It’s all over the press this week: Trump wants all NATO members to spend 4% of their GDP on defence – a doubling of the previously-agreed 2% target (which many countries don’t meet). To me this seems pretty wrong, for two reasons, but not the ones you might imagine. These aren’t new arguments, but since there’s a grain of truth in Trump’s criticism (NATO, for all the money spent on it, has demonstrably failed to deter agression on its borders), let’s unpack this.

Firstly, it is entirely legitimate to ask whether there is a level of defence spending that ceases to be sustainable in any society. To think about this we need to stop thinking of national budgets like ‘household expenses’ (pub, Sky subscription, holidays), and think of national spending more as a business-type investment. This is because most governmental spending is either meeting our national ‘running costs’ (recycling rubbish, paying the police to keep law and order), or a genuine investment (education improves skills and so wellbeing and productivity; health spending keeps more workers active longer; new railways get people to work faster and more reliably, etc.)

Defence spending is different, since it is essentially a form of insurance, e.g. an expense we incur not because we want to use tanks, ships, bombers on a daily basis, but because we hope that by having them, we can (let’s be honest) intimidate other nations into helping meet our foreign policy goals (or at least, not blocking them). Deterrent nuclear weapon systems are the ultimate example of this since – unlike an amphibious carrier, say, or a transport aircraft – they can’t be used for humanitarian work, or even counter-terror operations.

So is 4% of GDP sustainable for all NATO countries? Arguably not, or at least, not as equally. NATO is now a large organisation, and includes:

• small post-communist states like Bulgaria and Slovakia;
• larger mid-table nations such as Spain and the Netherlands;
• G7 members like the UK, France, Spain and Germany;
• the USA, a global hyperpower with self-imposed military commitments in every part of the world, currently engaged in the start of a (hopefully slow) relative decline in power compared to China; and outliers like
• Turkey (a massive country with one of the lowest per-capita incomes in Europe, but the only NATO state with a shooting war within its own borders, at least intermittently, in Kurdistan)
• and Canada (a geographically sprawling, rich nation with – let’s be honest – very few real direct foreign-policy competitors).

Even if we set aside the very differing nature of each countries’ political economy (Germany’s economy continues to motor along with export prices kept low by the Euro; Bulgaria is in dire need of investment in all sectors; the UK has stagnated for a decade with chronic underproductivity due to expensive housing stock and transport costs; the US continues to benefit from the dollar’s status as a global currency), there are clearly very different foreign-policy goals and motivations here.

Canada, for instance, could probably disarm their offensive military completely, retaining only a very small gendarmerie / expeditionary force – safe in the knowledge that the odds of an unprovoked US annexation occurring any time soon are minimal (any ratcheting of tension between the two neighbours would likely take decades to result in all-out war). On the other hand, the Baltic states not only have fresh memories of Soviet rule 28 years ago – many still remember the Nazi invasion of 1940, while the Great War and following Russian Revolution are also barely out of living memory. They are also effectively encircled by an increasingly hostile Russian autocracy, have no large territory into which they can withdraw to defend in depth, and have tiny populations – the total population of Latvia, Estonia and Lithuania, 6.1 million people, is barely half that of Moscow alone (11.8m). For NATO, successfully defending any of the Baltic states in the face of even a small Russian attack is probably impossible (Paul Mason has written well about this). I imagine the Estonian GHQ have many more sleepless nights than their Canadian counterparts.

There’s a second line of reasoning to Trump’s policy though, and it goes like this: whatever the defence spending target is, we’ve all agreed to meet it, and lots of you aren’t ‘paying your share’. On the face of it, this is more persuasive: if NATO is a club, shouldn’t all members pay the subscription fee?

Here we have to realise that not every dollar spent on defence by every country is created equal. Different countries have very different defence objectives, and economies of scale really matter. Military kit is really expensive – let’s take a look at some well-established costs for 3rd- and 4th-generation hardware. Remember, this isn’t the newest, shiniest, most-expensive-est stuff like drones, cyber capabilities, or energy weapons: just common-or-garden NATO workhorse kit of the sort that’s been in service for over three decades:

• US F-16 fighter: $14,000-000 – $18,600,000
• Italian-French-Moroccan FREMM multipurpose frigate: ~$500,000,000
• Tomahawk cruise missile: $250,000 per munition
• German Puma infantry fighting vehicle: €12,000,000
• AH-64D Apache attack helicopter: $33,000,000

The 2% of GDP would mean, for say Bulgaria or Lithuania, a total kitty of about $1000m per year. Just one frigate (or a squadron of fighters) would eat up half of that sum in capital expenditure alone, before we even think about running costs! The key thing to understand here is that a modern military works as a set of integrated parts. It doesn’t matter how great your tanks are; without air cover you’re boned, immediately, if your enemy has anything approaching a decent ground-attack capability. Then again, airfields in turn need protection from bombing, while all militaries have a very long tail of logistics to keep soldiers fed (and bandaged), trucks fueled, tanks mainteined in working order, and so on. Generally speaking, managing to mount one soldier in frontline operations for every 3-4 in support/logistics/training is considered really good*.

The current furore about scrapping some of our existing amphibious (e.g. beach landing) capability in order to pay for the two Queen Elizabeth-class carriers, their aircraft, and associated carrier battle groups highlights the stark choices even a large power like Britain faces: do everything on a shoestring, or focus on some capabilities (special forces; cyber; expeditionary warfare) at the expense of others (a nuclear deterrence force; large land operations).

Smaller powers such as Belgium or Poland feel this tension even more keenly, but since each countries’ citizens (and generals) want to feel their military can protect them on its own, without needing others’ help, we find that the European militaries consist of a succession of small-unit forces, e.g brigades and regiments, not division-scale forces – each with its own tankers, training institutions, intelligence corps and headquarters. This is incredibly wasteful, but predictable since the illusion of national independence (as opposed to interdependence) relies on the presumed ability  to act alone if need be.

There is a way to address some of this redundancy of course: have each nation contribute select components of a modern military capability, rather than trying to cover all bases. In extremis, this could mean a completely integrated NATO army where the Germans and Poles provided the massed land formations; the UK and French the surface fleet; the Hungarians the tankers, and so on – but while there is a long history of attempting to do this (the famed “NATO 7.62mm / 5.56mm” small-arms calibres, for instance, or the Franco-German Brigade) the reality is the people of NATO are, at present, more attached to national glory than military effectiveness – who wants to parade a squadron of petrol tankers on Armistice Day, anyway? Plus, this is just the sort of sovereignty-pooling Euro-nonsense many on the right recoil at – let alone Trump and his neo-isolationist MAGA crowd, happily powerful enough that this sort of impossible choice never confronts them.

In other words, throwing more money at the problem – and harming the long-term economic health which underpins any power-projection by so doing – without first reforming NATO so the dollars/euros we commit are spent effectively, without duplication and waste – is economic and military insanity. But I suspect we’ll muddle on – until, or unless – an external action forces NATO to wake up and smell the kvass.

*For instance, the UK armed forces managed to deploy forces in Iraq and Afghanistan for most of the last 15 years which peaked at around 10,000 in Afghanistan, and 46,000 in Iraq (the invasion itself). Sustaining this level of commitment (~30,000-50,000 troops, of a total UK Armed Forces headcount of around 250,000 personnel – including 80,000 reservists – e.g. about 1:4 deployed:support troops) over this length of time stretched the UK’s capability almost to breaking point. And the UK / France force size is probably about the smallest it’s possible for a modern military to be while still retaining the ability to conduct independent operations (e.g. without the Yanks) anywhere in the world at short notice (Libya arguably showed even this premise to be shaky – we relied on some US support, and acted in concert with France).