Washington Tightened Its Drone Supply-Chain Rules Last Week. Then It Admitted the Capacity Isn't There.

Three things happened in Washington in a single week, and read together they describe a problem the U.S. defense establishment has not yet solved.

On July 20, President Trump signed an executive order making it materially harder for defense contractors to obtain waivers to buy critical components and materials from China and other prohibited suppliers. Under the order, a contractor can no longer simply argue that a Chinese supplier is the cheapest or easiest option. It must prove it searched for alternatives, disclose where its materials originate, and submit a plan to move off prohibited sources or risk losing the contract. Default waivers end on January 1, 2027. The order also directs the Pentagon to require contractors to map their supply chains component by component, from raw material to finished system. As White House trade adviser Peter Navarro put it, the era of "we tried nothing and we're out of options" is over.

Three days later, on July 23, the Pentagon's Drone Dominance Program published version two of its supply-chain framework — the rulebook for a $1.1 billion commitment to buy more than 200,000 small attack drones across four phases. The revision was not cosmetic. Version one, from April, aimed to purge Chinese content "at every tier" of the supply chain. Version two keeps that and adds a new sentence naming the destination: the ultimate goal is now a "wholly domestic" small-drone supply chain, to be reached "as soon as is practicable." The July revision also reached deliberately toward the hardest parts of the problem — adding final assembly and semiconductors as newly tracked, phased line items, and extending the schedule with a new longest-horizon preference tier. The direction of every edit was the same: harder, and more domestic. We explained the changes component by component in a companion piece to this article. 

The framework is more revealing than the "wholly domestic" phrase alone, because it shows the path to the destination, and when. It sorts every major component into three sourcing tiers — allied (which it calls non-covered-country, or NCC), nearshored (defined as the Western Hemisphere), and U.S. — and then ratchets the requirement upward phase by phase, from August 2026 through 2027 and beyond. The near-term phases mostly demand allied sourcing: get China out. The later phases increasingly prefer U.S. or Western-Hemisphere production: move past allies too. Two different clocks are running inside the same document, and the gap between them is the whole story for Taiwan.

Then, on July 27, the official running that program said the quiet part aloud. Travis Metz, who leads the Pentagon's drone-industrial push, acknowledged in an interview that the United States remains in the earliest and hardest phase of building domestic capacity. The Pentagon's $1.1 billion program will have ordered just under 200,000 drones cumulatively by February against Ukrainian output that runs into the millions annually. He conceded that the vast majority of drones bought in the program's earlier phase likely contained Chinese motors, and that harder components, semiconductors especially, remain a bigger problem still.

Put the three together and the picture is clear: the rules now demand what the capacity cannot yet supply.

The gap is the opportunity

This is the opening for Taiwan's suppliers. U.S. policy has decided, firmly and on a deadline, that Chinese content must come out of the defense drone supply chain. And the first deadline is close. Under the framework, the Phase 2 minimum — the floor to compete for orders, with non-compliance disqualifying — requires allied, non-covered-country assembly for the core components as of August 2026. For motors and battery packs specifically, the framework demands non-Chinese assembly ahead of any statute that requires it; the motor requirement, the document notes, is a program rule with no current legal mandate behind it. This is aimed at Group 1 one-way-attack drones — the small, high-volume, low-cost systems Ukraine has made decisive — which is precisely the category where the sourcing problem is hardest and the volume largest.

But the domestic industry that is supposed to supply all this does not yet exist at the required scale. Motors, batteries, flight controllers, datalinks — the components the framework scrutinizes most closely — cannot all be sourced domestically today, and the official in charge says so plainly. Someone has to fill the space between the mandate and the manufacturing base. And for the near-term phases, the framework does not ask for domestic; it asks for non-Chinese. An allied supplier is not a fallback there — it is what the rule explicitly calls for.

Taiwan is well positioned for that role: it manufactures across most of the relevant categories, it is unambiguously non-Chinese, and its export capacity is already proven — the same components are already flowing in volume into the Europe-Ukraine corridor. For a U.S. contractor now required to disclose origins and demonstrate a move away from Chinese suppliers, a Taiwanese source is a defensible answer to a question the contractor is now forced to answer.

But the opportunity has a shape, and the shape matters more than the opening.

The window is real, and it is timed

The same framework that creates the opening also announces its intended end, and it does so on a visible schedule. The two clocks are the key to reading it. The first clock — get China out — is set for now: allied assembly is the disqualifying minimum from August 2026, and the requirements only tighten from there. The second clock — become domestic — runs much longer, and this is where Taiwan's position is exposed. In the later phases, the framework's preferred tier, the one that earns scoring weight and signals where orders will flow, shifts to "U.S. or nearshored" for nearly every core component. Nearshored means the Western Hemisphere. Taiwan is neither. A Taiwanese company that sells components across the Pacific satisfies the near-term rule handsomely and sits outside the long-term preference entirely.

That is the shape of the opportunity: Taiwan lives in the space between the two clocks. Wide enough to matter — the China-out requirement is immediate and the demand is urgent — but bounded, because the same document that rewards allied sourcing today prefers domestic sourcing tomorrow. Taiwanese suppliers who treat the U.S. market as a permanent export destination are misreading the opportunity. Those who treat it as a time-boxed opening to be captured quickly, and converted into something more durable, are reading it correctly.

There is an honest competitive caveat here too. Taiwan's obstacle in this market has never been only regulation; it is Chinese price and scale. Where a U.S. buyer can find a compliant domestic option, it now will. Where it cannot, the choice is between an allied supplier and a waiver that is now much harder to obtain. Taiwan wins in the second case, and its suppliers should be pricing and positioning for exactly that.

How real is "wholly domestic"?

There is a prior question underneath all of this, and it deserves a skeptical answer: is "wholly domestic" a genuine destination, or the latest version of a promise Washington has made, and missed, for more than a decade?

The skepticism is earned. The United States has been announcing the reshoring of critical manufacturing since well before drones were a defense priority, with results that have consistently trailed the rhetoric. The clearest recent parallel is semiconductors: years of policy attention and tens of billions in subsidies have produced real domestic capacity, but slower, costlier, and less complete than promised — and it is no accident that Metz named semiconductors as the drone program's hardest problem. Rare-earth magnets tell a similar story: genuine investment, multi-year timelines, and a persistent dependence on Chinese processing that money alone has not dissolved. The obstacles that blunted those efforts — higher domestic costs, thin pools of skilled labor, supply chains that run four and five tiers deep into materials no one makes domestically, and a Chinese competitor that has flooded the market with cheap, capable product — are precisely the obstacles the drone push now faces. None of them yields to an executive order.

A tension inside the policy itself pushes the same way. The government that wants a wholly domestic supply chain also wants hundreds of thousands of drones, quickly, because Ukraine has shown that drones are now decisive and the United States is behind. Those two goals collide the moment domestic capacity falls short — which, by the program's own admission, is now. When "buy domestic" and "buy quickly" conflict, the historical pattern is that urgency wins: waivers are granted, deadlines slip, and "wholly domestic" softens into "mostly domestic, eventually." A framework can state a destination. It cannot manufacture the capacity to reach it on schedule.

The framework itself half-concedes the difficulty. It acknowledges that non-Chinese components will cost more, "at least initially," and frames the whole exercise as a chicken-and-egg problem — costs stay high until volume arrives, and volume will not arrive until buyers commit — that its $1.1 billion demand signal is meant to break. That is a candid description of a hard economic climb, not a switch waiting to be flipped. It is also an implicit admission that on price, at the outset, the compliant supply chain loses to the Chinese one.

The opposite error is just as easy, though, and Taiwan's suppliers should avoid it too: concluding that this is merely talk. It is not. What is different this time is enforcement — waivers that actually expire, contract loss as a penalty, mandatory component-level supply-chain mapping, disqualifying minimums with dates, and real capital already committed. Earlier reshoring pushes were largely incentive and exhortation; this one has teeth and a schedule. The direction of travel is unambiguous even if the arrival time is not.

The honest synthesis is neither "wholly domestic is imminent" nor "it is only rhetoric." It is a real, enforced, well-funded direction of travel that will nonetheless take far longer, and arrive far less completely, than the language suggests — a long and contested transition rather than a clean switch. For Taiwan that is the crucial calibration, because it sets the width of the window. The transition being slow and incomplete means the allied-supplier role lasts longer than the framework's language implies. The transition being real and enforced means the window is genuinely narrowing, and will not reverse. It is wider than the rhetoric and closing anyway.

The durable position: co-production

Which raises the more strategic question. If the U.S. endgame is domestic production — even a slow, incomplete version of it — can a Taiwanese supplier secure a lasting position not by exporting into that market, but by producing inside it?

The logic does not depend on "wholly domestic" ever being fully reached. It depends only on the domestic-content preference being real and enforced, which the phase table proves it is. And the framework closes off the easier half-measure: the preferred tier is "U.S. or nearshored," and nearshored is defined as the Western Hemisphere. An allied supplier cannot reach that tier by remaining an allied supplier, however impeccable its non-Chinese credentials — only production in the United States (or the Western Hemisphere) crosses the line. For a Taiwanese firm, being the best non-covered-country source in the world still leaves it in the tier the program is designed to move past.

That is what makes U.S. production presence the durable play rather than a nice-to-have. A Taiwanese firm that manufactures in the United States — through a subsidiary, a joint venture, a licensing arrangement, or a co-production agreement with a U.S. partner — stops being a foreign supplier the policy wants to design out and becomes a domestic producer the policy is built to favor. It gains access to the preferred-sourcing tiers and the advanced market commitment the framework reserves for compliant domestic production. The temporary gap-filler becomes a permanent participant.

This is not a novel move in defense. Foreign primes have long established U.S. operations precisely to qualify for U.S. procurement; the drone-component layer is simply where the same logic is now arriving. For a Taiwanese company with a genuinely differentiated product — a proprietary datalink waveform, a hard-to-replicate manufacturing capability — a U.S. production footprint converts a technological edge into a durable market position that a pure export relationship can never provide.

But the co-production path is narrower than it looks, and it is honest to say so.

It is capital-intensive, slow, and operationally demanding — a profile that suits larger, well-capitalized Taiwanese firms far better than the small and mid-sized manufacturers who make up most of the ecosystem. It requires sharing something worth protecting: licensing or a joint venture means placing intellectual property in partner hands. And even a U.S.-producing, foreign-owned firm can face limits in the most sensitive programs, where foreign ownership, control, and influence reviews still apply — "wholly domestic," pressed to its conclusion, can shade from American-made toward American-owned.

There is also a tension that sits uncomfortably with Taiwan's own strategy. Taipei is spending NT$44.2 billion to build a drone industry and an industrial cluster at home. A Taiwanese firm that answers the U.S. domestic-content signal by moving production to America is, in a small way, moving in the opposite direction from its own government's industrial policy — a quiet leakage of exactly the capacity the subsidies are meant to build. For the firm, it may be the rational long-term play. For the national project, it is a cost. Both can be true at once, and any Taiwanese company weighing a U.S. footprint should weigh that friction honestly rather than pretend it away.

A necessary caveat: the demand has to be real first

None of this means a Taiwanese supplier should rush to spend against it. The honest reality on the ground tempers the urgency: after a year of executive orders, frameworks, and a bill written specifically to bring Taiwan into the Blue UAS supply chain, only one Taiwanese firm has actually cleared the certification that unlocks this market, with more still in progress. That is not a sign that Taiwan's suppliers are asleep. It is a sign they are doing the math.

Certification is not free, and for a smaller manufacturer the cost is far from trivial. Most importantly, it buys a credential, not a contract. On the other side of that ledger, the demand a firm would be spending against is still, by the plainest available measure, unproven: the Blue Skies for Taiwan Act, the legislation meant to accelerate Taiwan's entry, has not passed and even if enacted, it would commission a working group to study the question rather than place orders. The DDP's cumulative purchases, meanwhile, remain small against the millions of units Ukraine turns out. A firm that declines to pay a real, upfront cost against demand it cannot yet see is not being timid. It is being cautious, and it is reading the market accurately.

So the practical counsel is not "move now," but it is not "wait and see" either. Speculative investment — paying to certify, or building U.S. production, against no named buyer — is a bet only a few well-capitalized firms should make today, and most are right to wait. But two kinds of preparation cost little, commit nothing, and are where the urgency genuinely applies.

The first is readiness: knowing your own bill of materials, mapping where every component and subcomponent is made, understanding which non-covered-country requirements you already meet. That work lets a firm move in weeks rather than quarters when a concrete order appears. Certification should follow demonstrated demand, not precede it, but the supplier who has done the homework is the one who converts that demand when it comes, while others are still starting from scratch.

The second is visibility, and it is the part Taiwanese manufacturers most often neglect. The demand created by these rules will not find its way to a supplier who cannot be found. A U.S. contractor now forced to disclose origins and locate a non-Chinese source is actively looking but it can only choose from the suppliers it knows exist. A Taiwanese firm that is invisible to that buyer, however capable, is not in the running. Making yourself discoverable in the channels where non-Chinese sourcing decisions are actually made, in the language and terms those buyers use — directly determines whether the opportunity ever reaches you.

It is true that some of this matchmaking happens at the state level, and it is worth being clear-eyed about who it reaches. When Anduril signed a drone-development MOU in Taipei in June, it was witnessed by the Industrial Development Administration, and the IDA had already helped more than ten Taiwanese companies secure trial or production orders from the firm. That is real, and valuable, but it is the exception that proves the rule. A marquee U.S. prime gets a government agency to convene a shortlist for it; a mid-sized U.S. integrator, a European OEM, or any of the hundreds of buyers who are not Palmer Luckey do not. And the suppliers who benefit are the handful already on the agency's radar. For every firm on that shortlist, there are many more capable manufacturers the channel never surfaces, and many more buyers it never serves. Government facilitation handles the headline relationships; the long tail of demand — which is most of it — still has to find its own way to a supplier, and a supplier still has to make itself findable.

There is a subtle but decisive point buried here. The demand is not simply going to show up. It is being created by policy, but it does not flow automatically to Taiwan; it has to be captured, and often generated. The suppliers who understand what is happening — which rules come into effect when, which components are most in demand, which buyers are under the most pressure to re-source — are far better positioned not just to respond to inquiries, but to go and create them: to approach the right buyer, at the right moment, with the right thing to say. Understanding the landscape is not passive knowledge. It is the raw material of demand generation, and it is the difference between waiting to be found and going out to be chosen.

The conclusion

The sequence is what matters. The near-term opening is genuine, but the firms best placed to seize it are not necessarily those who spend earliest on faith — they are the ones ready to move fastest when the demand becomes concrete. The policy that creates this opening is real, enforced, and pointed in one direction; the demand that would justify investing against it is still, for now, mostly promise. Prudent suppliers will accept both facts at once: prepare quietly, and be ready to commit the moment a real buyer makes the opportunity tangible. The durable position beyond that — for the subset of firms with the capital and the differentiated product to reach for it — may be co-production inside the United States, entered with clear eyes about its cost, its control questions, and its tension with Taiwan's ambitions at home.

Europe remains the market where the relationship is structural rather than timed, and where the volume already is. The United States is where the margin could eventually concentrate, but capturing it durably will ask more of Taiwan's suppliers than shipping a container. And it may ask some of them to decide whether they intend to sell to the American drone industry, or to become part of it.

Buyers sourcing non-Chinese drone components from Taiwan can submit a sourcing request. There is no cost and we step out after the introduction is made. 

Photo credit: Department of War — Air Force Staff Sgt. Michael A. Richmond