The Great NATO Arms Race: What If Security Is Not Measured in Tanks?
NATO is preparing to spend 5% of GDP on defence. That may make the alliance harder to attack—but it could also make its societies harder to live in. The smarter question is not “How many weapons can we buy?” It is “What actually makes people safe?”
There is something almost surreal about the current NATO spending race.
Europe spent decades enjoying the post-Cold War peace dividend. Then Russia invaded Ukraine, Donald Trump spent years hammering European allies over defence spending, and suddenly the continent rediscovered an old economic formula:
Fear → weapons → contracts → more weapons → more fear.
Frigates. Submarines. Fighter aircraft. Air-defence systems. Missiles. Drones. Ammunition. Tanks. Radar. Cyberwarfare.
And the invoices keep getting bigger.
At the 2025 Hague summit, NATO members committed themselves to reaching 5% of GDP annually by 2035: at least 3.5% for core defence requirements and up to 1.5% for defence-related infrastructure, resilience, cybersecurity and other security investments.
That distinction matters.
Because the second 1.5% is not necessarily money wasted. Hardened infrastructure, resilient electricity grids, emergency communications, civil preparedness, cybersecurity, hospitals and disaster response can make citizens substantially safer without firing a single shot.
The 3.5% is another matter.
THE QUESTION NOBODY WANTS TO ASK
Why are we spending enormous sums on weapons whose greatest achievement, ideally, is that we never have to use them?
That's the peculiar economics of deterrence.
A successful weapon system is one that sits unused.
The submarine doesn't sink anything.
The fighter doesn't shoot anything down.
The missile doesn't destroy anything.
The tank doesn't cross the border.
The nuclear weapon—hopefully—never detonates.
So perhaps we should stop measuring defence exclusively by the number of spectacular machines governments can purchase.
The purpose of deterrence is not to win a war.
The purpose of deterrence is to make starting one irrational.
And that opens an entirely different debate.
BUT LET'S NOT MAKE THE NUCLEAR MISTAKE
Here is where the provocative argument needs a very important correction.
The tempting conclusion is:
Why buy conventional weapons? Just acquire some nuclear material, and suddenly you're as powerful as the United States and Russia.
No.
That is precisely the sort of shortcut that could turn deterrence into catastrophe.
Nuclear weapons are not a cheap substitute for conventional defence. They require highly sophisticated command-and-control systems, secure delivery systems, intelligence, early warning, physical security, trained personnel and extraordinary safeguards against accident, theft, miscalculation and unauthorized use.
And, critically, more nuclear-armed states do not necessarily mean more security.
Quite the opposite.
SIPRI estimates that the United States and Russia together still possess roughly 83% of the world's military stockpiled nuclear warheads, while all nine nuclear-armed states continued modernizing their arsenals in 2025. SIPRI also warns that the risks of miscalculation and escalation are increasing.
Copy-and-paste nuclear proliferation would therefore be a spectacularly bad version of economic efficiency.
We don't need ten more fingers hovering over the nuclear button.
We need fewer buttons—and much smarter deterrence.
THE PART THAT ACTUALLY WORKS
There is, however, a fascinating lesson hiding inside the nuclear argument.
NATO already recognizes that nuclear deterrence is part of its security architecture.
The alliance explicitly describes nuclear weapons as a core component of deterrence, alongside conventional and missile defence, cyber and space capabilities. NATO's nuclear posture relies substantially on U.S. nuclear forces, alongside British and French nuclear capabilities and allied infrastructure and participation.
In other words:
NATO already has a nuclear deterrent.
The question is whether every NATO country therefore needs to attempt to build every conceivable conventional weapons capability at enormous expense.
Perhaps not.
A smarter model would be:
Nuclear deterrence
Keep the existing strategic deterrent credible.
Conventional deterrence
Maintain enough conventional military capability to prevent an adversary from believing that a limited attack can succeed before nuclear deterrence ever becomes relevant.
Civil resilience
Make societies extraordinarily difficult to cripple.
Economic resilience
Reduce dependence on hostile states for energy, raw materials, technology and critical infrastructure.
Cyber resilience
Make hospitals, banks, communications, water systems and electricity grids difficult to paralyze.
Civil defence
Prepare populations for the increasingly likely non-military disasters of the 21st century.
That is a very different definition of defence.
And potentially a much more useful one.
THE MOST DANGEROUS WEAPON MAY BE A FRAGILE SOCIETY
Imagine two countries.
Country A possesses thousands of sophisticated weapons—but its electricity grid is fragile, its hospitals are overloaded, its drinking-water infrastructure is deteriorating, its communications can be disrupted and its population has little idea what to do during a prolonged emergency.
Country B possesses fewer exquisite weapons but has:
- redundant electricity systems;
- enormous water reserves;
- hardened communications;
- functioning emergency services;
- well-stocked hospitals;
- resilient food systems;
- strong cybersecurity;
- protected ports and railways;
- trained civilians;
- disaster-response capacity;
- diversified energy supplies;
- and an economy capable of continuing to function during prolonged disruption.
Which country is actually harder to defeat?
That's the conversation NATO should be having.
Because modern conflict doesn't necessarily begin with tanks crossing a border.
It can begin with a cyberattack.
A communications blackout.
A satellite disruption.
A pipeline failure.
A hacked water utility.
A disinformation campaign.
A critical infrastructure attack.
A supply-chain breakdown.
A prolonged power outage.
And sometimes the enemy isn't even another country.
It's a wildfire, flood, heatwave, pandemic, earthquake or infrastructure failure.
THE 5% QUESTION IS REALLY A PRIORITIES QUESTION
NATO's 2025 commitment is enormous, but there is an important detail often lost in the headlines.
The alliance doesn't say that every cent of the 5% must become a missile.
Up to 1.5% of GDP can go toward critical infrastructure, network protection, civil preparedness, resilience, innovation and strengthening the defence industrial base.
That is where this debate gets interesting.
Because imagine deliberately making the resilience component huge.
Clean drinking water?
Defence.
A modern hospital?
Defence.
A resilient electrical grid?
Defence.
Emergency communications?
Defence.
Flood protection?
Defence.
Wildfire prevention?
Defence.
Cybersecurity for hospitals?
Defence.
Public emergency shelters?
Defence.
Domestic manufacturing of essential medicines?
Defence.
Food-security reserves?
Defence.
Educated citizens capable of recognizing propaganda and disinformation?
Absolutely defence.
You could call it societal deterrence.
AND THEN THERE'S THE WATER
This may sound almost absurd in a conversation about NATO.
It isn't.
A country can possess the most advanced fighter aircraft on Earth and still have a catastrophic national-security problem if its drinking-water infrastructure is failing.
Clean water is not a luxury.
Neither is clean air.
Neither is functioning healthcare.
Neither is education.
Neither is resilient housing.
Neither is a stable electrical system.
These are the foundations upon which a functioning democracy—and therefore a functioning defence system—rests.
So if governments are going to spend extraordinary amounts of money preparing for the worst, perhaps they should prepare for all of the worst.
Not just the cinematic version.
THE MILITARY-INDUSTRIAL INCENTIVE PROBLEM
There is another uncomfortable issue.
Weapons are not merely defence policy.
They are an industry.
And industries like growth.
NATO itself says the alliance is expanding defence-industrial capacity and production lines as part of the new investment programme.
That makes perfect sense strategically.
But it creates an obvious political problem:
Once billions have been invested, somebody has a powerful incentive to keep buying.
Defence companies employ people.
Factories employ people.
Shareholders want returns.
Politicians want jobs in their constituencies.
Generals want capabilities.
Allies want interoperability.
And governments don't like admitting that the shiny new weapons system they spent €20 billion on may not have been necessary after all.
That's how spending can become self-perpetuating.
The question therefore shouldn't be:
Can we afford another weapons system?
It should be:
What specific threat does this system address, how likely is that threat, what cheaper alternatives exist, and what risks do we create by diverting those resources elsewhere?
That's a much less exciting procurement meeting.
It is also a much more responsible one.
THE MONEY HAS AN OPPORTUNITY COST
Every euro spent on a weapon is a euro that cannot simultaneously be spent somewhere else.
That's not pacifism.
It's arithmetic.
A government facing ageing infrastructure, housing shortages, expensive healthcare, climate disasters, stagnant productivity and deteriorating public services doesn't possess an infinite wallet.
And NATO countries don't all have Germany's fiscal room.
The defence surge is already happening: NATO reported that European Allies and Canada increased combined defence expenditure by nearly 20% in real terms in 2025, while five Allies were already expected to meet the 3.5% core-defence benchmark in 2026.
The real test will come later.
Because buying the weapon is only the beginning.
Weapons require:
maintenance.
personnel.
training.
ammunition.
infrastructure.
upgrades.
replacement.
fuel.
logistics.
more weapons.
The price tag on the brochure is not the price of owning the capability for 30 years.
WHAT IF WE BUILT A DIFFERENT KIND OF ARSENAL?
Here's the radical proposition:
Build societies that are extraordinarily difficult to intimidate, disrupt or break.
Spend aggressively on:
💧 Water
Modernize drinking-water and wastewater infrastructure before drought, contamination or infrastructure failure turns into a national emergency.
🌬️ Clean air
Reduce pollution and protect public health instead of treating healthcare as an afterthought.
🏥 Healthcare
Create hospitals and emergency systems capable of handling pandemics, heatwaves, mass casualties and infrastructure disruptions.
⚡ Energy
Build redundant grids, storage and distributed generation so one failure doesn't become a national crisis.
🎓 Education
Teach science, statistics, media literacy, cybersecurity and critical thinking.
🧑🚒 Civil protection
Train citizens and communities for blackouts, floods, wildfires, earthquakes and other emergencies.
🌾 Food security
Make supply chains less dependent on fragile international chokepoints.
💻 Cybersecurity
Protect hospitals, water utilities, power networks, banks and communications.
🏗️ Infrastructure
Build bridges, railways, ports and buildings capable of surviving the shocks we already know are coming.
🧠 Democratic resilience
Protect elections and public institutions from manipulation, corruption and information warfare.
Now there's an arsenal.
And almost none of it requires a missile.
DETERRENCE WITHOUT DELUSION
None of this means NATO should disarm.
That would be naïve.
Russia's invasion of Ukraine demonstrated brutally that conventional military force remains real. NATO's eastern members have legitimate security concerns. Deterrence matters.
But deterrence does not have to mean:
"Buy everything."
It should mean:
"Build the minimum credible military capability required to make aggression irrational—and maximize the resilience that makes aggression, coercion and disruption ineffective."
There is a huge difference.
NATO itself describes its deterrence posture as a combination of nuclear, conventional and missile defence, complemented by space and cyber capabilities.
That gives the alliance enormous room to think beyond the traditional arms race.
THE REAL SECURITY FORMULA
Perhaps the equation for 21st-century security should look something like this:
Deterrence + Defence + Resilience + Public Health + Infrastructure + Education + Energy Security + Democratic Stability
Not:
Missiles + Missiles + Missiles + Submarines + More Missiles.
Because a nation doesn't become powerful merely because it can destroy another nation.
A nation becomes powerful when it is extremely difficult to destroy.
That's a very different ambition.
AND THAT IS WHERE THE MONEY MATTERS
If governments genuinely believe they must spend substantially more to protect their citizens, fine.
But citizens should demand something in return:
Show us the threat assessment.
Show us the alternatives.
Show us the lifetime cost.
Show us what the system actually protects.
Show us what we're giving up to pay for it.
And then ask the most uncomfortable question of all:
Could the same money make our country safer in another way?
Sometimes the answer will be no.
Buy the frigate.
Buy the air-defence system.
Buy the ammunition.
Maintain the deterrent.
But sometimes the answer will be yes.
And when it is, we should have the courage to say so.
Because there is nothing particularly patriotic about purchasing a $20-billion weapons system while allowing drinking-water infrastructure, hospitals, schools, electrical grids and emergency services to deteriorate.
A weapon protects a country from a particular threat.
A resilient society protects itself from almost everything.
THE ADAPTATION-GUIDE BOTTOM LINE
The 20th century taught us that military weakness can invite aggression.
The 21st century is teaching us something else:
Fragility is itself a security risk.
Russia, cyberwarfare, pandemics, extreme heat, wildfires, floods, grid failures, supply-chain disruptions and misinformation don't politely wait their turn.
So yes—rearm if necessary.
But don't confuse more expensive weapons with more security.
Keep a credible deterrent.
Maintain enough conventional force to make aggression prohibitively difficult.
Cooperate with allies.
And then spend the next dollar where it produces the greatest increase in actual resilience.
Clean water.
Clean air.
Healthcare.
Education.
Energy security.
Infrastructure.
Cybersecurity.
Emergency preparedness.
A population capable of surviving a crisis without society collapsing.
That isn't soft security.
That is national security.
And perhaps the ultimate weapon of a democracy isn't the ability to destroy its enemy.
It's the ability to remain standing when everything around it starts falling apart.
One final reality check
The original argument that NATO's 3.5% target is unlikely to be reached needs updating for 2026: NATO now says five Allies are already expected to meet the 3.5% core-defence benchmark in 2026, and the formal deadline for the full 5% commitment is 2035, with progress to be reviewed in 2029.
So the more interesting question isn't whether NATO will spend more.
It almost certainly will.
The question is whether NATO's governments will spend that money intelligently.
And that is where citizens should be paying attention.
NATO: Defence investment and the 5% commitment · SIPRI: World nuclear forces
Possible killer headline:
THE MOST EXPENSIVE WAY TO FEEL SAFE
