NATO defence spending targets

NATO members agreed in 2014 to move toward spending two percent of GDP on defence within a decade, and in 2023 turned that aspiration into a floor rather than a ceiling. At the 2025 summit the alliance went considerably further, committing to five percent of GDP by 2035 - split into three and a half percent for core defence and one and a half percent for defence-related investment such as infrastructure, resilience and industrial capacity.

That split is the most important detail in the whole commitment, because the second component is defined much more loosely than the first. Roads and ports that serve a military purpose, cyber resilience and civil preparedness can all count, which means a member can move toward the headline number through reclassification as well as through procurement.

It is also a ratio, not an amount. The commitment is denominated in a share of an economy that grows, shrinks and gets revised - so a country can meet the target through a recession and miss it through a boom without changing its budget by a single unit.

Defence budgets in Europe have grown faster since 2022 than at any point since the Cold War, and that spending flows into a small number of industries with long order books and limited capacity. Order backlogs at European defence manufacturers stretch for years, which is why equity markets reprice these companies on political commitments rather than on delivered revenue.

The fiscal channel is the second one. Sustained defence spending at these levels changes borrowing requirements in countries already carrying high debt, and it competes with everything else in the budget. That is a genuine constraint on how fast the pledge can be honoured, and it is the main argument for treating the 2035 target with scepticism.

The third channel is industrial. Producing more ammunition, air defence and armoured vehicles requires factories, skilled labour and long-lead components that cannot be conjured in a budget cycle - so a spending commitment converts into capability with a lag of years, and the lag itself is a subject of markets.

NATO publishes defence expenditure estimates once a year, usually in the summer, covering the current year as an estimate and prior years as revised figures. Those numbers use the alliance's own definition of defence expenditure, which is not identical to any member's national budget line - it includes pensions for military personnel and certain paramilitary forces, and excludes some items national ministries would count.

That definitional gap is why a member can report one figure domestically and appear with a different one in NATO's table. Neither is wrong; they answer different questions. A market must name which source it uses, and in practice NATO's own publication is the only consistent one across all members.

The denominator comes from national accounts, which are themselves revised - sometimes substantially, and sometimes years later. A country recorded as meeting two percent in the summer estimate can fall below it in the following year's revision without having changed a single line of spending.