Air defence doctrine has spent two years being rewritten around a single number: the cost of stopping one incoming object relative to the cost of sending it. When that ratio runs the wrong way by two orders of magnitude, no inventory is deep enough.

The arithmetic

A long-range interceptor is a sophisticated piece of engineering with a unit cost to match. A mass-produced one-way attack drone is not. Using the first against the second wins the engagement and loses the campaign, because the attacker can replace losses faster and more cheaply than the defender can replenish.

Every serious air defence programme now sorts threats by what they are worth intercepting with.

The emerging layer structure

Four layers are becoming standard, from cheapest effector outward:

  • Electronic attack — jamming navigation or command links. Nearly free per engagement and effective against the least sophisticated threats.
  • Gun-based close-in defence — proximity-fused rounds at a cost per burst measured in hundreds rather than hundreds of thousands.
  • Short-range missiles — for what gets through, at a cost that is acceptable against a genuine threat.
  • High-end interceptors — reserved for ballistic and cruise threats, where the target's value justifies the expenditure.

The engineering challenge is not in any one layer. It is in the fire control that decides, in seconds, which layer handles which track.

Where programmes are struggling

Sensor fusion is the recurring problem. Assigning the cheapest adequate effector requires classifying the threat quickly and confidently, and classification errors are expensive in both directions — a high-end interceptor spent on a decoy, or a gun engagement attempted against something it cannot stop.

Several programmes have found that the fusion software, not the effectors, sets the pace of delivery. That is an uncomfortable finding for procurement systems built to buy hardware.

Magazine depth

The other lesson is unfashionable and simple. Sustained defence is a function of how many rounds are available, which is a function of production capacity rather than inventory on the day.

Countries that have rebuilt production lines are in a materially better position than countries that bought a larger stockpile from a line that has since closed. The stockpile is finite by definition. The line is not.