On the July earnings call, CEO Jim Taiclet laid out the arithmetic that has governed American munitions manufacturing for decades. Put $10 million of robotics into the missile plant at Camden, Arkansas, cut the cost per interceptor from $100,000 to $90,000, and the government reprices you down 10%. "Who would invest $10 million to get their price down 10% and bleed out all of that benefit?" he asked. "That's the issue with our industry, is essentially that accounting system. What the framework agreements do is take that away."
That mechanism, not the backlog headline, is the thing worth understanding about Lockheed Martin right now. This is a $124 billion company that drew 70% of first-half 2026 sales from the U.S. Government, 61% of them from the Department of War , across four segments: Aeronautics, which builds the F-35, F-16, F-22 and C-130 and accounted for roughly 40% of second-quarter sales; Missiles and Fire Control, the interceptor and tactical-missile business behind the Patriot PAC-3 and THAAD air-defense systems and the HIMARS rocket artillery family; Rotary and Mission Systems, housing Sikorsky helicopters, Aegis and naval combat systems; and Space, home to the Trident missile, the Next Generation Interceptor and Lockheed's half of United Launch Alliance . The quarter reported July 23 delivered an 11% sales increase to $20.1 billion, $7.94 of diluted earnings, $2.9 billion of free cash flow, $65 billion of new orders and a record $230 billion backlog . Every line of 2026 guidance went up.
The stock is trading at about $536.59 intraday, roughly 22% below the $692 high it set in early March . That gap is the argument.
The order book didn't just get bigger; the contracts changed shape
Late in June the Missile Defense Agency signed a seven-year, $35 billion contract to quadruple THAAD interceptor output. The Army added a $3 billion GMLRS rocket production contract and a HIMARS award worth up to $1.1 billion . Missiles and Fire Control backlog went from $46.7 billion at the end of 2025 to $87.9 billion at June 28 — an 88% step in six months .
Volume is not the interesting part. Defense order books have swelled and disappointed before. What is different is the instrument. These awards arrive as multi-year undefinitized contract actions, and the CFO was explicit about what that buys: they "allow us to get going at full speed," let Lockheed "optimize around a seven-year production run and get our suppliers aligned to that," and provide "a funding allocation" plus "the cash terms to allow us to protect the upfront investment" . The framework layered on top changes the economics. Taiclet described a fixed price to the customer with an escalator, under which Lockheed keeps efficiency gains it generates during the ramp and shares them with the government only after hitting "the top mark of our fully retained margin expansion" . Under the old cost-based repricing, the savings went straight back to the customer; here the contractor keeps them for a defined stretch. That is why a company that has spent two decades refusing to put capital in ahead of orders is now doing exactly that.
The near-term cost is visible and bounded. The CFO put steady-state Missiles and Fire Control margin in "the high 13s, low 14s" ; management put ramp dilution at 20 to 30 basis points "over the next few years with a goal to get to higher than historic margins" . The segment printed a 14.5% margin in the second quarter on 19% sales growth, though that included $190 million of net favorable profit booking-rate adjustments, so the underlying rate sits nearer the guided band than the headline suggests . The dilution is a ramp artifact. The retained-efficiency provision runs with the contract.
One objection deserves naming rather than burying: an undefinitized contract action is by definition not yet fully priced, and definitization is where the government can claw terms back. That risk is real and it is the single thing I would watch hardest at the Q3 print. But THAAD has already converted from framework to a signed multi-year contract, which is precisely the step that took $41 billion of funded work into the backlog, and PAC-3 is the next one management has said it intends to land in the second half.
What broke the stock in April, and what has been grinding it since
Two separate things moved this price, and conflating them is the error.
The first was the first-quarter print on April 23. Sales were flat year over year at $18.0 billion, earnings fell to $6.44 a share from $7.28, and free cash flow swung to negative $291 million from positive $955 million . Aeronautics absorbed $125 million of unfavorable adjustments on the F-16 and management merely reaffirmed full-year guidance rather than raising it. The stock fell from a $592.19 weekly close to $513.45 in the following week — a 13% move — and slid to an intraday $500.31 by early May . That is the market pricing the thing this franchise has repeatedly delivered: a good order book that converts into disappointing earnings.
The second has little to do with Lockheed's own disclosures. Shares recovered to an intraday $612.75 in the week of August 18 after the July raise, then drifted back to current levels with no company-specific release in between. What did happen is that the 10-year Treasury yield touched 5.041% on September 15, its highest since 2007, with the 30-year at 5.401% . Lockheed's dividend yield near 2.6% (annualizing the $6.90 per share paid in the first half ) and a free cash flow yield of 5.7% at the guidance midpoint compete directly against a 5% risk-free rate. I cannot rule out that something company-specific contributed, but nothing was disclosed, and the timing lines up with a repricing of duration across every long-dated asset.
The distinction matters because only the first is a claim about the business. The April de-rate was earned. The subsequent drift has handed back most of what a broad-based guidance raise bought.
Missiles carries the growth; Aeronautics carries the risk
Management raised the full-year sales outlook for all four segments in July and the profit outlook for three of them. Missiles and Fire Control grew sales 14% in the first half and is guided to a faster second-half rate still; Rotary and Mission Systems moved to $17.7–18.1 billion of sales on $1.86–1.89 billion of profit — roughly a 10.5% margin — on the strength of new radar awards, including a $2.3 billion contract signed in the quarter . Space was the only segment whose profit outlook was cut, to $1.34–1.38 billion, on reduced equity earnings from United Launch Alliance after the Vulcan Centaur anomaly . Rotary is also where the pending $3.45 billion Ultra Maritime acquisition, signed July 6, will land — undersea sensing and anti-submarine warfare, a domain the FY2027 request funds heavily.
Aeronautics is the segment that should worry a bull, and not for the reason the headlines suggest. F-35 deliveries fell to 19 in the second quarter from 50 a year earlier . On cost-to-cost percentage-of-completion accounting — revenue recognized as costs are incurred, not when an aircraft rolls out the door — that is largely noise; Aeronautics sales still grew 9%. The real signal is backlog, which has fallen from $59.4 billion at year-end to $54.4 billion over two quarters , and the top of the segment's 2026 range depends explicitly on the timing of the next F-35 production lot award . With the F-35 at 27% of consolidated sales and 67% of Aeronautics in 2025 , a segment that goes sideways while munitions compounds is a genuine mix shift. I think the market is discounting that correctly rather than over-discounting it. July's $1.6 billion F-35 spares contract, the largest in program history, is a real sustainment tailwind , but sustainment does not substitute for a production lot.
The capital going into this ramp is working capital, not concrete
Here is the piece I think is mispriced. Lockheed cut 2026 capital expenditure guidance to $2.0–2.4 billion from $2.5–2.8 billion while raising free cash flow by the identical $450 million to $7.0–7.2 billion, leaving cash-from-operations guidance unchanged at a $9.3 billion midpoint . The bearish read writes itself: the cash improvement is deferred capital, not created capital. But the cut was attributed to efficiencies in the munitions build-out, the framework awards are structured cash-neutral, and the customer is funding the tooling through the contract — which is what the retained-efficiency provision exists to let Lockheed exploit.
Run the return implied by that. First-half depreciation and amortization of $798 million annualizes near $1.6 billion , so the capex midpoint runs only about $600 million a year above replacement spend. Against that, Lockheed has been asked to triple PAC-3 MSE production and quadruple THAAD output . A 3-to-4x capacity multiple on the two largest interceptor lines for roughly $600 million a year of incremental fixed investment produces an incremental return far above any plausible 8–9% cost of capital — not because Lockheed is unusually efficient, but because it is not putting up most of the capital.
The capital it is putting up sits on the current-asset side. Contract assets — unbilled revenue recognized ahead of billing — consumed $3,037 million of cash in the first half against $2,178 million a year earlier, taking the balance to $16,038 million from $13,001 million . That, not the capex line, is the ramp's real funding requirement, and it is why first-half operating cash flow looked thin.
This is exactly why the second-half cash bar is lower than the market seems to believe. First-half free cash flow was $2,626 million, so the guide implies roughly $4.4–4.6 billion in the second half against $6.1 billion in the second half of 2025 (full-year 2025 free cash flow of $6.91 billion less $805 million in the first half). Management is guiding to second-half cash conversion about a quarter below last year's while growing sales double digits — which is to say the guide already assumes the contract-asset build keeps consuming cash. The October print does not need heroics to clear it.
Fixed-price development, not demand, is what can break this
The bear case here is not the budget and not the order book. It is that Lockheed turns fixed-price development work into large, sudden losses, and nothing in the munitions frameworks changes that. When estimated costs to complete a contract exceed the expected price, the entire loss is booked at once — a reach-forward loss — and Lockheed has taken several.
The evidence is on the balance sheet. Cumulative losses on a single classified Aeronautics program stood at approximately $1.8 billion across both phases at June 28, with $427 million still accrued in other current liabilities . Separately, a competitively bid classified contract at Missiles and Fire Control carries multiple fixed-price options that management expects "would each be at a loss" if performed, and certain options have already been exercised . The Canadian Maritime Helicopter Program carries roughly $585 million of contract assets whose recovery depends on restructuring talks with the Canadian government that may be "prolonged or unsuccessful" . Lockheed holds a 50% interest in United Launch Alliance carried at $617 million and in the second quarter agreed to guarantee ULA borrowings with maximum potential payments of $500 million; both partners expect to provide further support and could incur impairment and operating losses if Vulcan Centaur underperforms . A putative securities class action filed in July 2025 and amended in January 2026 is pending in the Southern District of New York .
The most damning single number is none of those. July's guidance raise added roughly $2 billion of sales at the midpoint for only $50 million of additional segment operating profit — a 2.5% incremental margin against a blended rate near 10.7%. On its face that contradicts the retained-efficiency story. I think the resolution is narrower than it looks: the sales raise came disproportionately from F-35 production volume and radar work, which carry Aeronautics and Rotary margins near 9.4% and 10.5%, while the Missiles and Fire Control profit range moved barely at all because early THAAD booking rates are deliberately conservative until risk-reduction milestones are taken. Management's own quantification of that drag — 20 to 30 basis points — caps the damage at roughly $50 million a year on a segment of this size. More to the point, at the current multiple I am not paying for margin expansion. The thesis needs the revenue to arrive at today's margins, not better ones.
Two other live threads. The continuing resolution signed in early September funds agencies through December 11 at roughly fiscal-2026 levels, restricting new program starts and rate increases; funded lots and already-signed multi-years blunt the near-term effect without eliminating it. And the FY2027 request seeking a $1.5 trillion topline with $760 billion for weapons procurement and modernization is a request, not an appropriation. If it lands materially short, out-year growth compresses toward 3–4% and the duration argument weakens.
Capital allocation is the softest part of the defense. Lockheed repurchased no shares in the quarter ended June 28 against $8,324 million of remaining authorization, having bought $1,250 million in the first half of 2025 . Cash is going to capacity and to Ultra Maritime. That is defensible; it is also a management team declining to buy its own equity after a 22% drawdown.
At about 16 times forward earnings, this is the cheapest of the large defense contractors
On the Financial Modeling Prep consensus panel and today's intraday prices, Lockheed trades at 16.4 times FY2027 estimated EPS of $32.71, against Northrop Grumman at 17.5 times ($533.30 / $30.46), General Dynamics at 19.4 times ($360.60 / $18.59) and RTX at 24.8 times ($196.31 / $7.91) . On enterprise value to forward free cash flow — the right lens for a percentage-of-completion contractor whose vendor-computed EBITDA is contaminated by the pension adjustment and by two reach-forward loss cycles in six quarters — the $140.6 billion enterprise value ($123.84 billion of market capitalization plus $16.7 billion of debt less cash, from the June 28 balance sheet ) is 18.6 times a next-twelve-month free cash flow base of $7.56 billion. The trailing-year price range of $437.25 to $692.00 spans 15.6 to 23.3 times on that same base .
Twenty times is the multiple I will underwrite: $151.3 billion of enterprise value, less $15.5 billion of net debt after a year of cash generation net of dividends, over 228 million diluted shares, gets to $595 — about 11% above the current price, plus the dividend. That sits below the $653 twelve-month consensus target , deliberately, because a 5% ten-year Treasury compresses what any 5-to-7% grower can be worth however good the backlog looks. The downside is not gentle. A PAC-3 slippage plus one new $800 million fixed-price loss would take next-twelve-month free cash flow to roughly $6.4 billion; at 16 times — the multiple the stock carried near its December low — that is about $375, a 30% decline. Lockheed has booked reach-forward losses larger than a full year of plausible margin expansion in each of the last two years, so that is a live scenario rather than a stress test.
What tips the balance is what the market is being asked to pay for. At 16 times forward earnings, this price embeds no margin expansion, no buyback, and a backlog that converts on the market's timetable rather than the contract's. The order book is priced as the same cycle investors have bought and regretted. It is not the same, because the repricing formula that made previous ramps economically pointless for the contractor has been taken out of the programs that matter, with seven years of funded visibility behind them and allied demand still converting — Sweden committed to HIMARS in September . The de-rate has paid for the F-35 mix shift, the fixed-price tail, and a higher discount rate. It has not paid for a business whose fastest-growing segment now carries an 88% larger backlog under terms that let it keep what it saves.
Lockheed Martin is mispriced to the upside, and the mispricing is a contract-structure insight the market has not yet underwritten.
Disclaimer: Tenzing Analytics, Inc. is not a registered investment adviser, broker-dealer, or financial institution, and all content provided by Tenzing Memo is AI-generated for informational purposes only. It does not constitute investment advice, a recommendation, or an offer or solicitation to buy, sell, or hold any security or financial instrument. Content is general in nature, not tailored to any individual's financial situation or objectives, and may contain errors, omissions, or inaccuracies; it should not be relied upon as a primary basis for any investment decision. Any forward-looking statements are subject to risks and uncertainties and may not reflect actual future results. Users should conduct their own research and consult qualified financial, legal, or tax advisors before making investment decisions, and by accessing this content, you acknowledge and accept these limitations.