Abstract: Since the 2023 Vilnius Summit, NATO has recognised industrial capacity as a core element of deterrence. The Alliance has endorsed the Defence Production Action Plan, adopted the Industrial Capacity Expansion Pledge, and aligned with parallel US and EU initiatives, the National Defence Industrial Strategy and the Act in Support of Ammunition Production to accelerate output, standardisation, and supply-chain resilience. Yet policy vocabulary has advanced faster than procurement logic. The return of industrial war, combined with a sensor-saturated battlefield, means that design choices now carry direct operational consequences. Manufacturability, supply substitution, repairability, and distributed production potential are variables of combat endurance. This article argues that NATO’s real gap is not only an industrial-capacity gap, but also an industrial-operational doctrine gap: the Alliance has not yet fully integrated production resilience and battlefield transparency into a mature procurement doctrine suited to prolonged high-intensity conflict.
Problem statement: Why do NATO defence-industrial reforms still focus primarily on expanding output within existing production structures instead of fully adapting procurement doctrine?
So what?: NATO defence ministries must add manufacturability, distributed production potential, and low-signature survivability as formal scoring criteria in capability requirements documents alongside the performance metrics that currently dominate. The NATO Defence Planning Process should be updated to require member states to demonstrate that major procurement programmes have assessed these criteria before approval. Throwing money at existing programme structures will not close the gap; the criteria themselves have to change.
From Shortage Awareness to Procurement Reform
NATO’s earlier concern that the Alliance was structurally blind to its own industrial capacity shortfalls is now largely resolved, at least in terms of recognition. NATO has, in fact, acknowledged the shortage problem. In its 2023 Vilnius Summit communiqué, the alliance explicitly endorsed a Defence Production Action Plan.[1] The plan’s initial focus on land munitions reflected a direct response to observed wartime consumption rates. That the Alliance endorsed this plan in a summit communiqué rather than a working-level document signals that industrial throughput, standardisation, and procurement speed are now treated as strategic concerns at the alliance level.[2]
The 2024 Washington Summit moved even further. The Washington Summit Declaration tied transatlantic defence-industrial cooperation directly to deterrence and defence. It also stated that stronger defence industries across Europe and North America improve the Alliance’s ability to promptly meet the requirements of NATO defence plans and linked industrial strength to immediate and enduring support for Ukraine. It also rooted national and collective resilience in Article 3 of the North Atlantic Treaty and committed Allies to integrating civilian planning into national and collective defence planning across peace, crisis, and conflict, and to addressing threats to democratic systems, critical infrastructure, and supply chains. Article 3 defines treaty obligation requiring Allies to continuously develop, individually and collectively, the military and civil capacity to resist armed attack; in contemporary NATO practice, this is understood as the legal foundation for resilience, preparedness, and the maintenance of credible defence capacity before war begins.[3] The NATO Industrial Capacity Expansion Pledge then reinforced this trajectory by explicitly committing Allies to accelerate the growth of defence-industrial capacity and production across the Alliance.[4]
NATO’s own public material now presents industrial expansion as a standing requirement for collective defence and for replenishing Allied arsenals. At the same time, support to Ukraine continues to ensure current conflicts do not undermine future capabilities.
The Gap has Shifted
Production is necessary, but it does not automatically produce wartime resilience. A larger version of a brittle model remains brittle. A faster version of a centralised model remains centralised. A better-funded procurement system can still privilege designs that are difficult to regenerate under pressure. NATO’s recent policy documents are strongest on questions of coordination, throughput, and cooperation. They are less mature on the question of whether the inherited architecture of Western defence production is itself aligned with the operational environment NATO now expects to face.
NATO’s industrial architecture has long prioritised sophisticated systems, long programme cycles, tightly controlled intellectual property, concentrated prime-contractor authority, and production runs optimised more for performance and compliance than for crisis elasticity.
The prevailing NATO acquisition model, characterised by high-end performance requirements, long programme cycles, and concentrated prime-contractor authority, reflected decades of military thought shaped by the Revolution in Military Affairs (RMA), a strategic paradigm that privileged precision strike, network-centric operations, and low-volume high-value procurement over mass production and industrial depth.[5] The result was a model that often excelled at producing highly capable systems, but not always one designed for rapid wartime expansion, dispersed regeneration, or tolerance for disrupted supply chains. NATO’s more recent official texts implicitly acknowledge this by repeatedly emphasising the need for urgent production, resilient supply chains, and reduced obstacles to defence trade and investment.
NATO has moved decisively on how much to produce and how fast. Yet the policy gap is in what it has not addressed: which design characteristics should be systematically rewarded in procurement, and whether production capability should be concentrated or dispersed.
More than An “Ammunition Shortage” Cliché
Ukraine is widely cited in Western defence policy literature and official alliance assessments as evidence that industrial war has returned at scale.[6] Specifically, it is argued that the scale and pace of materiel consumption in high-intensity conventional conflict exceed what post-Cold War procurement assumptions were designed to sustain.[7] That is true, but it is too often expressed in a flattened way, as though the lesson were merely that artillery consumption is high and therefore factories need to make more shells.
The war in Ukraine demonstrates the convergence of four pressures: attrition, adaptation, surveillance, and industrial asymmetry. Fourth, industrial asymmetry matters. Russia’s ability to sustain artillery shell production at scale, estimated at over three million rounds annually by 2024, while Western resupply to Ukraine repeatedly hit production ceilings, demonstrates that the side that can regenerate usable capability faster does not merely preserve stock; it dictates operational tempo.. Both sides have cycled through countermeasures at speed, from electronic jamming of drone guidance systems to field-improvised armour defeating new munition types, compressing the interval between technological introduction and obsolescence from years to weeks. Third, surveillance is persistent. Drones, commercial imagery, thermal systems, and networked intelligence, surveillance, and reconnaissance (ISR) compress the time between exposure and targeting. Fourth, industrial asymmetry matters. States and coalitions that cannot regenerate usable capability quickly enough lose more than stock; they lose initiative.[8]
Thus, the lesson for planners is not simply “produce more.” It is to produce in ways compatible with adaptation, redundancy, and contested use. The Ukraine case cannot be mapped directly onto NATO Alliance members, as they collectively possess an industrial base far beyond what Ukraine could draw on, and their constraints are primarily political and doctrinal rather than structural. That distinction sharpens rather than softens the argument: if NATO nations have the industrial capacity and are choosing not to fully mobilise or restructure it, the gap is in procurement doctrine and political prioritisation, not raw capability.
What Counts As Good Procurement?
The concept of the transparent battlefield has become increasingly difficult to ignore. Even where NATO’s official texts do not use that exact term, their repeated concern with resilience across all domains reflects a wider recognition that modern sensing technologies have fundamentally altered the conditions under which forces operate and equipment must be sustained.[9]
The growing accessibility of drones, thermal imaging, satellite observation, and layered ISR means that battlefield survival is increasingly tied to signature discipline and the ability to operate under persistent observation.[10]
This changes what should count as strategically efficient design. In a lower-transparency environment, an expensive and difficult-to-produce system might still be tolerable if it provides sufficient tactical advantage and can be fielded in manageable numbers. In a highly transparent environment, however, the equation shifts. Systems that are quickly detected and burdensome to resupply impose a larger operational tax. Likewise, systems that require rare inputs, narrow industrial bottlenecks, or excessively specialised labour impose a larger regeneration tax. In other words, future conflict may force a merger between two conversations that Western procurement often kept apart: the conversation about industrial robustness and the conversation about operational survivability.
In NATO, the gap is primarily doctrinal rather than legal or cultural. NATO’s legal framework points toward industrial preparedness. However, the political will to spend more on defence is increasingly present. What remains underdeveloped is the procurement doctrine that would translate both into systematic selection criteria. If those criteria remain anchored to performance metrics inherited from a lower-threat, lower-tempo era, NATO may expand its industrial base without fully reorienting it.
NATO Has A Capacity Agenda, but…
A central weakness in current Alliance thinking is the tendency to treat aggregate capacity as synonymous with resilient capacity.[11] NATO documents repeatedly call for boosting production capacity, strengthening the defence industry across the Alliance, and delivering critical capabilities urgently. However, capacity can still be expanded in highly centralised ways. A handful of large firms can receive more contracts, expand existing lines, and add shifts, yet leave the overall system concentrated in a small number of industrial nodes. That may ease the shortage in the short term, but it does not eliminate strategic fragility.
A dispersion agenda does not require changes to NATO’s legal framework. The North Atlantic Treaty obliges Allies to maintain their capacity to resist armed attack, including an adequate industrial base.[12] However, questions remain whether Allied governments are interpreting it narrowly as a floor on output or broadly, as a requirement for industrial structures resilient enough to sustain operations. A larger centralised system can remain vulnerable to cyberattacks, sabotage, transportation shocks, labour shortages, power disruptions, or targeted pressure on critical suppliers. Dispersion is about redundancy, substitutability, and industrial recoverability.
NATO Leans Too Heavily on the Prime-Contractor Worldview
None of this means large defence primes are unimportant. They are indispensable for many advanced capabilities, for cross-border integration, and for alliance-scale standards compliance. The problem is that if procurement doctrine remains psychologically centred on the prime-contractor model, it will under-reward designs and production structures that draw strength from wider industrial ecosystems.
NATO’s own Vilnius language explicitly called for understanding the defence industry across the Alliance, including small and medium-sized enterprises (SMEs). That detail matters because it shows the Alliance recognises that industrial depth cannot be understood solely through the lens of the largest firms. The Washington Summit Declaration’s stress on public-private cooperation and integrating civilian planning into defence planning points in the same direction. So does NATO’s emphasis on reducing obstacles to defence trade and investment among Allies.[13]
However, recognising the existence of SMEs is not the same as building procurement doctrine around ecosystem resilience. The ecosystem model asks different questions from the prime model. It asks not only who delivers the final item but also which network sustains the capability under pressure. It asks whether production depends on rare, proprietary chokepoints or can tolerate wider participation. It asks whether maintenance and regeneration can occur through a layered network or must revert to the source. It asks whether industrial scale is concentrated in programme ownership or distributed across a web of production competence.
NATO has taken important steps toward a broader industrial understanding, explicitly naming SMEs in the Vilnius communiqué, committing to reducing cross-border defence trade barriers, and framing civilian industrial capacity as part of collective defence planning in the Washington Declaration. The Washington Summit Declaration states directly that ‘national and collective resilience is an essential basis for credible deterrence and defence,’ and commits Allies to integrating civilian planning and framing preparedness across all hazards and domains as a collective imperative.[14]
Still, its current reforms remain stronger on cooperation with industry than on fully reclassifying which industrial forms are strategically preferable.
Resilience Rhetoric has Outpaced Acquisition Criteria
Another way to state the problem is this: NATO’s rhetoric on resilience is advanced, but its implied acquisition criteria are still incomplete. The Washington Summit Declaration makes resilience central. It describes national and collective resilience as an essential basis for credible deterrence and defence, calls for the integration of civilian planning, and frames preparedness across all hazards and domains as a collective imperative. The U.S. National Defence Industrial Strategy similarly aims to catalyse generational change toward a more robust, resilient, and dynamic modernised defence-industrial ecosystem, organised.[15] Yet once resilience enters acquisition systems, it is often reduced to narrower proxies: larger orders, stable contracts, supplier assurance, critical materials mapping, and stockpile depth. However, the limits of these proxies became evident during the early phase of Western support for Ukraine, when demand for 155mm artillery ammunition surged beyond what existing production lines could rapidly accommodate, despite stable contracts, established suppliers, and prior materials mapping. The bottleneck was not a lack of political will or funding. It was industrial architecture: production concentrated in too few facilities, with limited surge capacity and narrow supplier ecosystems.[16] That is the gap current resilience proxies do not systematically address. The Alliance speaks strategically about resilience while often buying in line with narrower industrial assumptions than that strategy logically demands.
The U.S. and EU: Progress without Doctrine
This is not just a NATO problem in the narrow sense of the headquarters. The wider transatlantic policy environment shows clear convergence toward industrial resilience. The U.S. Department of Defence’s first National Defence Industrial Strategy is notable precisely because it treats industrial questions as strategic rather than merely administrative. It is built around four priorities: resilient supply chains, workforce readiness, flexible acquisition, and economic deterrence.[17] The department has described the strategy as a framework for transforming the existing industrial base into a more robust, resilient, and dynamic ecosystem over the next three to five years.[18]
The European Union’s Act in Support of Ammunition Production (ASAP) regulation similarly exists because policymakers concluded that ammunition and missile supply chains required direct industrial reinforcement. ASAP was adopted to support ammunition production and strengthen the industrial base across relevant supply chains. That move reflected the same realisation visible in NATO and Washington: the old assumption that the market would automatically provide wartime scale is no longer credible.[19]
However, major shifts remain, at least so far, more mature as capacity-building agendas than as fully developed selection doctrines. One reading of this is that sequencing is intentional, that governments must first rebuild industrial muscle before they can meaningfully specify how to use it differently. Though the concern raised here is not that selection doctrine should have preceded capacity rebuilding, but that the two remain insufficiently connected. Without explicit procurement criteria that reward distributed production, rapid regeneration, and reduced industrial fragility, expanded capacity risks replicating the same concentrated, brittle architecture at a greater scale. Western strategic vocabulary has advanced, but procurement logic has only partially caught up.
Not Just an Industrial Gap
Calling this an industrial gap is accurate but incomplete. The stronger claim is that NATO faces an unresolved doctrinal gap because it has not fully fused three realities into one procurement logic.
The first reality is that production capacity matters again at scale; NATO has accepted this. The second is that resilience is not the same as output; it also includes substitution, recoverability, dispersion, and tolerance for disruption. NATO has accepted this in principle, though not yet always in acquisition practice. The third is that battlefield transparency alters the value of design itself. Systems that are difficult to sustain once observed and targeted impose different strategic costs than systems evaluated only in idealised conditions. This third step remains under-institutionalised.
A doctrine gap exists when official strategy begins to describe a world in which the underlying methods of selection, evaluation, and prioritisation are still partly inherited from another world. That is exactly where NATO stands. It is buying for deterrence plans in an environment of industrial war and persistent sensing. However, many of its evaluation habits still descend from an era shaped by expeditionary preference, centralised production, and lower expectations of prolonged attritional regeneration.[20] If the Alliance does not close that gap, it risks a familiar Western error: solving the visible shortage while preserving the deeper architecture that created strategic brittleness in the first place.
Achieving Policy Correction
Fundamental policy correction does not mean abandoning advanced systems or prime contractors. It means adopting a layered procurement doctrine. Some capability categories will always require concentrated excellence, tight control, long qualification cycles, and specialised industrial ecosystems, and NATO should continue to support those. But it should also carve out a protected procurement logic for capability classes whose strategic value lies significantly in their reproducibility, industrial accessibility, and regenerative potential. In other words, the Alliance needs not one acquisition philosophy but at least two: one for concentrated, high-end capabilities and another for resilient, scalable ones. NATO currently operates with 32 national procurement systems, and no formal alliance-wide selection doctrine exists beyond broad interoperability standards. However, nascent movement in this direction is visible; the US NDIS explicitly distinguishes between flexible and traditional acquisition pathways, and the EU’s ASAP regulation implicitly prioritises scalable, widely producible munitions over bespoke systems. The political will to formalise this distinction alliance-wide remains limited, but the analytical case for it is increasingly reflected in official frameworks. The analytical case for it is increasingly reflected in official frameworks. The clearest signals of nascent intent are the NDIS flexible acquisition pathway, the European Commission’s Defence Industrial Strategy (2024), and ASAP’s implicit prioritisation of widely producible munitions over bespoke systems, none of which yet constitute a formal doctrine shift, but each suggests the conceptual ground is being prepared.
That shift would have several implications: First, NATO and national ministries would need to treat manufacturability as a strategic criterion rather than a secondary technical convenience. A design that can be reproduced across a broader qualified industrial base offers wartime value even if it is less elegant in peacetime contracting logic. Second, distributed production potential should be recognised as a resilience asset. If only a tiny number of nodes can produce a capability, it remains a vulnerability, no matter how efficient those nodes are. Third, procurement should assess whether a system’s sustainment logic is sufficiently open to regeneration under pressure. Closed proprietary architectures may make sense for some advanced systems, but they should not be treated as the default virtue for every capability category. Fourth, Alliance-wide standardisation should not be confused with industrial centralisation. The DPD, DPP, and NDPP address standardisation at the interoperability layer, ensuring that Allied systems can operate together effectively.[21] What those frameworks do not systematically address is the production architecture layer: whether the inputs that generate interoperable outputs are themselves concentrated in fragile industrial nodes or distributed across a resilient ecosystem. The goal should be interoperable outputs with a wider range of producible inputs. Fifth, battlefield transparency should influence procurement choices more directly. Even at a high policy level, NATO must recognise that survivability in a sensor-contested environment is not just a tactical employment issue but a design and sustainment issue.
Coordination Is the Key Problem
The Alliance is advancing, it is pushing: standards, multinational cooperation, demand aggregation, trade and investment facilitation, and stronger defence-industrial ties across Europe and North America.
Selection reform must ask which design philosophies get rewarded and which capabilities endure once attrition, observation, supply disruption, and workforce strain are factored in. If procurement continues to reward solutions primarily for their performance on familiar technical metrics while underweighting regenerative value, then NATO may end up with a more coherent process for buying industrially fragile capabilities.
Thus, the doctrine gap matters more than the shortage headline. Shortages can be addressed with money, contracts, and political urgency. Doctrine gaps are harder because they require institutional reclassification. They require ministries and alliances to stop treating manufacturability, distributed production, sustainment openness, and observability as peripheral and to treat them as central aspects of wartime utility.
Conclusion
NATO has already crossed an important threshold. It now recognises that the defence industry is foundational to deterrence and defence. The Vilnius Summit communiqué, the Defence Production Action Plan, the Washington Summit Declaration, the NATO Industrial Capacity Expansion Pledge, the US National Defence Industrial Strategy, and the EU’s ASAP regulation all point in the same broad direction: the transatlantic community has re-entered the age of industrial preparedness.
The deeper problem is that NATO’s strategic vocabulary has advanced faster than its procurement doctrine. The Alliance is increasingly fluent in the language of resilience, industrial cooperation, and supply-chain security. Yet it still too often behaves as though wartime adaptation means merely producing more through the inherited forms of peacetime defence production. Contemporary peer-to-peer conflict, as illuminated by Ukraine, suggests something more demanding. The issue is no longer just how much can be built. It is whether what is built can be regenerated, sustained, dispersed, and employed under conditions of persistent surveillance and industrial strain.
That is why the true gap is doctrinal. NATO still needs a mature framework for selecting capabilities whose value extends beyond performance to include industrial reproducibility, distributed production fitness, sustainment resilience, and compatibility with a transparent battlefield. Until the Alliance aligns what it buys with the conditions under which it expects to fight, industrial policy and operational doctrine will remain only partially connected.
[1] Jack Watling and Nick Reynolds, Meatgrinder: Russian Tactics in the Second Year of Its Invasion of Ukraine (London: Royal United Services Institute, 2023); Michael Kofman and Rob Lee, “Making Sense of Russia’s Offensive Operations in Ukraine,” War on the Rocks, April 26, 2022.
[2] NATO, “Defence Production Action Plan,” endorsed at the Vilnius Summit, July 2023.
[3] NATO, “Washington Summit Declaration,” July 10, 2024, para. 29.
[4] NATO, “NATO Industrial Capacity Expansion Pledge,” July 10, 2024, para. 1, https://www.nato.int/cps/en/natohq/official_texts_227504.htm.
[5] Barry Watts, The Evolution of Precision Strike (Washington, DC: Center for Strategic and Budgetary Assessments, 2013), 1–15; Todd Harrison, Defense Modernization Plans through the 2020s (Washington, DC: Center for Strategic and Budgetary Assessments, 2016), 6–12.
[6] NATO, “Washington Summit Declaration,” July 10, 2024, https://www.nato.int/cps/en/natohq/official_texts_227678.htm.
[7] Mick Ryan, War Transformed (Annapolis: Naval Institute Press, 2022), 88–95; International Institute for Strategic Studies (IISS), The Military Balance 2024 (London: IISS, 2024), 67–72.
[8] Phillips Payson O’Brien, How the War Was Won: Air-Sea Power and Allied Victory in World War II (Cambridge: Cambridge University Press, 2015), 1–18.
[9] David E. Johnson, “The Implications of Persistent Surveillance,” Parameters 52, no. 1 (2022): 5–19.
[10] David E. Johnson, “The Implications of Persistent Surveillance,” Parameters 52, no. 1 (2022): 5–19; Robert O. Work and Greg Grant, Beating the Americans at Their Own Game (Washington, DC: Center for a New American Security, 2019), 12–18.
[11] Hannah Aries, Bastian Giegerich, and Tim Lawrenson, “The Guns of Europe: Defence-industrial Challenges in a Time of War,” Survival 65, no. 3 (June–July 2023): 7–24.
[12] North Atlantic Treaty, April 4, 1949, art. 3, 63 U.N.T.S. 243, https://www.nato.int/cps/en/natolive/official_texts_17120.htm.
[13] NATO, “Vilnius Summit Communiqué,” para. 36.
[14] NATO, “Washington Summit Declaration,” July 10, 2024, para. 46.
[15] U.S. Department of Defense, National Defense Industrial Strategy (Washington, DC: Department of Defense, January 2024), 3.
[16] Marcus Weisgerber, “A Lack of Machine Tools Is Holding Back Ammo Production,” Defense One, March 3, 2023, https://www.defenseone.com/threats/2023/03/us-artillery-production-ukraine-limited-lack-machine-tools-army-official-says/383615.
[17] U.S. Department of Defense, National Defense Industrial Strategy (Washington, DC: Department of Defense, January 2024), 3.
[18] U.S. Department of Defense, National Defense Industrial Strategy (Washington, DC: Department of Defense, January 11, 2024).
[19] European Parliament and Council, Regulation (EU) 2023/1525, Act in Support of Ammunition Production (ASAP), recitals 3–6.
[20] Hannah Aries, Bastian Giegerich, and Tim Lawrenson, “The Guns of Europe: Defence-industrial Challenges in a Time of War,” Survival 65, no. 3 (June–July 2023): 7–24.
[21] NATO, “NATO Defence Planning Process,” https://www.nato.int/cps/ua/natohq/topics_49202.htm.
