The Cost-Exchange Trap: How Cheap Drones Broke the Arithmetic of Air Defence

For half a century, the economics of air defence ran in the defender's favour. A surface-to-air missile costing a few hundred thousand dollars could destroy a combat aircraft costing tens of millions. The arithmetic rewarded the side doing the shooting down. Cheap drones have inverted that relationship, and the inversion is now reshaping procurement budgets, industrial policy and battlefield doctrine across every major military.

The most-quoted figures capture the shock. A first-person-view drone costing a few hundred dollars can disable an armoured vehicle worth twenty million. At the extreme, Iranian drones reportedly costing around $30,000 have disabled radar systems worth upwards of a billion — a cost-exchange ratio in excess of 30,000 to one. Dedrone's 2025 operational dataset, drawn from theatres across Europe, the Middle East and Asia, suggests the imbalance is widening rather than narrowing, with do-it-yourself drone detections rising several-fold year on year as more operators assemble custom platforms from commercial components.

These ratios are real, and they are alarming. But treated as the headline metric, they mislead. The cost-exchange ratio is a stock measure — a snapshot of one engagement. War is a flow.

The ratio is the wrong number

A favourable per-unit cost is strategically meaningless if you cannot produce at the rate your opponent consumes your inventory. This is the point that the viral statistics obscure, and it is best understood not as a pricing question but as a reserve-adequacy question — the kind an actuary or a risk engineer would recognise immediately.

Consider the consumption rates. Russia's winter 2025–2026 campaign of Shahed-type loitering munitions deployed roughly 19,000 drones over ninety days, against monthly production now estimated at around 3,000 units. Ukrainian layered defence performs well by any conventional standard, intercepting somewhere between 83 and 88.5 per cent of incoming threats across documented engagements; Gulf states intercepted 93 to 94 per cent of Iran's late-February saturation attack on multiple states in the region. And yet both defenders still lose on economics. Cheap mass defeats expensive defence even when the defender stops the overwhelming majority of what is thrown at it.

The binding constraint, in other words, is unit cost multiplied by units per month — on both sides simultaneously — and the attacker sets the tempo. Coalition forces in the Middle East have reported expending interceptors faster than industry can replace them. That is not a logistics hiccup to be solved with a production ramp. It is a structural mismatch between the economics of offence and the economics of defence.

The saturation trap

The most elegant part of the attacker's playbook is the deliberate mixing of cheap drones with expensive missiles in a single salvo. This is not crude volume; it is a calculated attack on the defender's inventory and budget at once.

Faced with a mixed salvo, the defender confronts a no-win allocation problem. Fire a three-to-four-million-dollar interceptor at a twenty-thousand-dollar drone and you lose on the wallet. Let it leak through to preserve your expensive interceptors for the genuine high-end threat, and you may lose the asset. The attacker chooses the mix; the defender must carry reserves against every tier at once and expend across all of them.

Framed in actuarial terms, this is a forced-consumption attack on a reserve held against a frequency-severity distribution in which the counterparty controls both the frequency and the severity mix. No insurer would willingly write that risk. The defender is compelled into a hedging programme whose expected value is negative by construction, and the only levers available are to change the cost of the hedge or to change the distribution at its source.

Choosing between cost curves

The escape routes from the trap are not really a menu of weapons. They are a choice between cost structures — different shapes of the curve relating fixed investment to marginal cost per engagement.

Cheap kinetic interceptor drones attack the ratio directly, flipping it back toward the defender. Ukrainian-built interceptors such as the Sting cost in the region of $2,100 and have been credited with thousands of Shahed shootdowns; Russia's handheld Yolka is reported at around $500; NATO buyers are paying roughly $15,000 a shot for systems such as Merops, which has logged more than a thousand intercepts since combat testing began in 2024. One AI-guided interceptor system has reportedly achieved a 13-to-1 cost-exchange ratio across nearly two thousand confirmed kills — concrete proof that the arithmetic can be reversed when the defensive munition is itself cheap and produced at scale.

Directed energy is the only candidate for a genuinely structural fix, because it moves defence off the consumable model altogether. The cost is front-loaded into the platform and the power supply; the marginal cost per engagement then collapses toward the electricity bill. Israel's Iron Beam is estimated at a few dollars per shot, the United Kingdom's DragonFire at around ten pounds. Against slow, low-flying targets the economics are transformative — an effectively unlimited magazine limited only by fuel for the generator. But lasers carry hard capability ceilings: atmospheric attenuation, weather, finite range, large power-generation demands, and an inability to engage fast or ballistic targets. They complement the kinetic layers; they do not replace them.

Electronic warfare offers, in principle, infinite rounds — a jammer is limited only by its battery, not its ammunition. Its weakness is that the threat is evolving out from under it. Fibre-optic FPV drones carry no radio link to jam, and fully autonomous, AI-guided munitions need no operator connection at all.

The strategic implication is that there is no single correct answer, only an optimal layering of cost curves matched to the threat distribution: guns and EW and lasers for the cheap, slow, high-volume tier; exquisite interceptors reserved for the genuine high-end threat they were designed for.

Why the equilibrium will not hold

It would be comforting to conclude that cheap interceptors and directed energy will restore balance and the matter will settle. They will not, because the attacker's cost curve is moving too.

Fibre-optic guidance has already neutralised much of the electronic-warfare layer. More consequentially, AI terminal guidance has reportedly lifted FPV mission-success rates from around 15 per cent to roughly 60 per cent. That is a multiplication of effective threat-per-dollar with no change in unit cost whatsoever — the same money buying three or four times the lethality. Both sides are now racing along their own curves, and any equilibrium is at best a temporary truce between two adaptation cycles.

This co-evolution is why the counter-UAS market is expanding so violently — most forecasts put it on a path from somewhere around four-and-a-half billion dollars in 2025 toward roughly fourteen-and-a-half billion by 2030 — and why procurement language has shifted decisively. The United States Army's Project Flytrap in late 2025 was described by officials, with deliberate bluntness, as the beginning of the end of firing four-million-dollar missiles at twenty-thousand-dollar drones.

The actual prize

The deepest lesson is that this contest rewards industrial mass and unit-cost discipline over exquisite capability. Modern aerial warfare has become less a question of technological superiority than of economic sustainability — of who can hold the favourable side of the cost-exchange at scale, over time, without exhausting either the magazine or the budget. That reframing favours the side that can manufacture cheaply and abundantly, which is precisely why mass-production initiatives have moved to the centre of defence planning.

And it points to where the leverage truly sits. In pure unit-economics terms, the drone's cost advantage is dispersed across thousands of cheap, individually disposable munitions — which is exactly what makes it so hard to defeat at the point of attack. That advantage concentrates, and therefore reverses, at one place only: the production node. Striking the factory rather than the munition is the single move that turns the attacker's own arithmetic against them. It is the one point on the curve where the defender, not the attacker, holds the favourable ratio.

For defence planners — and for anyone who builds reserves against an adversarial, adaptive risk — the conclusion is uncomfortable but clear. You cannot price your way out of a forced-consumption attack on your inventory. You can only change the cost of your hedge, or attack the distribution at its source.