In a fast-changing world, our military needs to keep up too

Monday, 27 April 2026

An Iranian army hand-out photo, showing the country’s army commander General Abdolrahim Mousavi, left, with Defence Minister General Aziz Nasirzadeh, with domestically built drones awaiting delivery to the country’s armed forces.

Dr Eric Crampton is chief economist at the New Zealand Initiative and a regular opinion contributor.

OPINION: For a long time, defence procurement could afford to be slow.

That was not ideal. But it was manageable. When militaries were buying large platforms expected to last for decades, haste was not the main virtue. Choosing the wrong frigate, transport aircraft or armoured vehicle could lock in a mistake for a generation. Better to move carefully through the long chain from defence review to supplier comparison, Cabinet approval, contracting, delivery and integration.

That logic made sense in a world where next year’s model was not wildly different from this year’s, and where upgrades and refits could keep equipment current.

Ukrainian servicemen load ammunition into a machine gun mounted on a combat ground drone during a training exercise, in this photo provided by the Ukrainian armed forces.

It makes less sense now.

Last week, Major-General (Retired) John Howard released his first report as a senior fellow with The New Zealand Initiative, the think tank where I work. His report, God Defend New Zealand, argues that if the government wants the 2025 Defence Capability Plan to succeed, it will need to get a lot better at procurement.

Howard’s point is not simply that New Zealand should spend more on defence. It is that spending more will not achieve much if the system turning money into capability is too slow and too rigid for the technologies that now matter most.

As Howard puts it, a procurement system “designed primarily around long-cycle platform acquisition should not automatically govern software-heavy, data-driven and rapidly evolving capabilities in exactly the same way”.

Ukraine demonstrates its Sea Baby drones, capable of inflicting massive damage on enemy naval vessels.

That is a dry sentence describing a very important problem.

Howard argues that small countries should focus where they can have disproportionate effect: stronger information capabilities, better intelligence, more use of open-source data and AI support, and clearer pathways for domestic dual-use technology. New Zealand’s space launch capability also gives it options that many countries our size do not have.

His report is worth reading in full. But one reason his argument matters is the speed at which the underlying technology is changing.

Casey Handmer, an Australian physicist, engineer and entrepreneur based in the United States, recently took aim at Australia’s defence settings. His basic question was whether Australian planning has kept up with the collapse in the cost of autonomous systems.

Anyone who has paid even modest attention to Ukraine will have seen the pattern. Cheap drones can now destroy or disable warships, tanks, and aircraft. The old relationship between cost and combat effect has shifted.

Handmer argues that the “electric stack” behind battery-powered drones and their guidance systems has become radically cheaper. Put less technically, capabilities that once required very expensive specialised systems can now often be assembled at far lower cost.

A 2004 file photo, when the New Zealand Defence Force had just taken delivery of dozens of light-armoured vehicles. The days of militaries and governments being able to run long-winded procurement processes for new weapon systems are over, Eric Crampton writes.

Or, as Handmer later puts it, the kind of one-off specialised raid that Australian commandos pulled off in Japanese-occupied Singapore in 1943 could today be executed by a competent high school robotics team at much higher scale and much lower cost.

For a long time, the Russian army had been feeding high-cost armour to low-cost Ukrainian drone systems. That is not a sustainable exchange rate for a rich country. It is even less sustainable for poorer ones. And the Russian government seems to care little about the human cost it imposes either on Ukrainian civilians or on its own conscripts.

If procurement systems remain geared to slow, infrequent purchases of a few large stable platforms, they will have a hard time keeping up with the fast-moving, lower-cost, higher-volume technologies that are increasingly important. Swarms of cheap autonomous systems effective against tanks, helicopters and ships require different ways of thinking and different procurement systems.

Handmer pushes the argument further than Howard does. He wants Australia to build greater domestic defence production capability and rely less on large foreign-supplied battleships, aircraft and tanks. He sees orbital launch and satellites as must-have sovereign capabilities, warning against having to “live under a sky controlled by others.”

Howard is more measured, but the direction is similar. He argues New Zealand should leverage its existing space capabilities while building stronger domestic dual-use technology pathways.

Read together, their arguments suggest a useful Anzac Day lesson.

Remembering sacrifice matters. So does remembering adaptation. And the value of working together.

Australia and New Zealand should not assume that yesterday’s procurement habits suit today’s strategic environment. Australia has greater scale and deeper industrial capacity. New Zealand has launch capability, useful niches in dual-use technology, and strong reasons to improve intelligence, information systems and interoperability.

On both sides of the Tasman, procurement will need to become faster, more modular and better matched to technologies that improve quickly and can be deployed in large numbers. That will also require better support for the people in uniform who have to understand and use those systems.

Anzac Day is about courage, service and memory. But if Howard and Handmer are right, it should also remind us of something else: in defence, failing to adapt can be its own form of negligence.