Security at what cost? Defence spending, growth, and the fiscal arithmetic

Europe is rewriting its security contract. Russia's full-scale invasion of Ukraine in February 2022 brought the post-Cold War ‘peace dividend’1 to an end, triggering a broad reappraisal of the European defence position. At the North Atlantic Treaty Organization (NATO) Summit in The Hague in 2025, Allies committed to raising defence spending to at least 3.5% of gross domestic product (GDP) for core defence requirements,2 well above the long-standing 2% benchmark. Governments have already begun adjusting their fiscal frameworks. Germany has overhauled its constitutional borrowing rules to free up defence funding, France has proposed further expanding its major multi-year military programming law, and at the European level the activation of national escape clauses for defence spending has provided flexibility. The Security Action for Europe (SAFE) programme creates a common borrowing instrument to support a coordinated expansion in defence investment.

The fiscal implications of this shift are substantial. For the euro area, raising defence spending towards 3.5% of GDP would imply additional annual expenditures in the order of €45 billion until 2035. In many cases, these increases coincide with already elevated public debt levels and competing fiscal demands, making the fiscal arithmetic appear daunting. A defence build-up could be financed through higher deficits in the short run, but debt sustainability ultimately requires a corresponding fiscal adjustment through either expenditure cuts or tax increases.   

A narrow focus on fiscal constraints captures only part of the picture. It abstracts from the macroeconomic effects of defence spending, including its impact on aggregate demand in the short term and on productivity over the medium term, particularly when expenditures are directed towards investment and innovation. These channels can support output and generate additional fiscal revenues, thereby offsetting part of the initial budgetary cost. Historical evidence from major military build-ups confirms that the fiscal impact of defence spending depends critically on its composition, financing, and economic context (see Box 2.1). 

This chapter examines the conditions under which such macroeconomic feedback can reduce the effective fiscal cost of defence spending, with the focus on how much the composition of spending matters and how fiscal financing choices shape the outcome. It combines novel firm-level evidence of productivity spillovers in defence-related sectors with a structural macroeconomic framework to estimate the degree of 'self-financing', defined as the share of additional spending ultimately recouped through higher economic activity and tax revenue (in present value terms). The main finding is that up to 53 cents of every additional euro of defence spending can be offset by macroeconomic feedback, reducing the effective fiscal cost from €1 to 47 cents in the long-term. 

The analysis proceeds as follows: Section 2.2 documents the defence spending gap. Section 2.3 presents firm-level evidence on productivity dynamics in defence-related industries and spillovers to civilian production. Section 2.4 incorporates these findings in a macroeconomic model to estimate self-financing under alternative scenarios. Section 2.5 concludes with policy implications.  

• 1

  The peace dividend refers to the political choice to reduce defence budgets after the Cold War, not an assertion that lower spending generated economic benefits.

• 2

  The Hague Summit commitment totals 5% of GDP: 3.5% to be allocated to core defence requirements and 1.5% to be accounted for across a broader range of activities, including critical infrastructure, civil preparedness, and innovation (NATO, 2025). This chapter focuses on the 3.5% component, which corresponds to the standard defence expenditure definition used in the fiscal analysis.

European defence spending has been structurally low for decades. After the end of the Cold War, NATO members gradually reduced military budgets in what became known as the 'peace dividend'. Among European Union (EU) Member States, for which comparable fiscal data are available from 1995, average defence spending fell from 1.6% of GDP in 1995 to a trough of 1.1% in 2014, the year Russia annexed Crimea and NATO formally adopted its 2% spending guideline. By 2025, all EU NATO members had reached the 2% threshold, a significant turnaround. But the challenge has shifted: most still face substantial gaps relative to the 3.5% core defence target agreed at the Hague Summit (Figure 2.1).3Only the Baltic states and Poland, the countries most directly exposed to the threat on NATO's eastern flank, are close to or exceeding that higher bar.

The issue is not only how much Europe spends on defence, but what it spends the money on. Over 40% of the EU defence budget goes to salaries and one-third to goods and services, with investment accounting for around 20% of total expenditure. Yet, within that investment envelope, research and development (R&D) remains chronically underfunded at roughly 3% of total defence expenditure. This compares to 12% in the United States (US). The consequences of this gap extend well beyond defence capabilities. Limited defence-related R&D investment contributes to Europe’s continued reliance on non-European suppliers for critical high-technology components, constraining both strategic autonomy and the scope for domestic productivity spillovers (Demertzis et al., 2025). 

• 3

  Data prior to 2025 are based on Eurostat/COFOG; 2025 data are based on NATO definitions, which may differ due to differences in accounting methodology including the treatment of military pensions and equipment downpayments. The broad trends are consistent across both sources.

Financial markets have already reacted to Europe’s defence build-up, anticipating sectoral growth. European defence equity indices have surged since Russia's invasion of Ukraine in February 2022, with a further sharp acceleration following the European Council's ReArm Europe announcement in March 2025 (Figure 2.2). This repricing has been broad-based across major listed defence firms, consistent with market expectations of a sustained increase in procurement volumes, which may extend along defence-related supply chains such as engines, electronics, and specialised materials. Whether this expansion translates into broader and more durable economic growth, however, depends on whether higher defence spending generates productivity gains and innovation spillovers beyond the defence sector itself. 

Defence spending can expand productivity beyond the defence sector through supply-chain linkages. A key question is whether the defence sector acts solely as a recipient of public demand or also as a source of innovation and learning for upstream firms. When defence contractors require stringent technological standards, demand customised inputs, and operate at scale, they can induce suppliers to improve production processes, adopt new technologies, and reorganise their operations. Empirical work on multinational firms shows that these vertical relationships generate supplier productivity gains through knowledge transfer, standard-setting, and “learning by supplying” (Javorcik, 2004 and Fons‑Rosen et al., 2017). Historical evidence suggests that similar dynamics can arise in the defence context. During World War II, when surging government demand for aircrafts drove rapid productivity gains at the production-line level (Ilzetzki, 2024), manufacturers improved their methods and reorganised supply chains under capacity pressure.  

The defence industrial base is broader than final weapons production. The analysis draws on a novel dataset for France, Germany, Italy, and Spain – the four largest euro area economies, which together account for around 75% of euro area GDP and the vast majority of European defence production – matching firms with documented defence-related business to Orbis balance-sheet data.4  This approach captures firms that conduct direct business with the public defence sector as their main line of activity, including both pure defence and dual-use firms, rather than focusing solely on weapons and ammunition production. In practice, this includes firms supplying key inputs and services to the defence sector, such as specialised electronics, advanced materials, engineering services, and maintenance and logistics. The resulting sample covers more than 1,300 defence firms and roughly 1.2 million non-defence firms from 2006 to 2022. Despite its relatively small share of total operating revenues (around 1.9% overall and more than 4.5% in manufacturing), the defence sector represents a disproportionately important high-tech segment of the euro area economy, based on the Organisation for Economic Co-operation and Development (OECD) classification of industries by technology intensity (Figure 2.3a). 

• 4

  The data covers both pure defence and dual-use firms, rather than focusing only on final weapons producers. Companies are identified through national defence-industry associations and ministry or contractor sources. Orbis is a comprehensive global database from Moody's (formerly Bureau van Dijk) containing financial and entity-level information on millions of private and public companies.

Defence firms tend to be more productive, more capital-intensive, and pay higher wages. Compared with civilian firms operating in the same industries, defence firms tend to be larger in terms of employment, invest more per worker, and exhibit statistically significant premia of around 40% in total factor productivity (TFP) and 35% in investment intensity, defined as investment as a share of total assets (Figure 2.3b). The defence sector, therefore, stands out as disproportionately composed of firms closer to the upper end of the productivity distribution, rather than a low-productivity enclave. From the perspective of upstream suppliers, exposure to the defence sector can be economically meaningful for overall productivity. This has direct implications for fiscal arithmetic: spending directed towards more capital-intensive and innovation-focused firms is likely to generate materially different macroeconomic returns than expenditure on wages, maintenance, or imported off-the-shelf equipment. 

Upstream suppliers constitute the main channel of productivity spillovers. The defence sector is deeply embedded in domestic supply chains, with around 85% of civilian firms’ intermediate inputs originating within the euro area (Figure 2.4a). The FREMM5 programme, a major Franco-Italian naval frigate initiative, illustrates how defence investment propagates through upstream supply chains. While prime contractors handle design and final assembly, hundreds of specialised suppliers provide key components such as radar systems and electronic modules. To meet demanding defence standards, these suppliers invest in new capital equipment and upgrade their technology and processes, generating productivity gains that extend beyond the defence sector. Consistent with this, the analysis suggests that a 10% increase in total defence investment intensity, weighted by input-output linkages, is associated with an approximately 0.08% increase in TFP of individual upstream civilian suppliers (Figure 2.4b). Although identified at the firm level, the magnitude of these effects implies relevance for aggregate productivity dynamics. The effects are concentrated among suppliers with higher ex‑ante productivity,6 pointing to a relevant role for absorptive capacity. Defence-related demand and technology transfer mainly raise productivity where civilian suppliers already possess the capabilities needed to internalise it. Breaking down total investment, both tangible components, such as machinery, and intangible components, such as R&D, contribute to these spillovers among suppliers with higher ex-ante productivity. 

Spillovers between defence and civilian firms operating within the same sector appear more limited.7 A complementary analysis explores the productivity effects among civilian firms operating in the same sector as defence producers, such as whether military truck programmes improve productivity in civilian vehicle production. The results suggest that, while defence firms invest more intensively and exhibit higher productivity, these gains do not systematically translate into productivity improvements for civilian firms in the same sector. Productivity remains largely firm specific, making within-sector spillovers smaller than those transmitted through upstream supply-chain relationships. 

The overall picture is therefore one of conditional opportunity. The growth impact of higher defence spending depends not only on its scale but on the structure of domestic production and the ability of upstream civilian firms to absorb and diffuse new technologies – conditions that vary considerably across the euro area and that policy can influence. These spillovers do not materialise automatically; they require deliberate choices about procurement design and domestic sourcing. 

• 5

  FREMM is the French abbreviation for Frégate Européenne Multi-Mission or Italian abbreviation for Fregata Europea Multi-Missione. The Franco-Italian family of warships is designed by Naval Group and Fincantieri.

• 6

  Ex-ante high-TFP suppliers are defined as firms with relatively high productivity prior to the increase in defence investment. 

• 7

  See Annex A9 for further robustness exercises and data on whether non-defence firms operating in the same two-digit sector and country as defence firms experience systematic changes in productivity when defence firms’ investment intensity increases.

Building on the firm-level evidence presented in the previous section, this section examines the implications of defence-related spillovers for growth and fiscal sustainability at the euro area level. A two-sector model, comprising a defence sector and a civilian sector, is used to capture how military spending interacts with the broader economy. The model incorporates three key features. First, defence and civilian production8compete for the same workers, capital, and intermediate inputs, so a build-up in one sector affects resource allocation in the other. Second, defence investment generates technological knowledge that diffuses gradually to civilian suppliers, consistent with the supply-chain spillovers documented in Section 2.3. Third, government borrowing affects private demand: households with finite planning horizons do not fully offset higher public debt with additional saving (Blanchard, 1985; Yaari, 1965), so deficit-financed defence spending raises output, but also puts upward pressure on interest rates and partially crowds out private investment. 

Our simulations consider a permanent increase in defence spending by 1.5 percentage points of GDP, phased in over 10 years, broadly consistent with reaching the NATO Hague Summit target by 2035. One-third is allocated to government consumption (personnel) and two-thirds to the investment and procurement of specialised military equipment. 9 This includes policies aimed at maximising the spillover by directly supporting higher investment in the defence sector.10 The build-up is initially debt financed, followed by a gradual fiscal adjustment that returns the debt-to-GDP ratio to its initial level over 20 years.11 This adjustment is assumed to take place through expenditure reallocation rather than tax increases.

Two scenarios are considered. In the first, productivity spillovers are absent, and the dynamics are driven solely by the demand impulse and the associated supply-side adjustments stemming from the reallocation of capital and labour towards the defence sector. In the second, spillovers are calibrated to match the firm-level productivity estimates from Section 2.3, leading to a long-run cumulative productivity gain of around 1.15% in the civilian economy, within the range of estimates in the broader literature on defence R&D spillovers (Moretti et al., 2025). The model assumes a closed euro area economy, so all procurement is treated as domestically sourced – an upper bound that holds if European suppliers capture the bulk of the additional demand. At the country level, procurement from other euro area Member States would shift spillovers geographically rather than eliminate them at the euro area level. 

The model is calibrated to reflect average conditions across the four largest euro area economies (France, Germany, Italy, and Spain). This aggregation abstracts from cross-country differences in defence sector size, civil-military integration, and cross-border spillovers, but is designed to capture the core mechanisms linking defence spending, productivity, and fiscal dynamics.  

Importantly, productivity gains are not confined to larger economies: what matters is the relative importance of defence-related activity and its integration into supply chains. In principle, the analysis is most directly applicable to economies with an established defence industrial base. However, smaller economies can benefit by specialising in upstream segments of European production networks, where stronger defence demand supports investment and technological upgrading. In this context, exports of intermediate goods can partly offset imports of final equipment. Productivity gains are likely to be more broadly shared across the euro area depending on national industrial structures and on deliberate policy choices to integrate into European defence supply chains. 

2.4.1 Growth and multipliers

The output response to the defence build-up reflects the interaction of short-run demand effects and longer-run supply-side adjustments. In the near term, higher government spending raises output above potential as stronger demand translates into higher production and employment. The initial expansion is positive but moderate. In the absence of productivity spillovers, the fiscal multiplier.12 peaks at about 0.7 after two years and remains broadly stable over the build-up period. With spillovers, it rises to around 1.0 after two years and keeps increasing gradually. 

As the initial demand impulse fades, supply-side adjustments drive the dynamics. The build-up of the defence sector increases demand for labour and capital, leading to a reallocation of resources away from civilian production. While employment rises overall, this reallocation constrains the expansion of the civilian sector, 13 limiting the output gains in the absence of productivity spillovers. Output remains durably above its initial level, but the long-term multiplier stays below 1.0, implying a reduction in private consumption.

Productivity spillovers can drive the multiplier above 1.0. As defence-related investment expands, technological knowledge diffuses along supply chains, raising TFP in the civilian sector, consistent with the mechanisms discussed in Section 2.3. This relaxes supply constraints and allows both the defence and civilian sectors to expand simultaneously. Output growth is then supported not only by higher input use but also by more efficient production, which in turn sustains a stronger expansion in capital. In this scenario, GDP rises to nearly 2.5% above baseline after two decades (Figure 2.5b). The cumulative multiplier increases to around 1.5 over longer horizons, compared with about 0.8 without spillovers (Figure 2.5a).  

• 8

  Firms in the defence sector produce military goods procured by the government, while firms in the civilian sector produce consumption goods and services.

• 9

  Investment and procurement of specialised military equipment include intermediate consumption and currently accounts for roughly 50% of defence spending in the four largest euro area economies (France, Germany, Italy, and Spain). New national defence plans point to a combined investment and intermediate consumption share of around 80%, with personnel accounting for only 15% (Checherita-Westphal et al., 2025), suggesting that our two-thirds assumption reflects a conservative spending composition.

• 10

  In the model, these policies are captured in a reduced form through an investment subsidy that increases capital deepening in the defence sector. In the technical annex A10 an estimate of the self-financing with and without the subsidy is provided. 

• 11

  In the model, these policies are captured in a reduced form through an investment subsidy that increases capital deepening in the defence sector. Annex A10 provides an estimate of the self-financing with and without the subsidy. 

• 12

  The fiscal multiplier is defined here as the cumulative discounted increase in output relative to steady state divided by the cumulative discounted increase in spending.

• 13

  The strength of this response would depend on initial labour market conditions. The expansion is larger and the demand impulse stronger where slack exists and participation rates have room to rise. In economies near full employment, the response would be more muted and wage pressure more pronounced, consistent with evidence that fiscal multipliers vary with the state of the cycle (Ramey, 2011; Ilzetzki, 2025).

Estimates of multipliers in the literature span a wide range, underscoring their conditional nature. The multipliers generated by our model are broadly consistent with recent European Central Bank14 and European Commission studies. Suites of models to assess military ramp-ups typically find values below 1.0 in the short to medium term. This is in line with empirical evidence available for the US, where short-run defence spending multipliers are generally estimated between 0.4 and 1.0 (Hall, 2009; Barro and Redlick, 2011; Ramey, 2011; Ramey and Zubairy, 2018; Antolín-Díaz and Surico, 2025). By contrast, our estimates are more conservative than several recent euro area-focused empirical studies reporting multipliers in the range of 1.5 to 2.0 (Furceri et al., 2026; García-Serrador et al., 2025; Ben Zeev and Pappa, 2017). The authors suggest that higher multipliers could be driven either by the presence of economic slack in the sample or by the weaker responsiveness of monetary policy in a monetary union, where area-wide interest rates are less sensitive to country-specific developments. In our framework, the transmission of fiscal policy operates through a relatively strong interest rate response, which implies more pronounced crowding-out effects on private investment and, therefore, more moderate short-run multipliers. At the same time, the higher long-run multipliers implied by calibrating productivity spillovers to microeconomic evidence are consistent with US estimates in Antolín-Díaz and Surico (2025), who report values between 1.7 and 2.0, and with the findings of Moretti et al. (2025) on R&D spillovers.15 

2.4.2  The fiscal arithmetic 

How much of the defence build-up can pay for itself? Debt sustainability requires that a permanent increase in defence spending is ultimately matched by either higher revenues or lower spending elsewhere. However, the necessary fiscal adjustment may be smaller than the headline increase in defence spending suggests. Macroeconomic feedback effects can improve fiscal outcomes, allowing part of the additional spending to be financed automatically, without the need for future consolidation.  

This 'self-financing' operates primarily through the expansion of the tax base. Stronger economic activity raises revenues without changes in tax rates, while higher employment and income reduce transfer payments. A second channel works through valuation effects, which reduce the real burden of outstanding public debt: higher economic activity tends to raise inflation in the short term while the maturity structure of debt delays the pass-through of higher interest rates to borrowing costs.16

Automatic fiscal recovery is measured through the self-financing ratio. While the multiplier captures how much output each euro of defence spending generates, the self-financing ratio measures how much of that spending is automatically recouped through the resulting increase in economic activity, without any change in tax rates. It is defined as the present value of all future automatic fiscal gains expressed as a share of the spending increase (Angeletos, Lian, and Wolf, 2023). A self-financing ratio of 25% implies that 25 cents of every euro of additional defence spending is ultimately recovered, reducing the net fiscal cost to 75 cents. The remaining 75 cents corresponds to the fiscal adjustment still required through expenditure or revenue measures, but it is significantly smaller than the initial outlay.   

Three key factors determine the extent of self-financing: the presence of productivity spillovers to civilian sectors, the composition of any additional fiscal adjustment, and whether public debt is stabilised or allowed to rise persistently. The following subsections examine each of these dimensions in turn. 

Defence productivity spillovers amplify the fiscal payback 

Productivity spillovers can bring the self-financing ratio to 53%. In the absence of spillovers, governments recover roughly 25 cents per euro of defence spending, primarily through a higher tax base driven by employment and capital accumulation. This payback is relatively stable over time, as the output expansion is limited by the reallocation of resources away from civilian production, keeping the multiplier broadly constant. When productivity spillovers to civilian suppliers are calibrated to empirical evidence, the ratio more than doubles, with 53 cents recovered per euro spent. In net terms, the effective fiscal cost of a €1 increase in defence spending falls from 75 cents to 47 cents (Figure 2.6). The additional recovery reflects a structural change in the economy that unfolds over time: when spillovers materialise and defence and civilian production expand together, almost doubling the output gains and generating multipliers that rise over time. The share of spending recovered therefore increases gradually over the long run, reflecting the permanent improvement in the economy's productive capacity.  

• 14

  European Central Bank Economic Bulletin, Issue 6/2025.

• 15

  Assuming fully Ricardian households in our setting would give higher multipliers (see technical annex), but this would imply counterfactual marginal propensity to consume.

• 16

  Assuming fully Ricardian households in our setting would give higher multipliers (see technical annex), but this would imply counterfactual marginal propensity to consume. 

Labour income expansion emerges as the dominant fiscal channel. Across both scenarios, most of the additional tax revenue comes from higher employment, defined as an increase in hours worked, which expands the wage bill. Capital income taxes and automatic stabilisers contribute but play a secondary role (Figure 2.7a). With productivity spillovers, gains in value added propagate through supply chains, reinforcing all fiscal channels. Labour income alone accounts for around 40 percentage points of the 53% self-financing ratio (Figure 2.7b), as productivity gains raise real wages, expanding employment income. Inflation effects remain limited and valuation effects on public debt are negligible.   

These gains depend critically on the composition of defence spending. Productivity effects are driven by investments in applied research, as well as the domestic development and retention of high-technology production, particularly dual-use technologies. Accordingly, the 53% self-financing estimate should be interpreted as contingent on a strong domestic supply chain and innovation-intensive procurement. If, instead, defence investment relies heavily on off-the-shelf imports, as has often been the case in the past, the productivity channel weakens and self-financing is likely to revert to, or even fall below the 25% baseline. 

Financing composition: spending reallocation or tax increases? 

How governments finance the increase in defence spending is as important as the size of the increase itself. Under expenditure-based adjustment (our baseline) the self-financing ratio is broadly stable and the overall effect on GDP remains positive. By contrast, financing through distortionary taxation – particularly labour taxes17 – substantially weakens the fiscal feedback. As the share of adjustment relying on labour taxation increases, self-financing declines sharply and can turn negative at high tax rates (Figure 2.7). The mechanism is straightforward: higher labour taxes reduce the after-tax return on work, weakening labour supply and compressing the tax base. Revenue gains are offset, or even reversed, by the contraction in economic activity (Furceri et al., 2026; Calò et al., 2026; Boullot et al., 2026). Importantly, the composition of future fiscal adjustment has ex-ante implications for the behaviour of economic agents, affecting labour supply, savings, and investment decisions already at the time expectations are formed.  

• 17

  The tax financing experiment assumes a permanently higher labour income tax rate. This is modelled as a shift in the steady-state tax rate, with the fiscal rule otherwise unchanged.

The cost of permanently higher debt 

Debt dynamics further shape the long-term fiscal cost of a defence build-up. Higher government borrowing affects the economy because households and firms do not fully offset this expansion. As public debt competes for a limited pool of savings, it puts upward pressure on real interest rates18 and crowds out private investment. In the baseline, this effect is temporary as the fiscal adjustment path brings debt to GDP back to its initial level, and interest rates revert. By contrast, allowing debt to remain permanently higher – equivalent to delaying part of the fiscal adjustment indefinitely – leads to persistently higher real interest rates (Figure 2.9), crowding out private investment and compressing non-defence expenditure as interest payments absorb an increasing share of fiscal resources. In net terms, the same defence build-up costs about 5% more if debt is stabilised at a higher level; the cost could be substantially higher if instead of following a fiscal rule, the debt is allowed to drift for longer.  

The choice is therefore not just about how much to spend on defence, but within what fiscal framework. The same build-up costs materially more in net fiscal terms when debt is allowed to drift than when it is stabilised through a credible medium-term adjustment path. Countries that couple a defence build-up with multi-year fiscal plans and enforceable fiscal rules capture more of the growth dividend than those that do not. The defence build-up and fiscal responsibility are mutually reinforcing, not competing priorities.

• 18

  The result follows from the Overlapping Generations structure of the model, which implies a break of Ricardian equivalence. 

The fiscal cost of Europe’s defence build-up is real but not automatic. It depends on the macroeconomic feedback, the composition of defence procurement, and the way governments manage the associated fiscal adjustment. This chapter shows that the effective budgetary cost of higher defence spending can be substantially lower than the headline figures suggest, provided spending is well-designed and embedded in a credible fiscal framework. In the short term, deficit-funded defence spending has a positive multiplier, leading to higher GDP, which generates tax revenues that partially offset the deficit. The key long-term transmission channel operates through technology: when defence investment raises the technological frontier, productivity gains can diffuse along supply chains, increasing output and broadening the tax base. This generates revenues and reduces the net fiscal cost. However, these long-term gains are not automatic and depend on policy choices, such as procurement design, supply-chain integration, and European-level governance.  

Four policy lessons emerge from the analysis:

First, well-targeted defence spending can generate meaningful fiscal payback, notably in the long run through growth. Higher defence expenditure raises output, employment, and tax revenues through demand effects and the compositional shift towards a higher share of defence-related activity. These effects can be significantly larger when procurement is directed towards capital – and R&D-intensive activities and euro area suppliers. In such cases, productivity spillovers to upstream civilian firms can bring self-financing up to 53 cents per euro spent. This payback materialises over the long term as productivity gains diffuse through the economy. Defence spending is therefore not merely a fiscal burden; it is a potential source of economy-wide productivity gains. This is not to say that it is the only or the most productive form of government investment. It should be driven first and foremost by security necessity, but efforts to maximise economic spillovers are a substantive side effect.  

Second, composition matters as much as scale. Spending concentrated on personnel, maintenance, or imported equipment delivers much smaller macroeconomic returns than spending directed towards technologically advanced European firms embedded in integrated supply chains. Fragmentation of the European defence industrial base limits not only strategic autonomy but also fiscal efficiency, making deeper integration of European defence markets both a strategic and fiscal priority (Mejino-Lopez and Wolff, 2025).  

Third, fiscal design is decisive. Financing the defence build-up through expenditure reallocation rather than tax increases avoids the additional economic burden of higher taxation and preserves fiscal payback. Heavy reliance on distortionary tax increases, particularly on labour, discourages work, eroding the tax base and weakening fiscal payback, potentially turning self-financing negative at sufficiently high tax rates.  

Fourth, defence spending increases should go hand in hand with a greater anchoring of fiscal responsibility. Allowing debt to drift permanently higher could raise borrowing costs and eventually lead to compressed non-defence expenditure as interest payments absorb an increasing share of resources. It may also amplify sovereign risk and financial stability vulnerabilities. By contrast, countries that couple a defence build-up with credible medium-term debt stabilisation could capture more of the growth dividend as the credible fiscal anchor limits borrowing cost increases. 

Greater coordination at the European level in defence markets could further amplify the benefits of positive growth.19 Defence capabilities generate cross-border security benefits – a European public good justifying more security cooperation. Lack of cooperation and full reliance on nationally organised defence spending risks burden shifting and the under-provision of defence, since the cross-border nature of security benefits creates incentives to free-ride on others' spending. Here the vicinity to a country posing a security threat matters a lot. But defence-related supply-chain spillovers can also extend beyond national borders, creating positive growth dynamics. Defence spending exclusively directed at national suppliers perpetuates the inefficient fragmentation of the defence market, raising costs and limiting scale, particularly when overall expenditure increases. It may also mean that positive productivity spillovers are missed, as innovative firms may remain constrained in their relatively small market instead of deeply integrating with firms across all of Europe. Pooling procurement and opening defence markets to competition would allow EU Member States to place larger combined orders, enabling longer production runs and substantially reducing unit costs (Wolff, Steinbach, and Zettelmeyer, 2025). 

European coordination is essential to ensure that benefits reach all EU Member States. Countries with an established defence industrial base are best placed to activate productivity gains directly. However, smaller economies can benefit by specialising in upstream segments of European supply chains where they hold comparative advantages. Participation in integrated production networks allows these countries to gain both directly, through supplying advanced intermediate goods, and indirectly, by benefiting from stronger aggregate euro area growth. Building on the universe of European firms can also accelerate the production cycle. 

Realising these gains requires a European defence industrial policy that integrates supply chains and directs procurement towards innovation. Reducing national fragmentation, increasing competition, and expanding the role of innovative firms are essential steps to strengthening the technological dynamism that underpins productivity spillovers (Kapstein, Ospital, and Wolff, 2026). Procurement requirements explicitly targeting advanced, innovation-intensive capabilities, including dual-use and artificial intelligence-enabled technologies, maximise the growth dividend from higher defence expenditure. Joint R&D programmes extending across borders can generate scale effects that no single EU Member State can achieve alone, as illustrated by the European Defence Fund. Common procurement frameworks such as the European Defence Industry Programme can further strengthen supply chains and reduce unit costs. Common financing instruments such as SAFE can ease national fiscal constraints and reinforce the growth dividend from higher defence expenditure. In this setting, European coordination carries both strategic and fiscal benefits. 

Taken together, these findings reframe the fiscal debate around Europe’s defence build-up. Done well, the effective cost of meeting NATO's 3.5% target is considerably lower than the headline numbers suggest. Done poorly, the macroeconomic channels that could make higher defence spending more fiscally sustainable are largely foregone. The design of Europe's defence build-up is therefore not a second-order implementation detail, but rather one of the central fiscal policy questions of this decade.  

• 19

  A public good is non-excludable and non-rival in consumption. Defence meets both criteria at the European level: security benefits cannot easily be confined to the country providing them, and one country's security does not diminish another's. For the case for treating defence as a European public good, see Beetsma, Buti, and Nicoli (2024).

Abelshauser, W. (2004). Die deutsche Wirtschaft 1945–1989. Munich: C.H. Beck. 

Angeletos, G.-M., C. Lian, and C.K. Wolf (2023). Can Deficits Finance Themselves? NBER Working Paper 31185. 

Antolín‑Díaz, J. and P. Surico (2025). The long‑run effects of government spending, American Economic Review, 115(7), 2376-2413. 

Baker, S. R., N. Bloom, and S.J. Davis (2016). Measuring economic policy uncertainty. The Quarterly Journal of Economics, 131(4), 1593-1636. 

Barro, R. J. and C.J. Redlick (2011). Macroeconomic effects from government purchases and taxes. The Quarterly Journal of Economics, 126(1), 51–102. 

Beetsma, R., M. Buti, and F. Nicoli (2024). Defence as a European Public Good: Delivery and Financing, EconPol Forum. 

Ben Zeev, N. and E. Pappa (2017). Chronicle of a war foretold: The macroeconomic effects of anticipated defence spending shocks. The Economic Journal, 127(603), 1568-1597. 

Blanchard, O.J. (1985). Debt, deficits, and finite horizons, Journal of Political Economy, 93(2), 223-247. 

Boullot, M., C. Cahn, E. Challe, and J. Matheron (2026). Aggregate and Distributional Implications of a Military Buildup, Banque de France Working Paper No. 1039. 

Calò, S., R. Kraemer, and F. Tomasone (2026). The Innovation Channel of Fiscal Space, ESM Working Paper No. 75, February 2026. 

Checherita-Westphal, C., C. Huber, M. Rodríguez-Vives, and G. Müller (2025). European Central Bank Economic Bulletin, Issue 5/2025, Box 7. 

Eichengreen, B. (2007). The European Economy since 1945: Coordinated Capitalism and Beyond. Princeton University Press. 

European Commission (2025). White paper for European defence readiness 2030, JOIN (2025) 120 final, 19 March 2025. 

European Commission (2025). 2026 European macroeconomic report: Underpinning resilience in a geoeconomic landscape, European Economy Institutional Paper No. 310, November 2025. 

Demertzis, M., A. Fiorito, and K. Panitsas (2025). Strategic autonomy and European competitiveness: Security now comes first. European Parliament, PE 764.371. 

Fabrizio, K. R. and D.C. Mowery (2007). The federal role in financing major innovations: Information technology during the postwar period. Research Policy, 36(4), 449–469. 

Field, A. J. (2003). The most technologically progressive decade of the century. American Economic Review Papers and Proceedings, 93(4), 1399–1413. 

Fons‑Rosen, C., S. Kalemli‑Ozcan, B.E. Sørensen, C. Villegas‑Sanchez, and V. Volosovych (2017). Foreign investment and domestic productivity: Identifying knowledge spillovers and competition effects, NBER Working Paper No. 23643. 

Furceri, D., P. Juarros, S. Mishra, A.D.M. Nguyen, A.S. Pessoa, and A. Sollaci (2026). Macroeconomic impacts of EU defene spending, IMF Working Paper No. 26/53. 

García‑Serrador, A., D. Sarasa, and C.A. Ulloa (2025). Buy guns or buy roses? EU defence spending fiscal multipliers, SUERF Policy Brief No. 1209. 

Ilzetzki, E. (2024). Learning by Necessity: Government Demand, Capacity Constraints, and Productivity Growth, American Economic Review, 114(8), 2436-2471. 

Ilzetzki, E. (2025). Guns and growth: The economic consequences of defence buildups, Kiel Report No. 2. 

Javorcik, B. S. (2004). Does Foreign Direct Investment Increase the Productivity of Domestic Firms? In Search of Spillovers through Backward Linkages. American Economic Review, 94 (3), 605-627. 

Kapstein, E., J. Ospital, and G.B. Wolff (2026). Reforming European defence procurement to boost military innovation and startups, Bruegel Policy Brief 04/2026. 

Mejino‑Lopez, J. and G.B. Wolff (2025). Boosting the European defence industry in a hostile world, Intereconomics, 60(1), 34-39. 

Moretti, E., C. Steinwender and J. Van Reenen (2025). The intellectual spoils of war? Defence R&D, productivity, and international spillovers, Review of Economics and Statistics, forthcoming. 

Mowery, D. (2010). Military R&D and innovation, in B. Hall and N. Rosenberg (eds.), Handbook of the Economics of Innovation, Vol. 2, 1219-1256. 

Ramey, V.A. (2011). Identifying government spending shocks: It’s all in the timing, Quarterly Journal of Economics, 126(1), 1-50. 

Rockoff, H. (1984). Drastic Measures: A History of Wage and Price Controls in the United States. Cambridge University Press. 

Ruttan, V. W. (2006). Is War Necessary for Economic Growth? Military Procurement and Technology Development. Oxford University Press. 

Wolff, G.B. (2025). Defending Europe without the US: First estimates of what is needed, Bruegel. 

Wolff, G.B., A. Steinbach, and J. Zettelmeyer (2025). Financing and governing European rearmament: The case for a unified procurement fund, Bruegel. 

Yaari, M.E. (1965). Uncertain lifetime, life insurance, and the theory of the consumer, The Review of Economic Studies, 32(2), 137-150.