Defence Technology and the Strategic Industrial Base

The state of the sector

The defence technology sector has entered its most sustained period of expansion since the end of the Cold War. Russia's full-scale invasion of Ukraine in February 2022, intensifying strategic competition in the Indo-Pacific, and a broadening perception of threat among treaty allies have converted a decade of stagnant or declining real defence budgets into a multi-year upcycle. Global military expenditure, on the widely cited estimates of the Stockholm International Peace Research Institute (SIPRI), passed roughly USD 2.7 trillion in 2024, the steepest year-on-year real increase in over a decade. The portion of that spending directed at equipment, research, development, testing and evaluation, and the sustainment of fielded systems — the addressable market for defence technology firms — is expanding both in absolute terms and as a share of the whole.

Yet the defining feature of this cycle is not the money. It is the discovery, in Ukraine and in every serious wargame since, that the industrial base built for the post-Cold-War era cannot produce at the rates a protracted, high-intensity conflict demands. Western artillery-ammunition output at the start of the war was a fraction of monthly battlefield consumption; propellant, forgings, machine tools and skilled tradespeople emerged as the true limiting factors long before budget authority did. The central claim of this report is that the sector's constraint has migrated from demand to supply — from whether states will fund rearmament to whether the base can absorb the funding and convert it into delivered capability at pace and at scale.

Two structural shifts run through the analysis. The first is the tension between mass and sophistication: cheap, attritable, software-defined systems — loitering munitions, uncrewed aircraft and vessels, autonomy stacks — are altering the cost of imposing effects on an adversary, without yet displacing the exquisite crewed platforms that anchor incumbent revenues, export relationships and alliance politics. The second is the entry of venture-funded, software-native firms that have shown they can design and field capability quickly, but have yet to prove they can sustain certified, high-rate production and long-tail support across decades. How incumbents, entrants and governments resolve these tensions will determine where value accrues through 2030.

This report sizes the market with transparent, clearly labelled estimates, maps its structure and regional contours, and sets out three scenarios to 2030. Our base case is a durable but uneven expansion in which capacity slowly catches up to appropriations, autonomy captures a growing but still-minority share of procurement, and the Western-aligned blocs deepen industrial integration while decoupling further from strategic competitors. The risks — fiscal reversal, supply shocks, and the gap between political commitment and delivered output — are substantial and treated candidly.

USD 2.7tn

Global military expenditure, 2024

SIPRI estimate

USD 700–800bn

Addressable defence technology market, 2024

authors' estimate; working base ≈ USD 750bn

USD 632bn

SIPRI Top 100 arms revenue, 2023

bottom-up anchor

25–33%

Share of outlays that is equipment + R&D

rule of thumb across advanced militaries

Headline conclusions

• The addressable defence technology market is an estimated USD 700–800 billion in 2024. This covers equipment procurement, defence RDT&E, and sustainment/MRO, and is anchored to SIPRI's Top 100 arms-producing companies (combined arms revenue of roughly USD 632 billion in 2023) plus non-Top-100 producers and defence-specific services. It is growing in the high single digits in real terms — well above the sector's long-run average.

• Capacity, not capital, is the binding constraint. The most instructive lesson of the Ukraine war for buyers has been that appropriated funds do not convert into shells, missiles and vehicles on the timelines assumed. Bottlenecks recur in energetics and propellants, large-calibre forgings and castings, solid-rocket motors, and — pervasively — skilled manufacturing labour.

• Autonomy and attritable mass are reshaping cost curves, not yet the order of battle. Uncrewed and loitering systems have demonstrated disproportionate effect at low unit cost, and doctrine is adapting quickly. But crewed combat aircraft, submarines, surface combatants and armour still dominate spending and will continue to for the horizon of this report.

• Input dependence is now a first-order strategic risk. Rare-earth processing, advanced-node and legacy semiconductors, specialty chemicals and certain alloys are concentrated in ways that buyers regard as unacceptable exposure. Reshoring and "friend-shoring" of these inputs is under way but is measured in years, not quarters.

• New entrants have proven capability; scaling remains unproven. Software-native firms have compressed development timelines and attracted significant venture capital. The open question is whether they can achieve and sustain the certified, high-rate, long-supported production that defines a strategic industrial base.

• The market is consolidating into blocs. Export controls, alliance frameworks such as AUKUS, and the European Union's push for industrial autonomy are drawing clearer lines between an interoperable Western-aligned market, a Chinese-anchored sphere, and a contested middle of large buyers hedging between them.

• Value is migrating toward the "digital thread." Sensing, command-and-control, electronic warfare, secure communications and autonomy software are growing faster than heavy platforms, and increasingly determine platform effectiveness. This favours firms that own the integration layer.

1. Context and why it matters

For roughly three decades after 1991, the defence industrial base of the advanced democracies was managed for efficiency rather than resilience — a logic coherent for its time. Great-power war was judged improbable, the strategic problems were counter-insurgency and expeditionary operations, and finance ministries pressed for a "peace dividend." Governments consolidated suppliers, ran production lines at low and predictable rates, held minimal inventories, and optimised for unit cost and sophistication rather than surge capacity. In the United States, the 1993 gathering of defence executives that became known as the "Last Supper" signalled official encouragement of consolidation, and dozens of prime contractors coalesced into a handful; European states pursued a parallel, slower rationalisation, complicated by the imperative to preserve national champions and domestic jobs.

That settlement has broken down. The proximate cause is Russia's invasion of Ukraine, which returned industrial-scale, attritional land war to the European continent and exposed how thin Western production capacity had become. But the shift is broader and predates 2022. Strategic competition with China, North Korea's and Iran's programmes, and a general erosion of confidence in the durability of the post-Cold-War order had already begun to move threat perceptions. Ukraine converted a gradual reassessment into an urgent one.

The consequences are visible in budgets and in doctrine. NATO members that had treated the alliance's 2 percent-of-GDP guideline as an aspiration began to meet and exceed it; at the alliance's 2025 summit, members endorsed a substantially higher long-run ambition, structured as core defence spending plus a broader security-and-resilience component. Germany's 2022 announcement of a special fund for its armed forces and its embrace of a Zeitenwende, or turning point, marked a decisive break with post-war restraint; Japan moved to raise spending toward 2 percent of GDP, a level without modern precedent for Tokyo; and Poland pushed toward some of the highest defence ratios in NATO. Across the board, the political ceiling on defence outlays rose.

1. 1991
   The peace dividend

   After the Cold War the advanced democracies' industrial base is managed for efficiency, not resilience: consolidated suppliers, low and predictable production rates, minimal inventories.

2. 1993
   The 'Last Supper'

   A gathering of US defence executives signals official encouragement of consolidation; dozens of prime contractors coalesce into a handful.

3. Feb 2022
   Russia invades Ukraine

   Industrial-scale, attritional land war returns to the European continent and exposes how thin Western production capacity had become.

4. 2022
   Germany's Zeitenwende

   Berlin announces a special fund for its armed forces, a decisive break with post-war restraint; Japan and Poland also move to raise spending.

5. 2025
   NATO raises its ambition

   At the alliance summit members endorse a substantially higher long-run goal — core defence spending plus a broader security-and-resilience component.

Why this matters beyond the sector is straightforward. Defence procurement is one of the largest discretionary claims on the public purse, a significant employer of skilled labour, a driver of advanced manufacturing and a locus of dual-use innovation. The choices made now — what to buy, from whom, and how much resilience to pay for — will shape fiscal trajectories, alliance cohesion, industrial geography and the technological frontier for a generation, and are being made under genuine uncertainty about the character of future conflict.

2. Market structure and scale

Sizing the market requires care, because "defence spending" and "the defence technology market" are not the same thing. Total military expenditure includes personnel, operations, maintenance, infrastructure and pensions; only a portion is available to industry as equipment procurement and RDT&E. As a rule of thumb across advanced militaries, equipment and R&D together account for between a quarter and a third of total outlays, the remainder consumed by people and running costs. Applying that logic to global military expenditure of roughly USD 2.7 trillion in 2024, and adjusting for the fact that procurement shares differ sharply between spenders, yields an addressable defence technology market — equipment, RDT&E and defence-specific sustainment — of an estimated USD 700–800 billion.

This top-down figure is consistent with a bottom-up anchor. SIPRI's Top 100 arms-producing and military-services companies recorded combined arms revenue of approximately USD 632 billion in 2023; adding producers outside the Top 100, defence-specific services and the portion of government RDT&E that flows to industry supports an estimate in the same USD 700–800 billion band. We treat the midpoint, around USD 750 billion, as our working 2024 base, and stress that it is an estimate assembled from public data, not a measured total.

The table below decomposes that base into segments. Segment figures are the authors' estimates, derived by cross-referencing procurement records, corporate segment disclosures and published industry analyses; they are indicative shares with meaningful uncertainty, not audited accounts. The 2030 column expresses a range under our base-case assumptions (Section 3), not a point forecast.

Read as shares of the working base, the composition is platform-heavy but visibly tilting:

• Aerospace platforms26%
• C4ISR, electronics & EW16%
• Land systems13%
• Naval & maritime13%
• Missiles & munitions11%
• Sustainment, MRO & services10%
• Defence space & satellites7%
• Cyber, autonomy software & AI5%

Two features of this structure deserve emphasis. First, concentration. The prime-contractor tier — Lockheed Martin, RTX, Boeing's defence arm, Northrop Grumman and General Dynamics in the United States; BAE Systems, Airbus Defence and Space, Leonardo, Thales and Rheinmetall in Europe — captures a large share of platform and systems revenue and sits atop supply chains reaching thousands of tiers. Engineered deliberately for efficiency, it is now viewed, in part, as a resilience liability. Second, the growth is uneven. Munitions, C4ISR, space and cyber/autonomy are growing considerably faster than the mature platform segments, even though platforms remain the largest single claim on spending. Value, in other words, is migrating toward sensing, connectivity, software and expendable mass, even as the balance sheet is still dominated by heavy metal.

Plotting each segment's 2024 estimate against its illustrative 2030 base-case range makes the uneven growth explicit — cyber/autonomy, missiles, space and C4ISR expand fastest even as aerospace remains the largest single claim:

• 2024 (est.) (USD bn)
• 2030 (est.) (USD bn)

3. Demand: the return of mass and the limits of budget authority

The demand side of this cycle is easy to describe and easy to overstate. Budgets are rising, and the political consensus behind them is broader and more durable than at any point in a generation. But two qualifications matter.

The first is that appropriated money is a leading indicator, not a delivered capability. The Ukraine war made this vivid in artillery ammunition: at the start of the conflict, combined Western monthly production of 155mm shells was a small fraction of the rate at which they were being expended. The United States set out to raise output several-fold over a period of years; the European Union launched instruments to expand production and pledged very large shell deliveries to Ukraine, targets initially missed and then partially recovered. The lesson generalises. Solid-rocket motors, air-defence interceptors, loitering munitions, armoured-vehicle hulls and large-calibre gun barrels face similar dynamics: the constraint is the physical line, the specialised input, the qualified supplier and the trained worker, none of which scales at the speed of a supplemental appropriation.

The second qualification is that "mass" has returned to strategic thinking after decades of emphasis on precision. High-intensity war consumes materiel at rates exquisite, low-rate production cannot replace, reviving interest in magazine depth — how many rounds and effectors a force can fire before running dry — and in the affordability of the systems that deliver them. It is here that uncrewed and attritable systems enter the demand picture: a loitering munition or a modest uncrewed aircraft can impose costs far exceeding its own price, produced by supply chains closer to commercial electronics than to traditional defence manufacturing. Programmes to field large numbers of inexpensive autonomous systems on short timelines — the US Department of Defense's Replicator initiative among them — reflect an explicit bet that quantity, networked and made smart, has a quality of its own.

Yet the demand for mass has not cancelled the demand for sophistication. Crewed combat aircraft, submarines and integrated air-and-missile defence remain central to every serious force design, both for their operational role and because they anchor alliance interoperability and export relationships that are themselves instruments of statecraft. The realistic near-term picture is additive: buyers want more of the cheap and attritable and continuity in the exquisite, at a moment when the industrial base struggles to deliver either at the desired rate. That combination is the source of both the sector's growth and its strain.

4. Supply: consolidation, capacity and the new entrants

If demand is the easy part of the story, supply is the consequential part. Three dynamics are in play.

Incumbent capacity and the economics of surge. Prime contractors and their supply chains were optimised for steady, low-rate production. Surging output requires capital investment in new lines and tooling, qualification of additional suppliers for safety-critical components, and, above all, hiring and training skilled labour in trades — welding, machining, precision assembly, systems integration — where the workforce has aged and thinned. Firms are reluctant to fund large expansions against demand they fear may prove temporary, which is why multi-year procurement commitments and government co-investment in capacity have become central policy tools: where they exist they de-risk the expansion, where absent or slow, capacity lags. The result is a sector in which order backlogs have swelled — good for revenue visibility — while delivery timelines have lengthened, a less welcome signal for the forces waiting on equipment.

Input and supply-chain concentration. Modern defence systems depend on inputs concentrated in ways buyers now consider strategically unacceptable. Rare-earth elements, and especially their processing and magnet manufacture, are dominated by a single supplier state to a degree that creates real leverage. Advanced semiconductors are produced at scale in a handful of locations, several exposed to the very contingencies defence planners worry about; legacy chips, ubiquitous in military systems, carry their own concentration risks. Specialty chemicals, energetics precursors, certain titanium and tungsten supplies and optics round out the choke points. Diversifying these inputs — through reshoring, allied "friend-shoring," stockpiling and qualification of alternative sources — is under way and is a genuine growth driver for parts of the base, but it is expensive and slow, and no serious analyst expects it complete within this report's horizon.

New entrants and the venture-capital wave. A cohort of software-native firms has entered defence, attracting substantial venture funding. Their distinctive contribution is speed and a software-first architecture: autonomy stacks, sensor fusion and mission software developed on commercial timelines and updated continuously, against the multi-year cycles of traditional programmes. In autonomy, uncrewed systems and battlefield software, several have moved from concept to fielded capability with a rapidity incumbents have struggled to match. The pivotal question for where value settles is scale and endurance. A strategic industrial base is defined not only by building a capable system but by certifying it, producing it at high and sustained rates, supporting it for decades, and absorbing the compliance and security obligations defence buyers impose. Whether the new entrants graduate into that role, are absorbed by incumbents, or settle as software suppliers to prime-led programmes remains open; the most likely outcome is a blend of all three.

The interaction of these three dynamics is producing a base that is simultaneously more concentrated at the top (primes and their tier-one suppliers) and more porous at the edges (new entrants, commercial-technology suppliers, and a widening set of software vendors). Managing that hybrid — extracting the speed of the newcomers without sacrificing the reliability and accountability of the incumbents — is the central industrial-policy problem of the decade.

5. Regional and comparative lens

The market is not global in any unified sense; it is increasingly organised into blocs whose boundaries are drawn by alliance politics and export control.

United States. The United States remains the largest single market and the anchor of the Western-aligned base, accounting for a substantial share of both global military spending and top-tier arms revenue. Its industrial-policy instruments — multi-year procurement, defence-production authorities, targeted investment in munitions and shipbuilding — are the most developed, though shipbuilding and munitions capacity remain acknowledged weaknesses. Through its export-control regimes, Washington effectively sets the terms on which allied and partner states access much of the frontier, making it the gatekeeper of the wider Western market.

Europe. Europe is the fastest-moving region by rate of change. The combination of the Ukraine war, higher NATO commitments and a policy drive for greater industrial autonomy — expressed through the European Union's defence-industrial strategy, ammunition-production and joint-procurement instruments, and rising national budgets — has produced the sharpest demand surge. Rheinmetall, Saab, BAE Systems, Leonardo, Thales and Airbus's defence business are principal beneficiaries. Europe's structural challenge is fragmentation: multiple national programmes, duplicated platforms and non-harmonised requirements raise costs and dilute scale. The central European debate — how far to consolidate demand and production across borders versus preserving national champions and sovereign capability — is unresolved and consequential.

Indo-Pacific. The Indo-Pacific is where great-power competition is most directly shaping procurement. Japan's move toward higher defence spending, South Korea's emergence as a cost-competitive arms exporter, Australia's participation in the AUKUS partnership for nuclear-powered submarines and advanced capabilities, and Taiwan's emphasis on asymmetric, resilient defence all point to a durable, technology-intensive build-up. South Korea in particular has shown it can deliver capable systems at competitive prices and on short timelines, positioning it as a rising force in the exporter tier.

China and the competitor sphere. China has built a large, increasingly capable and substantially indigenous defence industrial base, and is the world's second-largest military spender on standard estimates, though the opacity of its accounting means figures carry wide uncertainty. Its base anchors a distinct sphere with its own supply chains and export relationships. Russia's industry, under sanctions and the demands of an active war, has sustained high output of some categories while depending on sanctioned or grey-market inputs for others — a live demonstration of both the resilience and the vulnerability of a base cut off from the Western-aligned network.

The contested middle. A large group of buyers — across the Gulf, South and South-East Asia, and parts of Africa and Latin America — is hedging, purchasing from multiple spheres and using supplier competition for leverage. For exporters, these are the markets where growth and margin are contested most directly, and where price, financing, offsets and speed of delivery increasingly outweigh a marginal technological edge.

6. Risks and what could turn out differently

An honest assessment must foreground what could go wrong or turn out differently.

Fiscal reversal. The current spending consensus rests on a heightened perception of threat. A change in that perception — a negotiated settlement in Ukraine, a shift in domestic politics, or a fiscal crisis forcing consolidation — could slow or reverse budget growth faster than industry can adjust the capacity it is now, tentatively, building. The memory of the post-Cold-War drawdown is precisely why firms hesitate to expand without multi-year commitments.

The commitment-to-delivery gap. Political pledges to spend are outrunning the base's ability to deliver. If large appropriations translate into long backlogs and slipped timelines rather than fielded capability, the credibility of deterrence — and public support for the spending — could erode even as headline budgets rise.

Supply shocks. An interruption to a concentrated input — a rare-earth export restriction, a semiconductor squeeze, an energetics-precursor shortage — could stall production across multiple programmes at once. This is the scenario buyers most fear and are least able to fully insure against on a short horizon.

Technological mis-bet. The character of future war is genuinely uncertain. Over-investing in exquisite crewed platforms that prove vulnerable to cheap mass, or in attritable autonomy that proves brittle against sophisticated electronic warfare, are symmetric risks. The prudent hedge — a balanced portfolio — is more expensive than betting correctly, and correct bets are only visible in hindsight.

Escalation and use. The most serious uncertainty sits outside the market. A wider conflict would not be a demand event to be modelled; it would be a discontinuity that overrides every projection in this report. Our figures assume no systemic great-power war within the horizon.

Scenarios for the base to 2030

We present three scenarios. They are illustrative framings, not forecasts, and each rests on stated assumptions.

Scenario A — Durable, uneven expansion (base case). Threat perceptions remain elevated without a systemic great-power war. Western budgets hold near or above current commitments; capacity investment continues; multi-year procurement and government co-investment gradually relieve the worst bottlenecks. Autonomy and uncrewed systems grow into a meaningful minority of procurement, complementing rather than replacing exquisite platforms. Bloc consolidation deepens. Under this case, the addressable market grows in the mid-to-high single digits in real terms, reaching an estimated USD 1.0–1.2 trillion by 2030, with munitions, C4ISR, space and autonomy software growing fastest. This is our central expectation.

Scenario B — Constrained plateau. A partial de-escalation, a fiscal squeeze in one or more major economies, or persistent failure to convert budgets into delivered capability slows the cycle. Budgets flatten in real terms; some capacity expansions are paused; new entrants consolidate as venture funding cools. The market still grows, but modestly, reaching an estimated USD 850 billion–1.0 trillion by 2030, concentrated in sustainment, munitions replenishment and the highest-priority modernisation. Value accrues to firms with recurring-revenue franchises and diversified backlogs.

Scenario C — Accelerated, contested build-up. A further deterioration — a new crisis in the Indo-Pacific, a widening of conflict in Europe short of systemic war, or a sharp escalation in strategic competition — pushes budgets and urgency higher. Emergency instruments, wartime-style production authorities and stockpiling drive rapid capacity expansion, though input choke points and labour shortages cap how fast output can rise. Autonomy scales quickly under operational pressure. The market could reach USD 1.2–1.4 trillion or more by 2030, but with heightened supply-shock and mis-bet risk, and with delivery persistently lagging appropriation.

A — Durable, uneven expansion

Central case

Threat perceptions stay elevated without systemic great-power war; capacity slowly catches up to appropriations; autonomy grows into a meaningful minority of procurement; bloc consolidation deepens.

Market 2030

USD 1.0–1.2tn

Real growth

mid–high single digits

B — Constrained plateau

Downside

Partial de-escalation, a fiscal squeeze, or persistent failure to convert budgets into delivered capability slows the cycle; some expansions pause; venture funding cools.

Market 2030

USD 850bn–1.0tn

Concentrated in

sustainment, munitions

C — Accelerated build-up

Upside / higher risk

A further deterioration pushes budgets and urgency higher; emergency instruments and stockpiling drive rapid expansion, though input choke points and labour shortages cap output.

Market 2030

USD 1.2–1.4tn or more

Key risk

supply-shock, mis-bet

The three cases fan out from a common 2024 base of roughly USD 750 billion:

Across all three, several directional judgements hold: value migrates toward software, sensing and connectivity; supply-chain resilience commands a durable premium; and the base becomes more, not less, an instrument of alliance and industrial policy rather than a purely commercial market.

Implications across the sector

For governments and policymakers. The decisive lever is not the size of the appropriation but the credibility and duration of the demand signal. Multi-year commitments, capacity co-investment and a broader qualified supplier base do more to expand output than one-off supplementals, which industry rationally hesitates to build against. Treat munitions, energetics, castings and skilled labour as strategic reserves warranting deliberate investment, and fund the diversification of rare earths, semiconductors and specialty chemicals, accepting that resilience carries a cost efficiency-era accounting will resist. Design acquisition to admit the speed of new entrants without abandoning the accountability the sector requires — a balance current processes handle imperfectly.

For business and industry. Incumbents should read the migration of value toward the digital thread as a directive: ownership of the integration, autonomy and C4ISR layers will increasingly determine platform relevance and margin. Backlog is not delivered revenue; the firms that invest early in capacity and workforce, and partner intelligently with software-native entrants, will convert this cycle into durable share. New entrants should be clear-eyed that fielding a capable system is the beginning, not the end, of joining a strategic industrial base — certification, high-rate production, sustainment and security compliance are where the deeper moat lies.

For investors and financial institutions. The upcycle is real and broad, but uneven and cyclical, and the sector's own memory of the post-Cold-War drawdown is instructive. Recurring-revenue sustainment franchises and diversified backlogs offer resilience across scenarios; pure-play exposure to a single programme or national budget carries concentrated policy risk. In the venture segment, distinguish carefully between demonstrated fielded capability and the far rarer achievement of sustained production at scale. The un-glamorous input tiers — energetics, castings, motors, specialty materials — may offer some of the more durable exposure to the theme.

For international organisations and allied partners. Consolidation into blocs raises the value of interoperability, joint procurement and pooled capacity, and raises the cost of fragmentation. For smaller and partner states, the strategic questions are how to secure access to the frontier through alliance membership and co-production, and how to build sovereign capability in the segments — increasingly software, autonomy and sustainment — where barriers to entry are falling even as they rise elsewhere.

Method and evidence base

This report is a synthesis and analytical assessment, not a primary statistical study. Its evidentiary base comprises published estimates of military expenditure and arms-industry revenue from established, non-partisan research bodies (notably SIPRI's military-expenditure database and its Top 100 arms-producers series); government budget and procurement documentation; the segment disclosures of listed defence contractors; and structured interviews with sector specialists under non-attribution terms.

Readers should be explicit about which numbers are which. Drawn from the published record, and established if still estimated by their originators, are: global military expenditure of roughly USD 2.7 trillion in 2024; combined SIPRI Top 100 arms revenue of approximately USD 632 billion in 2023; and the existence and broad character of the national and alliance commitments described. The authors' own estimates, assembled by triangulating those sources and stated as ranges because the underlying data are incomplete or definitional boundaries contestable, are: the USD 700–800 billion addressable-market figure; every segment size and share in the Section 2 table; and all figures in the scenarios. The segment table reflects analytical judgement about how to allocate spending across categories that overlap in practice (a combat aircraft embeds C4ISR, munitions and software), and its shares should be read as indicative rather than precise.

Three limitations warrant emphasis. First, defence data are inherently uncertain: classification, off-budget financing, purchasing-power differences and definitional inconsistencies mean even "established" figures carry wide error bars, and those for less transparent spenders more so. Second, the forward scenarios are conditional framings, not predictions, invalidated by the one contingency — systemic great-power war — that would matter most. Third, this is a fast-moving sector; figures reflect information available at the time of writing, and the assumptions behind every estimate are stated so readers can substitute their own.

Sources consulted

The expenditure and arms-revenue series, alliance figures and strategy documents below were read directly; corporate segment disclosures and non-attributable specialist interviews informed the segment allocation.

• Stockholm International Peace Research Institute (2024). Trends in World Military Expenditure, 2023 and SIPRI Military Expenditure Database. SIPRI, Solna.
• Stockholm International Peace Research Institute (2023). The SIPRI Top 100 Arms-Producing and Military Services Companies, 2022. SIPRI, Solna.
• International Institute for Strategic Studies (2024). The Military Balance 2024. IISS and Routledge, London.
• NATO (2024). Defence Expenditure of NATO Countries (2014–2024). North Atlantic Treaty Organization, Brussels.
• US Department of Defense (2023). National Defense Industrial Strategy and the Replicator initiative. US DoD, Washington, DC.
• European Commission (2024). European Defence Industrial Strategy (EDIS) and the Act in Support of Ammunition Production (ASAP). Brussels.
• Government of Japan (2022). National Security Strategy and National Defense Strategy. Cabinet Secretariat, Tokyo.
• Segment disclosures of leading prime contractors (2023–2025) and structured, non-attributable interviews with sector specialists, used to allocate spending across overlapping categories.