The World Pays for Weapons: Who Overspends, Who Accumulates, Who Goes Under

At first glance, just columns of numbers. View it through Giovanni Arrighi’s lens, however, and the picture changes entirely. This is not a report on who spent how much on tanks. It is a snapshot of where each major player currently sits within the systemic cycle of capital accumulation.

Arrighi described how a hegemon in decline begins spending more on sustaining the global system than the system returns to it. Spain in the sixteenth century. Holland in the eighteenth. Britain in the early twentieth. Each, at a certain point, transformed from producer into rentier, and their military expenditures relative to economic weight began to spike — because only force could still hold what economics no longer held. In parallel, a challenger rose — one accumulating productive capacity without prematurely converting it into military power.

The year 2025 is, in effect, a documentary snapshot of this model in action. Place a country’s share of global military spending alongside its share of global GDP, and you get a diagnosis.

The big players: hegemon, challenger, parasite

The United States spent $954 billion — 33% of all global military spending. American economy makes up 26.1% of world GDP ($30.6 trillion). Overspending coefficient: 1.27. Meaning Washington spends roughly a quarter more than its economic weight on defence. And this is after a 7.5% reduction in 2025 — a temporary one, since “the decline in US military spending in 2025 will likely be short-lived,” as SIPRI programme director Nan Tian commented. The cut did not stem from prosperity: Congress simply failed to pass a new aid package for Ukraine. For 2026, Trump is proposing a record $1.5 trillion.

Here it is — classic Arrighian financialisation of decline. An empire no longer able to fund its dominance from a productive economy starts pushing money into the military sector because that is the only place where consensus to spend remains. On education, infrastructure, healthcare — fights. On “containing China” — everyone agrees.

China spent $336 billion — 11.6% of the world pie. Against an economy of 16.6% of global GDP. Coefficient: 0.70. Beijing spends 30% less than its economic weight. China has increased military expenditure by 7.4% — the thirty-first consecutive year of uninterrupted growth. Not a single year of pause. Yet even with that growth, the coefficient stays below one.

This is the conduct of a challenger, not an aggressor. The United States behaved this way from the 1880s through the 1910s. Accumulate. Build ships. Expand industry. Stay out of wars prematurely. Military conversion will come — but later, when productive power can sustain it without strain. A note: some experts argue China’s actual figure may run considerably higher, since Beijing does not fully disclose military expenditure. Even with that adjustment, the coefficient unlikely tips above one.

Russia spent $190 billion — 6.6% of global spending. Against an economy of 2.17% of world GDP ($2.54 trillion). Coefficient: 3.04. Moscow spends slightly over three times its economic weight. Russia commits 7.5% of its GDP to military needs — more than triple the global average of 2.5%. Military spending makes up 20% of all state expenditure — the highest level on record.

This is not “great-power conduct.” This is convulsion. In Arrighi’s theory, regimes like this occupy a separate niche — the parasitic intermediary, living off the rent on resources passing from colonies to the global hegemon. As the cycle nears completion, the parasite stops receiving its share and begins cannibalising its own body. Precisely what we observe: defence now devours 32% of the budget, while social policy, education, medicine get whatever is left. Reserves burn, oil and gas rent fails to cover the structural gap, and 2026 has further growth scheduled.

The middle group: those converting

Germany — $114 billion, 3.9% of global spending against an economy of 4.27% of world GDP ($5 trillion). Coefficient: 0.92. Almost parity. The dynamic is more telling: spending grew by 24% in a year, and Germany exceeded 2% of GDP on defence for the first time since 1990, reaching 2.3%. Berlin lived for decades under the American umbrella and paid the minimum. When the umbrella began to disappear, the Bundestag urgently changed fiscal rules, exempting military spending above 1% of GDP from the debt brake. Classic signal: the client realised the suzerain no longer guarantees security — time to pay yourself.

Britain — $89 billion, 3.1% of global spending against an economy of 3.38% ($3.96 trillion). Coefficient: 0.92. Almost like Germany. Yet the British dynamic runs the other way: spending in 2025 fell by 2%. London is cutting aid to developing countries — reducing assistance from 0.5% to 0.3% of gross national income, redirecting funds to defence. Even own defence budget runs short, and the money gets stolen from “soft power.”

France — $68 billion, 2.4% of global spending against an economy of 2.73% ($3.2 trillion). Coefficient: 0.86. Growth of 1.5% in 2025. Spending on armaments for the armed forces — mostly French-made — rose 11% to $21.1 billion. Paris, unlike London, holds the line of self-sufficiency in defence as an industrial strategy.

Japan — $62.2 billion, 2.2% of global spending against an economy of 3.65% ($4.28 trillion). Coefficient: 0.59. The lowest among major players. Historically, this is the cheapest security in the world — Article 9 of the post-war constitution plus the American umbrella regiment. Two decades of “won” time invested not in divisions but in Toyota, Sony, Hitachi. That model is now falling apart. Prime Minister Sanae Takaichi has pledged to raise defence spending to 2% of GDP — the current 1.4% already the highest figure since 1958. Japan is lifting its export embargo on lethal weapons, signing its first contract with the Australian navy. The era of free security is ending.

India — $92.1 billion, 3.2% of global spending against an economy of 3.52% ($4.13 trillion). Coefficient: 0.91. Delhi has climbed to fifth place in the world, overtaking Britain. Spending grew 8.9%, driven by the May 2025 conflict with Pakistan in which combat aircraft, drones and missiles were used. India does what China does, but with a twenty-year lag — building productive economy without prematurely shoving money into military capital expenditure. Another candidate for regional centre of the next cycle.

South Korea — $47.8 billion, 1.7% of global spending against an economy of around 2% ($1.9 trillion). Coefficient: 0.85. Seoul deserves a separate look. It keeps military spending minimal relative to economy yet has built one of the most powerful defence industries on the planet: Hanwha Aerospace shares soared 193% in 2025, Hyundai Rotem 278%, LIG Nex1 91%. K9, K2, KF-21 sell across the globe. Korea quietly became the arsenal of those who refused Russian gear and cannot wait for American. Classic “producer” model, as Arrighi called it — sell the means of production without sliding into the role of major customer.

A separate category: those without a choice

Israel — $48.3 billion, 1.7% of global spending against an economy of $0.55 trillion (0.47% of world). Coefficient: 3.56. The highest among “normal” countries. Military spending dropped 4.9% after the January 2025 ceasefire with Hamas — yet even with that drop, it stayed 97% above the 2022 level. This is a garrison economy. It cannot afford a coefficient of 1, because the Golan wall and the Gaza border will not vanish from the Knesset passing a peaceful budget.

Poland — around $47 billion, 4.5% of GDP. Poland continues to carry one of the highest defence burdens in NATO at 4.5% of GDP. Growth of 23% over the year. Overspending coefficient above 2. Warsaw is doing what Moscow would be doing, were it a normal state — preparing for possible conflict with a real enemy, not an invented one. The Poles understand geography better than the Germans and the British: between them and the Russian flag lie two hundred kilometres of plain.

Turkey — $30 billion, around 1% of global spending against an economy of around $1.2 trillion. Coefficient: roughly 0.9. Turkey raised spending by 7.2%, partly because of military operations in Iraq, Somalia and Syria. Ankara does what Korea does — converts military expenditure into industrial policy. Bayraktar, Atmaca, Akıncı — the Turkish defence industry has become an export brand over a decade. Recep Erdoğan, for all the chaos of his rule, here acted by Colbert’s classic model: buy weapons from us, get Turkey as an ally.

Ukraine — $84.1 billion, 2.9% of global spending against an economy of around $0.21 trillion (0.18% of world). Coefficient: 16.1. The highest share of GDP on defence anywhere in the world — 40%. As a share of GDP, Ukraine is the world’s largest military spender at 40%, in the fourth year of war against Russian invasion. This is no longer “overspending” in the Arrighian sense. This is a separate category: existential mobilisation, where peacetime metrics lose their meaning. When you are the arena on which big players decide whose cycle continues, your figures fall outside the model.

Who is wise, who goes under

Take the warring countries — Ukraine and Russia — out of the equation, and the picture sharpens.

Spending wisely — that is, considerably below their economic weight — are Japan (0.59), South Korea (0.85), France (0.86), India (0.91), Britain and Germany (0.92), Turkey (around 0.9). Countries either still living under another’s umbrella, or investing in their own defence industry as an export sector, or lacking, for now, an existential reason to strain.

Overspending out of forced necessity — the United States (1.27), Poland (above 2), Israel (3.56). Either an empire defending a global position, a frontline state, or a permanent garrison.

Going under — Russia (3.04). Because its coefficient is justified neither by hegemonic position nor by frontline status in the classical sense. This is the coefficient of a country no longer holding itself up on its own productive base. Moscow gives 7.5% of GDP to war on an economy of 2.17% of world GDP — a gap closable only through cannibalising domestic resources. National Wealth Fund reserves dwindle, oil and gas revenues fall, the central bank’s rate has strangled the private sector, inflation refuses to retreat. In 2026, defence is set to grow further — while no new sources of revenue have appeared. Classic dynamic of a cycle running out of breath.

The United States is a separate story. Technically, the overspending is moderate (1.27), yet the trajectory looks alarming: Trump’s $1.5 trillion proposal for 2026 means a sharp jump in the coefficient. This is financialisation: the money exists, but increasingly it gets pushed into the unproductive military sector, because productive investment requires a political consensus that does not exist, while for defence the standard formula suffices — “China threatens, so vote yes.”

China is not yet a challenger-in-action. It is a challenger-in-ambush. A coefficient of 0.70 is the figure of a country that has not yet decided to step out of the shadow. Japan in the 1930s had similar numbers before the Manchurian gambit. The United States in the 1900s — before the Spanish-American War. The very lowness of the coefficient means powder is stockpiled, but the spark has yet to fly.

What this means in practice

To look at absolute military spending figures — who spent how many dollars — is to see the surface. To look at the overspending coefficient — the ratio between a country’s share of global defence and its share of global economy — is to see the phase of the cycle. The empire overspends in absolutes, but that is a symptom, not strength. The challenger underspends, but that is a symptom of strength, not weakness. The parasite-in-convulsion overspends threefold against an economy that does not carry such ambition — a symptom of departure from the system. The arena state overspends sixteenfold — a symptom that the fate of an epoch is being decided on it.

The world of 2025 is a world in which the American cycle of accumulation runs out of breath before our eyes, the Chinese accumulates without haste, the Russian completes its final convulsion, and around them Japan, Germany, Poland, Britain, Korea, India and Turkey rebuild their long-term calculations to fit a new configuration in which there will be no one’s guarantees. This is the moment between hegemonies. Arrighi wrote that such moments are always periods of wars. The SIPRI numbers confirm it.

Press enter or click to view image in full size

World GDP 2025: $117.2 trillion World military spending 2025: $2,887 billion (2.5% of world GDP) Overspending coefficient = country’s share of global military spending ÷ country’s share of global GDP. Above 1 — the country carries a military burden heavier than its economic weight. Below 1 — underspending, life under someone else’s umbrella, or accumulation without conversion.

Average values:

• Arithmetic mean across 13 countries: 2.53
• Excluding Ukraine (clear outlier): 1.40
• Excluding countries in active wars (Ukraine, Russia, Israel): 1.02
• Median across all 13: 0.93

Sources: SIPRI Military Expenditure Database, April 2026; IMF World Economic Outlook 2025–26.