Newsletter · · Ashutosh Agarwal
Boeing Finds Its Footing as the Missile Math Stops Adding Up - Aerospace & Defense Weekly - Week of July 18, 2026
Aerospace and defense newsletter for the week ending July 18, 2026. The whole sector converged on one theme, magazine depth and manufacturing: the arithmetic of shooting down $30,000 drones with multimillion-dollar interceptors, Boeing's defense-business turnaround, and a wave of drone founders promising million-strong fleets they still have to learn to mass-produce.
Aerospace & Defense Weekly
Week of July 18, 2026: Boeing Finds Its Footing as the Missile Math Stops Adding Up
Week ending Saturday, July 18, 2026
Last week was the NATO summit and the argument about who pays. This week the conversation moved from the politicians to the people who actually build the hardware, and it got a lot more specific. Two words showed up on podcast after podcast: attritable and exquisite. Once you hear the distinction, you can't unhear it, and it explains almost everything happening in the sector right now: the restocking scramble at the big U.S. contractors, the flood of money into drone startups, and the quiet panic about whether America still knows how to run a factory.
A quick note on who's talking, because it matters all the way through. We flag whether a speaker is a company operator (someone actually running the business, building the thing) or a pundit/analyst/investor (someone commentating or allocating capital from the outside). An executive telling you their factory is on schedule is a different kind of evidence than an analyst modeling it from a spreadsheet. This week was unusually rich in operators, the founders and executives were doing most of the talking.
1. The one idea that ties the whole sector together: cheap and disposable vs. expensive and perfect
The clearest single explanation of the moment came on The Bid (July 17), BlackRock's markets podcast, where Rolf Heitmeyer, who heads the industrials team in BlackRock's fundamental equities group (a pundit/investor, not an operator), laid out the framework in plain language.
Start with the two words. "Exquisite" weapons are top-of-the-line: more capable one-for-one, but expensive, complicated, and therefore only buildable in small numbers. A stealth fighter. A Patriot interceptor. "Attritable" weapons are the opposite: "low-cost, mass-produced, almost always unmanned," built from cheap off-the-shelf parts and "engineered to be just good enough, not perfect." They're meant to be used up in huge quantities. They win, as Heitmeyer put it, "by attrition, which is the root of the word." Ukraine is the live example: cheap drones "have totally taken over the battlefield in a way that has made a lot of legacy weapon systems obsolete or close to it."
Here's why investors should care, and it's just arithmetic. The U.S. arsenal is "mostly exquisite." So the country is defending against $50,000 drones by shooting them down with Patriot missiles that cost about $4 million apiece, and often firing more than one to be safe. "It's wildly asymmetric economically and unsustainable," he said. But the bigger problem isn't cost, it's volume. His stat is the one to remember: in the first 16 days of the current Iran war, the U.S. and allies used about 1,800 Patriot missiles, roughly three years of production at current rates. The military has a word for the problem, "magazine depth." The plain-English version is: running out of ammo.
So what happens next? Heitmeyer says the system is "absolutely not capable of keeping up with demand," and fixing it is slow and unglamorous. One bottleneck he singled out: solid rocket motors, which are in short supply partly because the manufacturing methods "haven't changed since World War II." The plan is to lift production of key interceptors like the Patriot by 200% to 300% by the early 2030s, and the fact that it takes until the early 2030s "gives you a sense as to how difficult this problem is."
His investment case rests on two ideas worth internalizing. First, this is a super cycle, not a short cycle: "Let's say in a best-case scenario, all active conflicts ended tomorrow," you would still have to massively ramp production just to refill emptied inventories and build the bigger buffer stocks that modern war requires. Second, defense is a genuine portfolio diversifier because defense budgets are "totally uncorrelated with the macroeconomy," the driver is the threat environment, not GDP. He also flagged the flip side of the coin, the new-entrant wave: U.S. venture capital into defense tech in 2025 was up almost 300% versus five years earlier, and the U.S. military-equipment-and-R&D budget in 2026 was 60% higher than five years ago. His structure call: after decades of consolidation, the industry is now at "a big inflection where we're entering a period of deconsolidation as these new entrants come into the picture."
Why it matters: this is the mental model for everything below. The exquisite incumbents have to ramp; the attritable upstarts have to prove they can scale. The whole sector is a bet on who converts the demand into delivered hardware.
2. The "replacement trade": what actually got fired, and who has to rebuild it
If the Iran war emptied the shelves, someone gets paid to restock them. On WorldWide Markets (July 7), equity-research host Simon Brown (a pundit/analyst) made the most concrete version of this case, and his pick was Lockheed Martin.
The numbers behind it: the U.S. fired roughly 150 THAAD interceptors in 12 days, about half the entire stockpile, plus nearly half its Patriot PAC-3 inventory. Replenishment is already contracted as a seven-year deal worth up to $35 billion, to quadruple THAAD output from 96 to 400 a year, plus 2,000 PAC-3 missiles. But, he stressed, "this rebuild is going to take years rather than quarters." On valuation, Lockheed screened as the cheapest of the primes: a forward price-to-earnings ratio around 17 versus a decade average just under 22, a backlog of about $194 billion, and a 2.5% dividend yield. The catch he was honest about is single-customer risk, Lockheed's one big buyer is the Pentagon, which cuts both ways on pricing power. His other U.S. pick was RTX (which makes the Patriot system itself), notable as "the only one of the primary manufacturers who have raised guidance at this point in time." He described Northrop Grumman (B-21 bomber) and General Dynamics (submarines and Gulfstream jets) as also potentially re-rating.
On Europe, Brown framed it as growth-versus-value: European names have "faster top line… because they are rearming from a very low base," but many "have run hard already." His quick tour of the European primes: Leonardo the cheapest (forward P/E just under 20, order expectations up 15%), Thales the priciest (forward P/E near 23) but with strong earnings growth, BAE Systems guiding revenue up 7–9% on a backlog around £84 billion, and Rheinmetall, the one that "absolutely went crazy," now down about 45% from its highs even on 30–40% revenue growth, trading near 30x earnings. His summary line on the region: "the easy money in Europe has perhaps been made."
The single strangest development of the week made the restocking problem vivid. On Good Revenue News (July 10), host Neeta Bidwai (a pundit/journalist) broke down the president's surprise announcement at the summit: the U.S. is granting Ukraine a license to build its own PAC-3 Patriot interceptors, a weapon that until now only the U.S. made, and one built by Lockheed Martin. The framing to Zelensky, quoted directly, was blunt: "we're going to give a license to you to make Patriots… This way you can't complain that we're not giving you enough. I said, make them yourself." The president added that "we have great power over the companies" and, remarkably, that he had not yet told Lockheed Martin. As Bidwai dryly noted, "I'll bet that was quite the phone call."
The context she assembled is the real story. Ukraine is so short of interceptors it's now firing as few as one per incoming warhead instead of the usual three, which lowers the odds of a hit, because the U.S. "used as many as 1,400 Patriot missiles in the war against Iran." The price gap is the whole problem in one line: Patriots run about $5 million each, THAAD up to $15 million, Tomahawks about $3.5 million, against roughly $30,000 adversary drones. The Pentagon is pushing Lockheed to get the Patriot's cost down toward $1 million, and Lockheed has reportedly reached out to General Motors for help shoring up its supply chain. And the reason the whole system is so brittle: in 1990 there were about 51 prime defense contractors; today there are five, and three of them control 90% of the weapons systems the country needs. Germany, per the Financial Times reporting she cited, is separately pushing to build PAC-3s on its own soil, with a maintenance facility already announced.
For the counter-view on where the restocking money is not a slam dunk, The Wise Investor Show (July 11), from Raymond James's Wise Investor Group (pundit/advisor), took a careful look at Northrop Grumman ahead of its Q2 report (due July 21). The stock was trading about 20% below where it sat right before its April results, which were actually a beat, and had gone from a premium to the S&P 500 earlier this year to roughly a 10% discount. The bull points: its three anchor programs, the Sentinel nuclear missile, the B-21 bomber, and missile defense, together make up over 30% of revenue; the working backlog is around $100 billion; and margins improved to 10.8% last quarter, guided above 11% for the year. The hesitation: Northrop is spending heavily to accelerate B-21 deliveries (a roughly $200 million CapEx bump) and did not reaffirm its 2027 cash-flow guidance, which is why the host cited Goldman Sachs trimming its price target from $740 to $635 and both Goldman and JPMorgan sitting at neutral despite calling it "exposure to the best growing parts of the defense budget." One eye-opener on how these programs balloon: Sentinel's contract has grown from an original ~$78 billion when Northrop won it in 2020 to about $140 billion, with first delivery pushed into the early 2030s.
Why it matters: the restocking trade is real and contracted, but it's a years-long grind priced into names that have already begun to move. The nuance the pundits keep landing on, cheap valuations at Lockheed and RTX, execution and cash-flow questions at Northrop, is where the actual work is.
3. The operator turnaround story of the week: Boeing's defense business
Amid all the outside commentary, the richest operator interview came from inside a prime. On Aviation Week's Check 6 (July 14), Boeing Defense, Space & Security boss Steve Parker (operator) walked through a genuine turnaround, and he was refreshingly direct about both the wins and the scar tissue.
The word he used for the change was "momentum," rooted in "the basics": production stability, engineering excellence, and, his phrase, bidding fixed-price contracts "unashamedly" only when the risk is understood. The proof he pointed to is a run of milestones that had been stuck for years and finally cleared in the last 12 months: the T-7 Red Hawk trainer hit Milestone C (the Air Force's green light to start production); the KC-46 tanker's troubled Remote Visual System 2.0 finished its first flight-test phase; and the MQ-25 Stingray, the Navy's autonomous refueling drone, got its production milestone. His governing line, and a good one: "At the end of the day, the warfighter is the customer, it's not the contract."
He didn't dodge the KC-46's history, "I'll acknowledge KC-46, we've lost money. It's a reach-forward loss program," but said the path out is to "price it appropriately" going forward, and that Boeing is raising the build rate to two tankers a month, helped by capacity freeing up as the 767 freighter line winds down in 2027. (When the interviewer noted the tanker's unit cost has climbed sharply, Parker declined to discuss pricing but acknowledged "a lot of inflationary pressures" against a contract "signed a long time ago.")
The most striking passage was about the F-47, the Air Force's next-generation fighter, the program formerly known as NGAD, which Boeing won. Parker's strategy was to "invest ahead of need," building the St. Louis facility before the contract was even awarded. What's there now, in his description: the roof, exterior panels, and hangar doors are complete; 25,000 tons of U.S. steel across more than a million square feet. He called it "the largest secure manufacturing facility in the United States" and "the world's most advanced all-digital aerospace design and manufacturing system." And he pushed back on anyone who wrote off the older jets: the F-15EX is enjoying a "resurgence," and St. Louis routinely runs three or four production lines at once.
Then the part that ties Boeing directly into the attritable-autonomy shift: the MQ-28 Ghost Bat, an uncrewed "loyal wingman" aircraft developed with the Royal Australian Air Force. Last year's air-to-air missile shot got the headlines, but Parker said the real news was validating the full "kill chain" with no human at the controls: an F/A-18 Super Hornet's radar spotted the target, passed it to an E-7 Wedgetail command aircraft, which told the Ghost Bat to take the shot, "there's nobody on a joystick. It's all fully autonomous." At the recent Valiant Shield exercise, Boeing shipped the aircraft on a C-17, set it up, refueled it off a C-130, and flew it alongside F-35s and F-15EXs. His pitch on maturity: "Nobody else showed up with [collaborative combat aircraft]. We did because it's mature." The airframe has a swappable nose (radar, electronic-warfare jammer, or air-to-air sensor), internal weapons carriage for two AIM-120 missiles or four small-diameter bombs, and it "just finished its stealth characterization."
Why it matters: this is what a prime clawing its way out of a bad decade sounds like from the inside, and a reminder that the incumbents aren't ceding autonomy to the startups. The Ghost Bat is Boeing playing the attritable game with a heritage balance sheet behind it.
4. The disruptors' answer: build the cheap stuff yourselves, at scale
If the primes are ramping the exquisite weapons, a wave of founders spent the week arguing that the future belongs to whoever can mass-produce the cheap ones. Three operator interviews stood out.
The biggest capital-flow signal came from Shield AI. On Founder's Story (July 15), president and co-founder Brandon Tseng (operator, and a former Navy SEAL) confirmed the company "just closed a $2 billion round." His prediction is the one worth writing down: "Every modern military in the world in the next 5 years will declare that they're going to build a 1 million drone army." And the logic that turns it into a business: "To field a million drones, you're not going to have a million pilots. You need high levels of AI and autonomy so that a single person can field… up to thousands of drones at any given moment." Shield AI, he said, is "operating missions every single day in Russia, Ukraine… the Middle East… the Asia Pacific region," and its VBAT drone helped the U.S. Coast Guard "interdict 100,000 pounds of narcotics last year, a record." He was candid about what taking the money means: raising is "one of the sadder days of the journey because you're just signing up for really, really incredible expectations… you are going to deliver top 0.1% results." (Some founder history for texture: he was rejected by all 30 investors he pitched in 2015 before getting his first yeses in 2016.)
For the ground truth on why scale is so hard, Valley of Depth (July 8) with Soren Monroe-Anderson, CEO of Neros (operator, a former professional drone racer), was the most grounded conversation of the week. Neros has raised about $120 million and expects to hit a 100,000-drones-a-year run rate by year-end, with a million-a-year factory as the goal. The scale gap he described is sobering: China is "easily at 100 million drones a year" and DJI alone makes 10–15 million consumer drones, while "America is in the hundreds of thousands… there is no company that will domestically build 100,000 drones this year. Neros is trying to get close." He was withering about the competition's battlefield record: "basically every American drone that went to Ukraine failed, like flat on its face," and on the public "drone dominance" leaderboard, "the majority of the 25 didn't score a passing grade." His most quietly alarming stat: over the last few months, around 90% of casualties on the Russian side are from drones. And a warning for the money pouring in: on the roughly $15 billion of defense-tech funding year-to-date, he expects "a bit of a day of reckoning… there's going to be the people who built real businesses and then there's going to be a lot of folks who didn't" and are "burning tons of venture dollars."
The investor's-eye version came from Bloomberg Tech (July 14), where Joe Lonsdale, managing partner at 8VC and a co-founder of Palantir (investor/pundit), described a genuine culture change inside the Pentagon over "the last 5 or 6 years." Palantir, SpaceX, Anduril, and the autonomous-boat maker Saronic "broke through" by showing solutions "10 times better," and "the admirals learned, the generals learned, and people in procurement said, wait a second, we can't just keep giving these… giant documents written by Raytheon and Lockheed and all the other primes… the full power to block everyone else." But he was clear the incumbents still dominate the money, and explained exactly why: "The primes have these Congressional districts locked down where they're producing things in 49 states, and it's kind of bolted into the [defense bill] that they automatically get the contract even if their stuff sucks because it's a jobs program. It's not about deterring your enemies." He also flagged, live on air, a fresh European data point: Germany's Helsing closed a $1.8 billion round at an $18 billion valuation (investors including Goldman Sachs and JPMorgan), cementing it as Europe's largest defense startup while staying predominantly European-owned.
Why it matters: the startups' entire thesis is that cost and volume beat heritage and scale, that you can build "100 or 1,000 more of these for the same cost," in Lonsdale's words. The open question, which Monroe-Anderson named himself, is which of these venture-funded names survive the coming shakeout.
5. The bottleneck nobody can dodge: America forgot how to run a factory
Cutting across every conversation this week was a single anxiety: even with the budgets and the venture money, can the West actually make the stuff? Three operators tackled it head-on, and it's the theme most under-covered by the market.
The sharpest framing came from Welcome to the Arena (July 8), where Joe Musselman, founder of the defense-focused venture firm BVVC and of a new company called Union (operator/investor), argued that the industrial base itself is the missing piece. His observation from working with his own portfolio: a founder builds one great prototype, wins a $20 million order for 100 units, "and then my question was, okay, well, how do you build 20,000 of those things? Insert awkward pause." His blunt diagnosis: "No one in the world fears the United States turning on their factories anymore because they don't exist." So Union's thesis is to "build factories as a stockpile. We don't just need to stockpile weapons… we have to stockpile factories that build things at scale." The company has raised a blended cap table of nearly $100 million, moved 53 families to Dallas, and claims to be the first privately owned, venture-backed factory in U.S. history to produce the 155mm artillery shell (and its variants, down to tank rounds), deliberately a commercially owned and operated plant, not the traditional government-owned model. Factories two through eight are already being mapped. His guiding line: "The United States should never have a fair fight ever."
The same problem shows up one layer down, in the chips inside the drones. On Drone Radio Show (July 14), Mike Horton, CEO of HYFIX (operator), explained a new regulatory reality: in December 2025, the FCC ruled that drones sold to U.S. consumers and commercial customers need U.S.-origin technology, but most drone radio components come from overseas. HYFIX's answer is a single system-on-a-chip, "a drone on a chip," that combines the radio, GPS/positioning, flight controller, and AI compute that today require several separate parts, most of them foreign. He noted that the only company to have built custom silicon specifically for drones so far is DJI itself, and that repurposing a heavy AI-datacenter or phone chip "isn't the right approach" for something that has to be light and power-efficient. His pushback on the whole "just add it to the banned list" instinct: "we're going to innovate more in the United States and… build drones here that are globally competitive. That's the right answer."
And on the endurance-and-independence frontier, Digital Currents (July 10) featured Fatema Hamdani, CEO of Kraus Hamdani Aerospace (operator), whose bet a decade ago was a 100% U.S.-industrial-base drone with no Chinese parts. Her K1000 aircraft holds the world record for its class at 75-plus hours of continuous flight, is fully electric (so it's near-silent and hard to spot), and needs a footprint of just two to three people to run six to twelve aircraft. Two months ago the company demonstrated charging the aircraft from a laser beam on the ground, which she says puts a nonstop 65-day flight within reach by early 2027. Beyond endurance, she described the drone as "one of the most effective cell towers in the sky": supporting the U.S. Army in a Hawaii exercise, it connected up to 2,500 handheld radios through a single aircraft, solving the problem that a soldier holding a radio on a hilltop "is like a bullseye," and that the first thing an adversary destroys is the cell tower, "that's what they did in Ukraine."
Why it matters: the money is the easy part. Turning it into shells, chips, and airframes runs straight into a manufacturing base the U.S. spent 50 years hollowing out. The names that own real, scaled production capacity, not just clever designs, are the ones that convert this cycle into revenue. This is the same point The Bid made about solid rocket motors stuck on 1940s methods, seen from the factory floor.
6. Europe's capital catches up to its rhetoric
One more thread worth pulling, because it's the flip side of last week's "seven of every eight venture dollars go to America" story. Europe is trying to fix that, and the plumbing is only now being built. On The Brand Called You (July 17), Thorsten, a founding partner of the NATO Innovation Fund (investor/pundit), described just how recent the shift is. Russia's invasion of Ukraine was the catalyst for NATO to stand up a billion-euro-plus multi-sovereign venture fund, backed by NATO's member nations, including newcomers Finland and Sweden, to invest in dual-use deep tech. His detail on how allergic Europe was to this only two years ago is the memorable part: "in 2023, in Europe, no one would do defense investing… it was an absolute no-go area. In fact, the first two months when we started we didn't even have a bank because the bank wouldn't want to bank us." He was firm that it's "not a charity project": the money comes from pension funds and sovereign wealth funds and has to earn a return. Read alongside Helsing's $1.8 billion round and Simon Brown's European primes rallying hard, the picture is of a continent scrambling to build a defense-capital ecosystem from a standing start, and the honest question, as last week's episodes framed it, is whether it can do so fast enough to matter before the next threat window.
The bottom line
Every strand this week wove back to the same knot. The demand is settled; the fight is over magazine depth and manufacturing, whether anyone can produce enough, fast enough, at a cost that isn't insane. The exquisite incumbents (Lockheed, RTX, Northrop, Boeing) are locked into a multi-year restocking super cycle and, in Boeing's case, a real operational turnaround, but the money is largely priced in and the cash-flow questions are real. The attritable upstarts (Shield AI, Neros, and the factory- and chip-builders behind them) are racing to prove they can scale from clever prototypes to millions of units before the venture "day of reckoning" arrives. And the quietest, most important story is the least glamorous: a Western industrial base that forgot how to forge steel, print silicon, and pour rocket motors, now trying to relearn all three at once. As one founder put it, nobody fears a factory that doesn't exist.