🛡️ Defense

The $13-to-$4-Million Ladder: Every Counter-Drone System's Real Cost Per Kill

Farnborough opened with three new counter-drone announcements in a single morning. We built the full cost-per-engagement curve and ran every system against Ukraine's 222-drones-a-night tempo. Only two price tiers survive the math.

By Elena Vasquez · Defense & Security Tech · July 20, 2026 · ☕ 11 min read

Three announcements hit the floor at Farnborough Airshow within the first four hours on Monday. MBDA unveiled a Counter Mass Interceptor designed to give Europe a sovereign answer to drone saturation. Lockheed Martin rolled out the PAC-3 ACE, a new Patriot missile priced at less than half the $4 million PAC-3 MSE. And Lockheed's MORFIUS X-Rotor, a reusable high-power microwave system, was shown for the first time in its production-ready configuration.

All three exist because of the same arithmetic. Iran launched more than 771 ballistic missiles in the opening days of the current Middle East conflict. Russia sent 6,663 drones at Ukraine in April alone, an average of 222 per night. The United States burned through hundreds of Patriot interceptors shooting down Shaheds before anybody with a spreadsheet pointed out the problem: each Patriot costs more than $3 million. Each Shahed costs somewhere between $4,000 and $50,000 to build.

That cost mismatch is the defining operational constraint of modern air defense. Nobody at Farnborough said it plainly. So we did something nobody presenting at Farnborough did today. We put every counter-drone system on the same chart, from the cheapest shot fired in anger to the most expensive, and ran the numbers against real attack tempos.

The Full Cost-Per-Engagement Ladder

Here is what it actually costs to kill one drone, ordered from cheapest to most expensive, using publicly available pricing from government budgets, Congressional testimony, manufacturer disclosures, and contract awards:

Rung to rung, the ladder spans a factor of roughly 300,000. Staggering. Its most expensive engagement costs nearly a million times what the cheapest one does, a price dispersion no other weapons category in history has ever produced for a single tactical function.

The Exchange Ratio: Who's Winning the Spending War?

A strategic question lurks behind the price tags: what each interceptor costs relative to the thing it's destroying. Defense planners call this the exchange ratio, and it determines who goes broke first.

Every system above needs a benchmark target, and that target is the Shahed-136, the most widely deployed one-way attack drone in both theaters. Its cost is genuinely contested. Iran's HESA factories in Isfahan produce them at what one Tehran-based academic estimated as IRR 6 billion, roughly $4,000 at current exchange rates, according to a detailed analysis published by the Phenomenal World. Washington's official range is $20,000 to $50,000, which is also the estimate printed in most Western reporting. America's LUCAS knockoff costs $35,000 per unit despite using more expensive labor and composite materials, which at minimum confirms the Shahed is cheaper than $35,000 to build.

We'll use three price points: $4,000 (lower bound, Iranian production costs), $20,000 (conservative Western estimate), and $50,000 (upper bound).

Read it this way: a ratio above 1:1 means the defender spends less than the attacker. Below 1:1 means the attacker is bleeding the defender's budget faster than its own. At the $20,000 Shahed price point, the most widely cited Western estimate, only three systems put the defender on the right side of the cost curve: DragonFire, Ukrainian interceptor drones, and Merops. Everything from Coyote upward is a losing trade.

Army Secretary Dan Driscoll put it bluntly in his April Congressional testimony about Merops: "That puts us on the right end of the cost curve, and we will make that trade all day long."

He's right. But he's also right at the edge. If Shaheds actually cost $4,000 to build instead of $20,000, even Merops at $15,000 per shot becomes a losing bet because the attacker spends $4,000 while the defender spends nearly four times that amount. Do that 222 times a night and the math is ruinous.

The Nightly Bill: 222 Drones, Seven Price Tags

Russia launched 6,663 drones at Ukraine in April 2026, a record, averaging 222 per night. That's the current operational tempo for a sustained large-scale drone campaign. Iran's Middle East attacks have been spikier but can exceed 100 drones in a single salvo. We'll use 222 as the stress test.

Here is the nightly cost to defend against that tempo, assuming 100% engagement (every drone gets a shot):

Numbers speak for themselves, starting with DragonFire at $2,886 per night, which is almost free by any military standard. Ukrainian interceptor drones at $444,000 a night are painful but manageable for a nation at war. Merops at $3.33 million a night is the ceiling of what a well-funded military can sustain, and the Pentagon's $600 million counter-drone budget would last exactly 180 nights at that rate, or about six months.

That's a losing trade. Everything above Merops is fiscal suicide against drone saturation. Using Coyotes at 222 a night eats $9.7 billion a year. That's more than the entire annual procurement budget of the Royal Navy. It isn't even worth discussing the Patriot column. $888 million per night is what happens when you use a system designed for ballistic missiles against a target that costs less than a used Toyota.

The Capital Trap: DragonFire's Hidden Denominator

If $13 per shot sounds too good to be true, it is, at least partly.

Britain awarded a £316 million ($413 million) contract to MBDA for DragonFire systems to be fitted to four Type 45 destroyers by 2027. Roughly $103 million per ship in capital costs before the first shot is fired. At $13 per shot, DragonFire covers only electricity: about ten seconds of power from a 50kW laser, equivalent to running a household microwave oven.

So the real question: at what engagement count does DragonFire's massive capital investment break even against cheaper kinetic systems?

Consider a single DragonFire installation at $103 million capital. Against Merops at $15,000 per interceptor (zero capital, truck-mounted system), DragonFire breaks even after intercepting roughly 6,867 drones. At Ukraine's 222-per-night tempo, that's 31 nights.

The arithmetic: $103,000,000 ÷ ($15,000 – $13) = 6,867 engagements. At 222 per night = 30.9 nights to break even.

Against Ukrainian interceptors at $2,000 each, the break-even is 51,526 engagements, 232 nights, or about 7.7 months. Still less than a year of sustained operations.

For a fixed military installation expecting prolonged drone attack (an airbase, a port, a forward operating base), the laser pays for itself remarkably fast. DragonFire's constraint isn't cost; it's the physics of a 50kW laser that needs clear atmospheric conditions, takes seconds to burn through a target (during which it can't engage another), and has a range limited by diffraction and weather that renders it useless against ballistic missiles or twenty drones arriving simultaneously.

That's why the military term of art is "layered defense," and it's why every system on this ladder continues to exist despite the price spread.

The Strongest Counterargument: Price Per Kill Isn't the Whole Story

The chart above makes it look obvious: buy DragonFire and Ukrainian drones, skip everything else. But cost per engagement is the wrong metric if the engagement fails.

Probability of kill (Pk) matters at least as much as cost per kill, and it's the number nobody at Farnborough will give you. A $2,000 Ukrainian FPV interceptor that misses 40% of its targets effectively costs $3,333 per kill. A $120,000 Coyote with a 95% Pk costs $126,316 per kill, still more expensive, but the defender doesn't have to worry about the ones that got through.

In saturated environments, the math compounds. If 222 drones attack and your interceptor has a 60% Pk, you need to fire 370 rounds to achieve 222 kills (assuming each target receives one shot). Surplus cost of 148 wasted interceptors at $2,000 each is $296,000. At $15,000 each, it's $2.22 million. At $120,000 each, it's $17.8 million.

Available data is limited but suggestive. Merops has logged more than 4,000 Shahed-type intercepts in Ukraine since mid-2024, according to DefenseScoop. General Cherry's Bullet logged 3,296 confirmed kills in February 2026 alone. Ukraine's overall drone interception rate runs around 90%. But those are aggregate figures. Nobody publishes Pk by system, by target type, by conditions.

Without reliable Pk data, the cost-per-kill chart understates the advantage of more expensive, more reliable systems in scenarios where a single leaker can cause disproportionate damage. One Shahed hitting a runway, a fuel depot, or an ammunition storage facility changes the calculus entirely.

Where This Leaves Western Air Defense

Three realities emerge from those numbers, and they reshape every procurement decision being discussed on the Farnborough floor this week:

An old model is dead. Using PAC-3 or THAAD against drones is not a cost-ineffective choice. It is a logistical impossibility at sustained tempo. America physically cannot produce Patriot interceptors fast enough to keep pace with Iran's Shahed output, let alone Russia's. Lockheed's Patriot line produces roughly 500 PAC-3 MSE missiles per year. Iran and Russia together can produce more drones than that in a single month.

Only $13 to $15,000 survives. At sustained 200+ drone-per-night tempos, only three systems survive the budget: directed-energy weapons (DragonFire class), sub-$3,000 Ukrainian-style interceptors, and Merops-class AI-guided kinetic interceptors at or below $15,000. Everything above that tier is reserved for high-value threats: cruise missiles, ballistic missiles, manned aircraft, where the exchange ratio inverts because the target costs millions.

Production capacity, not unit cost, is the binding constraint. General Cherry told Military Times it can scale to 100,000 drones per month. America bought 13,000 Merops in eight days. Ukraine's interceptor output in the first four months of 2026 already surpassed all of 2025. The question isn't whether cheap interceptors work. It's whether the factories exist. MBDA's new Counter Mass Interceptor won't be available until 2028. Lockheed's PAC-3 ACE needs 36 months. Right now, the systems that can ship at the right price point are being built in Kyiv and Menlo Park, California, not by the traditional defense primes.

Limitations

This analysis uses publicly reported unit costs, which are imprecise. Government contract prices bundle logistics, spares, training, and integration costs into per-unit figures. Merops' $15,000 price may or may not include the ground station, radar, and launch infrastructure. DragonFire's $13 figure covers only the energy cost of firing, not the amortized capital cost of the laser, the ship, or the crew. Shahed cost estimates span a 12× range ($4,000 to $50,000), and the true Iranian production cost is not independently verified.

Probability of kill data by system and by engagement condition is not publicly available for any system in this analysis. Our nightly cost calculations assume one interceptor per drone and 100% Pk, which understates true costs. We also assume a constant attack tempo, when real-world drone campaigns involve clusters, pauses, and mixed salvos of drones and missiles that require different interceptors simultaneously.

Our DragonFire capital break-even calculation uses the total contract value divided by four ships. Actual per-unit installation costs will vary based on integration complexity, support infrastructure, and multi-year maintenance contracts not captured in the initial award.

The Bottom Line

Modern air defense has become a spreadsheet problem where the technology works fine but the arithmetic breaks under sustained pressure.

At Farnborough today, the traditional defense primes announced systems that will be available in 2028 or later at prices that still lose the exchange ratio against $20,000 drones, while meanwhile Eric Schmidt's startup already has a $500 million contract, 4,000 combat kills, and a unit price that puts the defender on the right side of the ledger, and Ukrainian companies nobody had heard of two years ago are producing 100,000 interceptors a month and signing NATO production deals across three continents.

Expensive missiles are not obsolete, and that's not the lesson here because a THAAD that stops a ballistic missile carrying a 1,000-pound warhead from hitting an airbase is worth every dollar of its $12 million price tag. What matters is that the counter-drone layer, the one that has to work 222 times every single night for months on end, cannot cost more than $15,000 per shot. Anything above that line is a system designed for a different war.

What You Can Do

If you're a defense procurement official or uniformed planner: the $15,000 threshold is your budget ceiling for the counter-drone tier. Any system that costs more is a cruise-missile interceptor being miscast in a drone-defense role. Build your layered defense with explicit cost-tier assignments: directed energy and sub-$15K kinetics for drones, Coyote-class for larger UAS and cruise missiles, Patriot and THAAD for ballistic threats only.

If you're a defense investor: the growth curve belongs to the sub-$15,000 tier. Perennial Autonomy's $500 million JIATF 401 contract, the Pentagon's $54–$74 billion FY2027 drone budget (per Secretary Hegseth's testimony), and Ukraine's interceptor export framework all point the same direction. Companies building $2,000 to $15,000 interceptors at scale are the ones with the addressable market. Traditional primes like RTX and Lockheed will capture the premium tier, but the volume play is lower on the ladder.

If you're watching this from a country that imports its air defense: don't wait for 2028. Systems that work at the right price exist now, are combat-proven, and are being sold. Lithuania bought 48 Merops without competitive bidding. Ukraine is opening arms exports. A window to build drone-defense capability before the next crisis closes is measured in months, not the procurement cycles Farnborough vendors are quoting.