not an opportunity

Investment Scorecard

What moved and what deliberately did not.

This is Amentum’s first investment grade on this site. It has been carried as a catalogue row since the 2024 backfill and has never appeared in a monthly report.

The backlog contradiction is charged once, to Competitive Position. It would be easy to charge it again to Financial Profile — a shrinking enforceable book is a financial fact — but the flat revenue line is already doing that work there.

The governance items are charged once, together, to Management. The missing return-on-capital metric, the sponsor-affiliated lead director and the vanished legal matter are three faces of one question, and they are worth one markdown.

Strategic Rationale is where the structure is charged, and only there. The index churn, the leveraged dividend and the parent stake are all consequences of how this deal was built.

Rating: BUY. Target US$25. This is a cheapness-and-catalyst call, not a quality call, and the distinction matters. The business is not improving — revenue is flat to down and the enforceable backlog has shrunk a third. But the market has taken 47.5% out of the equity in seven months while the sector rose and while the company raised its earnings guidance, and it did so against a company that was legally forbidden from buying a single share of its own stock throughout. On Monday that prohibition ends. At 8× earnings and an 11.4% cash yield, the price already assumes the deterioration continues; the guidance says it is not, yet.

The overhang is the reason it is cheap, and it is also the risk. Roughly 36% of this company is registered for sale. Jacobs’ precedent says it clears at about a 17.5% discount — and that Jacobs’ block marked the low before a 128% run.

Sell the thesis, not the stock, if remaining performance obligations fall again in the November 10-K. That is the number that decides whether 8× is cheap or correct.

Nothing on this page is investment advice. Figures are drawn from the primary filings, from SEC XBRL company facts and from this site’s own price archive; where a figure or an attribution is my own calculation or inference rather than a company statement, that is said in the text.