The Economic Impact of Arms Spending in Germany, Italy, and Spain

1 Introduction

European countries are on a road to militarisation. In the last 10 years, military expenditures of NATO EU countries (according to NATO definitions and data) have increased by almost 50 %, from €145 billion in 2014 to €215 billion in 2023 (measured in constant 2015 prices). This total is greater than the annual GDP of a country such as Portugal. With the war in Ukraine, 2023 outlays are expected to increase by almost 10 % in real terms over the previous year. NATO EU countries as a whole now spend 1.8 % of GDP on their militaries, close to the 2 % target set by the US and NATO.

In this context, it is important to assess the economic consequences that the current increase in military spending is likely to have on Europe’s economies. In this article we focus on Germany, Italy and Spain, the three largest EU countries – excluding the nuclear power status of France – and we concentrate on arms acquisitions. Over a decade, Germany has increased its real military spending by 42 %, Italy by 30 %, Spain by 50 %. In all countries, this expansion has been entirely due to higher acquisitions of arms and equipment. Such a rise in military expenditure and arms procurement contrasts starkly with the stagnation of EU economies. In the aggregate of NATO EU countries, between 2013 and 2023, real GDP has increased by 12 % (just over 1 % per year on average), total employment by 9 %, and military expenditures by 46 %, four times faster than national income. The picture in the area of new investment is even more dramatic: while capital formation has risen by 21 %, arms acquisitions have increased by 168 % – eight times as fast – in NATO EU countries. In Germany, Italy and Spain, the disparities in growth rates are broadly similar. Arms are absorbing a rapidly increasing proportion of the resources that countries devote to new production capabilities, new technologies, and new infrastructures.

At a time of concerns about public finances, such a rise in military spending comes at the expense of other types of public expenditures. In this article we investigate the economic effect of military expenditure on growth and employment and we compare it to the impact of public expenditures for education, health and the environment. We use an input–output methodology – already adopted by several studies – to assess the relevance of imports and demand towards different sectors providing intermediate inputs. We can therefore assess the likely impact on output and jobs of one billion euros of extra spending in arms, and compare it to the outcomes of the same amount spent in education, health and the environment. Our findings show that for all countries non-military public expenditures have a greater impact on the economy and employment than spending for arms acquisition.

The article is structured as follows. Section 2 assesses the state of the art on the economic impact of arms spending; Section 3 presents data, patterns and the methodology used; Section 4 shows the results; Section 5 concludes.

2 The State of the Art on the Economic Impact of Military Expenditure

Military expenditures are the outcome of the military and political strategies, and economic processes. They represent the quantity of economic resources that a country’s government allocates to the national military system: the armed forces, the acquisition of armaments, military infrastructures, and the implementation of military operations.

Their military and political drivers include security and foreign policy objectives, military alliances and external threats.^[1] In the economic system, military expenditures are funded by tax revenue or government debt and compete with other public expenditures – for education, health, welfare, research, the environment, etc. Government policies define the relative importance of the military as opposed to other economic, social or environmental priorities, shaping a country’s growth trajectory. Military expenditures create demand for arms and supplies produced by corporations and fund the research, development and exports of armaments. Profits in arms production are usually higher than average and a country’s ‘military industrial complex’ is often a major force driving the growth of military expenditures (Caruso 2017; Nascia and Pianta 2009).

The debate regarding the impact of military expenditures on growth and employment reflects their contradictory nature – a potential to expand the economy on the one side, and on the other side, a subtraction of resources to consumption and investment.

On the one hand, the argument for a positive effect of military expenditure on output growth rests on the Keynesian principle of effective demand and the Marxian “underconsumption” problem. In this view, military expenditure may boost capacity utilization and profits, compensating for capitalism’s tendency towards stagnation (Baran and Sweezy 1968). At the same time, military expenditure may also have positive supply-side effects by supporting R&D, contributing to the development of new technologies, enhancing the skills of the labour force, as in the case of the US since the Second World War (Dunne and Tian 2013; Dunne, Smith, and Willenbockel 2005; MacNair, Murdoch, and Sandler 1995).

On the other hand, for a given level of productive capacity, an increase in military expenditure may have negative effects on growth. On the demand side, there may be a crowding-out of productive investment and other types of public expenditure. On the supply side, military expenditures may absorb a significant part of a country’s limited capabilities in research, technology, human skills, capital accumulation and finance. In the case of the US, this has led to business practices that have inflated costs, prices and profits, and reduced efficiency (Markusen 1986; Melman 1988).

Authors in the Marxist tradition have pointed out the contradictory role of arms spending in world capitalism. Military expenditures are required for ensuring military superiority and hegemonic power but, at the same time, their expansion imposes substantial costs in terms of lower capital accumulation and slower economic growth as it diverts resources from productive uses (Smith 1977).

Empirical investigations have produced mixed results. Survey of studies of the impact of military spending have pointed out that findings are highly dependent on the theoretical approach, the nature of the countries analysed, the time horizon, and the methodology used (Cepparulo and Pasimeni 2024; Dunne 1996).

Dunne and Uye (2010), reviewing 103 studies, found that about 45 % of cross-country and case studies show an unclear association between military expenditure and economic growth; 39 % of cross-country studies and 35 % of case studies find a negative effect of military spending on economic growth, with only about 20 % finding a positive effect. Dunne and Tian (2013, 2016 updated the survey, including developing countries, and confirmed that it is more likely for the empirical evidence to show negative or at most no statistically significant effects of military expenditure. When focussing on post-cold war data, most cross-country studies show a negative impact of military spending on economic growth. Yesilyurt and Yesilyurt (2019) conducted a meta-analysis of empirical studies using the share of military expenditure on GDP as independent variable and found evidence of non-statistically significant effects. In their meta-analysis of 48 studies, Churchill and Yew (2018) supported the existence of a negative military expenditure-growth association. In their meta-analysis of 32 empirical studies, Alptekin and Levine (2012) found a positive effect of the share of military expenditure in GDP on economic growth for developed countries and an unclear association in developing countries and in the sample as a whole. Sandler and Hartley (1995) argued that works focussing only on supply-side effects tend to find positive or insignificant effects, while models allowing for crowding-out effects of investment on the demand side are more likely to find negative effects.

D’Agostino, Dunne, and Pieroni (2018) and Saeed (2023) addressed the endogeneity problem arising from reverse causality and concluded that an increase in military expenditure leads to a statistically significant reduction in GDP growth; the results are robust to the use of different instrumental variables, the application of different estimators, different time periods, and different cross-section compositions.

The prospects of disarmament opened up by the end of the Cold War gave rise to literature on the “peace dividend”, i.e. the economic benefits of cuts in military expenditures (Gleditsch et al. 1996). Rather than looking at growth effects, this literature focused on the short-run output and employment effects of shifting public resources from the defence sector to alternative types of public spending (Dunne and Tian 2013; 2016). For the case of UK, Dunne and Smith (1984) found that a reduction of the share of military expenditure in GDP by 1.5 percentage points, matched by a proportional increase in the non-military components of public expenditure, would lead to a net employment increase of about 100,000 jobs; Barker, Dunne, and Smith (1991) showed that halving the military budget and relocating the released “peace dividend” to the civilian sectors would have resulted in significant output and employment increases.

The role of changing security threats and military alliances in shaping arms spending has been investigated by several studies. Dunne and Smith (2020) accounted for internal and external threats among the determinants of military expenditure and identify different possible configurations of the relationship between arms spending and growth according to world regions and historical periods. Caruso and Di Domizio (2016) found a positive association between US and European military spending, driven by the dynamics of the NATO alliance. The composition of military expenditure – armed forces, arms acquisition, etc. – may have an impact on aggregate growth (Becker and Dunne 2023) and is significantly affected by government ideology (Olejnik 2024) and by the strategic role of a country within a military alliance (Bove and Cavatorta 2012). The expansion of military expenditure is also positively associated with economic inequality (Biscione and Caruso 2019) and may contribute to authoritarian turns in a country’s institutions, political culture, and society, with a possible erosion of democracy (Galtung 1985; Thorpe 2014).

A focus on the economic impact of different types of public expenditures has come from the literature on fiscal multipliers. For the case of the US, Auerbach and Gorodnichenko (2012) found a positive multiplier effect of defence spending, which is generally higher than that for consumption spending but lower than that for non-defence and investment expenditures. Deleidi and Mazzucato (2021) investigated US defence R&D spending as a proxy for “mission oriented” programmes, finding – consistently with Keynesian insights – that the multiplier is higher in periods of low growth or recessions than in periods of expansion (Arin, Koray, and Spagnolo 2015; Auerbach, Gorodnichenko, and Murphy 2018; Deleidi, Lafrate, and Levrero 2023).

In the case of Europe, Saccone et al. (2022) assessed the impact of public nvestment in 31 countries over the 1995–2019 period using the local projection method and found a strong multiplier effect of aggregate spending on output. In the breakdown by type of public activities, they find that investment in education, public order and general public services have the highest multipliers, while defence investment turns out with a negative sign. This can be related to the specificity of military investment in recent years and to Europe’s reliance on arms imports.

A different method is offered by the input–output (I–O) approach, which looks at the intersectoral flows of goods and services of an economy. I–O tables provide information on the input requirements in a given industry and allow estimation of the output and employment impact that an increase in production in one industry has on all other sectors.

Wassily Leontief, the inventor of the input–output approach, carried out the first study on the economic impact of demilitarizing the US economy, showing that a shift of resources from the defence to non-defence sectors would have doubled the number of jobs (Leontief and Hoffenberg 1961). Medoff (1993) used the input–output model for the U.S. economy to estimate the employment impact of different variables – consumption, private investment, state expenditure, military spending – finding that personal consumption and defence expenditures have the lowest effects on jobs and wage increases. Anderson, Bischak, and Oden (1991) rely on an augmented model that integrates the I–O methodology with other statistical techniques. In a scenario of gradual reduction in military spending, starting with a cut of $35 billion in 1990 up to a $105 billion cut in 1994, they find that the U.S. economy would gain 477,000 additional jobs. Pollin and Garriet-Peltier (2009) showed that spending $1 billion on personal consumption, health care, education, mass transit, and construction for home weatherization and infrastructure repair will all create more jobs than devoting the same amount of public resources to military spending. Peltier (2017, 2019, 2023 found similar results when considering other types of non-military expenditure (e.g. investment in wind and solar energy).

Input–output analysis refers to real quantities in the economy and is highly appropriate to assess the short-run economic impact of military and non-military spending.^[2] Most I–O studies have focused on the US – where the Bureau of Economic Analysis produces I–O matrixes including data for the defence sector – and less attention has been devoted to European countries (Gentilucci 2010; Scandizzo, Nardone, and Ferrarese 2015). This article aims to fill this gap, using the I–O methodology to explore the economic impact of arms spending in three major EU countries – Germany, Italy and Spain.

3 Military Spending in Europe: Data, Patterns and Methodology

3.1 Data

NATO EU countries have significantly increased their levels of military expenditures over the last 10 years, both in absolute terms and as a share of GDP.^[3] The increase in military expenditures has been associated with a decline in the share of personnel and a parallel increase in the share of arms and equipment (Becker et al. 2024; Bonaiuti et al. 2023).

Figure 1 shows the evolution of total military expenditures and outlays for arms, equipment, operations and research for Germany, Italy and Spain. From 2013 to 2023, Germany increased military spending from €36 billion to €51 billion (+42 %), Italy from €20 billion to €26 billion (+30 %), Spain from €10 billion to €15 billion (+50 %).^[4] In all countries, the increase has been entirely accounted for by higher expenditures on arms and equipment. Germany tripled its spending on arms and equipment from €4.5 billion in 2013 to €13 billion in 2023; Italy went from €2.5 billion to €5.9 billion; Spain raised its outlays from €1.2 billion to €4.3 billion. NATO EU countries as a whole increased their expenditures for arms and equipment from €24.1 billion in 2013 to €64.6 billion in 2023, with an increase of 168 %.

Figure 1:

Total military expenditure and military expenditure on arms, equipment, operations and R&D.

In decade of slow economic growth, the increase in military expenditures has meant a rise in their share of GDP, as shown in Figure 2. For NATO EU countries, the percentage went from 1.4 % in 2013 to 1.8 % in 2023. In 2023, Germany was at 1.56 %, Italy at 1.45 % and Spain at 1.25 %. The US has long asked NATO EU countries to reach the 2 % of GDP target for military expenditures; after the start of the Ukraine war, Europe is getting close to such levels.

Figure 2:

Military expenditure (percentage of GDP).

The international context for such patterns is provided by SIPRI data (based on a slightly different definition). From 2014 to 2022, military expenditures as a share of GDP decreased in the United States from 3.7 % to 3.4 %, in China from 1.7 % to 1.6 %, and in India from 2.5 % to 2.4 %, also due to their strong GDP growth over the decade. In Europe, Poland is the largest spender, moving from 1.9 % to 2.4 %, with a plan to reach 3 % in a few years; facing the war, Ukraine experienced a jump from 2.2 % to 33.5 % of GDP devoted to military activities.

It should be pointed out that over this decade, Italy and Spain were experiencing difficult situations in their public finances with strict European constraints on government deficits and debt. In spite of this, military budgets and arms procurement were able to increase at an unprecedented pace, further reducing the space for social and environmental public expenditures.

Arms procurement has the nature of capital investment as it is part of a country’s gross fixed capital formation. An additional investigation can then compare the investment dynamics in different areas of public expenditures. Figure 3 shows the percentage changes (in real terms) in public investment in arms acquisition, environmental protection, education, and health for the 2013–2023 period.^[5]

Figure 3:

Arms expenditures versus investment in the environment, education and health percentage change in real terms, 2013–2023.

In NATO EU countries, total government capital investment increased by 35 % over the period, with education showing moderately lower (+24 %) and health showing moderately higher (+45 %) patterns; remarkably, investment in environmental protection fell in real terms by 5 % during the decade. At the same time, arms procurement increased by 168 %, 4.8 times faster than total public investment.

Germany shows a 67 % increase in public investment, similar to the expansion of health capital expenditures (+78 %), while environmental investment did not grow. In contrast, arms procurement has increased by 184 %, almost three times the rate of government capital expenditure as a whole.

Italy’s government finance crisis is clearly visible in public investment data; the 105 % increase in total public capital outlays is entirely accounted for by increases in the most recent years, when EU funds for the Recovery Programme have become available. Investment in health has grown by 33 %, investment in education is unchanged, there is a fall in environmental spending, while arms procurement increases by 132 %.

In terms of total public expenditure, Spain presents a more balanced distribution of government investment across the different areas, with an overall increase of 55 %, equally distributed to education and health; Spain is the only country with some growth in environmental investment (+21 % over the decade). The exceptional growth of 266 % in expenditures for arms and equipment is the result of the rapid increase in arms procurement in 2023.

The widespread problems with public finance and poor economic performances experienced by Europe in the last decade have shaped the context for difficult policy choices regarding public spending priorities. It is remarkable that all NATO EU countries – even those in greater economic difficulties, such as Italy and Spain – have allowed military expenditure and investment in arms procurement to take priority over environmental and social objectives. The increasing military orientation of European economies is likely to be problematic not only in terms of social quality and environmental sustainability but also in terms of its impact on economic and employment performance, that we assess in the next sections.^[6]

4 Results

What are the output and employment effects of €1,000 million expenditure in arms as opposed to non-military alternatives in the economies of Germany, Italy and Spain? Table 2 provides the results for the overall demand multiplier (domestic and foreign), for the share of imports and for the domestic output multiplier effect of each sector – the focus of our investigation.^[11]

The first column of Table 2 shows that arms procurement has a significant multiplier effect on both arms-producing sectors and supplying industries, in the domestic and foreign economies; a €1 increase in final demand generates €1.62 of output in Germany, €1.65 of output in Spain, and €1.82 of output in Italy. However, investment in environmental protection has a higher effect than arms in all three countries, with output multipliers ranging from 1.77 in Germany to 1.91 in Italy. Health has an intermediate effect – with output multipliers ranging from 1.38 in Germany to 1.56 in Italy – while education has the lowest multiplier, as it requires fewer intermediate inputs in goods and services from other industries, and is more labour intensive.

The second column of Table 2 shows the propensity to import in each sector; in all countries, the arms industry exhibits the highest percentages, ranging from 59 % in Italy to 22 % in Spain, while the other three sectors have shares of imports lower than 7 % (with the only exception of the education expenditure in Italy).

The multiplier effects on the domestic economy is reported in column 3 of Table 2, showing the output effect of a €1 increase in expenditure in each of the sectors. In Germany the domestic effect of arms acquisition is 1.23, significantly lower than that of investment in environment and health, and similar to the one in education. In Italy the arms multiplier falls to 0.74, due to the high share of arms imports, the lowest value of all sectors. In Spain arms expenditure has an output effect of 1.28, lower than environment and health, and higher than education.

Table 3 presents for the three countries the detailed effects of a €1,000 million expenditure on domestic output and employment, identifying in each sector the impact on the sector itself and on other industries. For Germany a €1,000 million expenditure in arms procurement sets in motion an increase in domestic output of €1,230 million, for two thirds in arms-producing industries. In Italy the resulting increase is €741 million only, 60 % of which in arms-producing industries. In Spain the impact is 1,284 million, similar to the German results. The high concentration of the effects within the industries that produce arms points out the limited expansionary effects that such expenditure has on the rest of the economy.

In addition, Table 3 shows the employment requirements (in FTE) per million euros of direct output in the industry concerned. The arms industry is highly capital-intensive and employs significantly fewer employees per unit of output than environment, education, and health activities, which are more labour-intensive.

The key result shown in Table 3 is the domestic employment effects of the initial €1,000 million expenditure, broken down in the direct impact within the same industry and the indirect effects in the rest of the economy. A €1,000 million expenditure in arms acquisition is associated to 3,800 additional jobs in Germany, 2,900 in Italy, and 6,000 in Spain, generally half in the arms-producing industry and half in the rest of the economy.

How do these results compare with the three alternative destinations of the original €1,000 million in public expenditure? Environmental protection, education, and health are all characterised by service activities in the domestic economy, with a much lower relevance of imports, less need for intermediate inputs, and a higher labour intensity. The multiplier effect in terms of output and employment for each of the three alternative public expenditures is generally greater than the economic effect of increased arms procurement, except for the output effects of education expenditure in Germany and Spain. In terms of output, the highest results are found for environmental protection, with an increased output of €1,723 million in Germany, €1,866 million in Italy, and €1,717 million in Spain. For education and health, the additional output ranges from €900 million to €1,508 million.

Looking at the impact in terms of additional employment, in Germany the original €1,000 million in public expenditure could lead to the creation of 6,200 new jobs in the environmental sector, 10,700 jobs in education, and 13,000 jobs in health services. The employment impact of education and health is about three times the one resulting from arms spending, and is mainly concentrated within the sectors themselves.

In Italy, the new jobs created by a €1,000 million expenditure would range between 9,000 in environmental services to 10,900 in health services – three to four times higher than the employment impact of increased arms procurement; the share of the new jobs created within the industries ranges from 60 % in environmental protection to 90 % in education and health.

In Spain, the employment effect would range between 10,200 new jobs in the environment to 14,700 in education – two and a half times the number of jobs associated to arms procurement.

These results show that increasing arms expenditure has a significantly lower effect in terms of total additional jobs created in the domestic economy compared to the three alternative destinations of public expenditures – environmental protection, education, and health activities. The results for the three countries are broadly similar, with differences due to the relevance of arms imports and economic structures. In terms of the indirect effects – i.e. the additional jobs created in other industries through the supply chain – in all countries the highest results are found for environmental protection, which has the largest positive employment spillover effects in the rest of the economy.^[12]

4.1 Sensitivity Analysis

How solid are the results we have obtained, and how sensitive are they to the methodology we have adopted? The multiplier effects of arms acquisition we calculated are linear combinations of the multiplier effects of the three main component sectors of arms industry: “Computer, electronic and optical equipment”, “Motor vehicles, trailers and semi-trailers”, and “Other transport equipment”. This implies that the results are sensitive to the weights assigned to each component sector. Additionally, the estimated domestic multiplier effects of investment in arms procurement, environmental protection, education, and health are highly dependent on each country’s and sector’s propensity to import out of final investment demand. Therefore, in order to assess the robustness of our findings, we conducted a sensitivity analysis by generating three scenarios (A, B, C), based on different assumptions, that can be compared with the baseline one (BL) leading to the results shown above.

In Scenario A, we assume that the arms industry is comprised entirely of the most labour-intensive industry (i.e. “Computer, electronic and optical equipment”, which includes military electronics). Thus, the domestic output and employment effects of increasing arms expenditure in this scenario are determined entirely by the output multiplier of the “Computer, electronic and optical equipment” industry and the labour intensity of this sector and its intermediate input suppliers.

In Scenario B, we assume that the arms industry is comprised entirely of the least labour-intensive industry (i.e. “Motor vehicles, trailers and semi-trailers”, which include tanks and armoured vehicles). Therefore, the economic impact of arms acquisition in this scenario is determined entirely by the output multiplier and the labour intensity of this sector and its intermediate input suppliers.

In Scenario C, the propensity to import out of final investment demand was set to 0 for all military and non-military expenditures. This scenario simulates that all final investment goods in the four sectors are produced domestically. Therefore, to calculate the output and employment effects of each type of expenditure, we used the multipliers of final demand provided in Table 2.

Figure 4 shows the employment effects of a €1,000 million expenditure in each sector for the baseline (BL) and the three alternative scenarios (A, B, C).

Figure 4:

Employment effects of 1,000€ million expenditures (in thousands FTE employees).

The sensitivity analysis demonstrates that our findings are robust to different weights for the sectoral structure of the arms industry and to different assumptions about the domestic content of military and non-military sectors. The new jobs created by a €1,000 million expenditure for arms acquisition would range between 3,400 and 5,100 in Germany, between 2,700 and 7,100 in Italy, and between 5,400 and 7,700 in Spain. In all three alternative scenarios, the employment impact of the same amount of increase in public investment in environmental protection, education, and health would be about 1.5–4 times greater than that in arms procurement. In the education and health sectors, the larger employment effects are due to higher labour intensity than in the arms industry. In the environmental protection, the larger employment effects are due to both higher labour intensity and a higher output multiplier than in the arms industry.^[13]

5 Conclusions

In the last decade, Europe has significantly increased its military expenditures, both in absolute terms and as a share of GDP, with a dramatic acceleration since the start of the war in Ukraine in 2022. In all NATO EU countries, this expansion has been driven by higher acquisitions of arms and equipment. At a time of economic stagnation and constraints on public finance, Europe’s increased arms procurement is likely to reduce the space for social and environmental public investment.

In this article, we employed input–output analysis to estimate the economic and employment impacts of increasing expenditure in arms production in Germany, Italy, and Spain, comparing them with the effects of increasing public capital expenditure in environmental protection, education, and health services.

Our findings show that the current drive to increase European military expenditures has problematic economic consequences. Policies that concentrate limited public resources in military activities have negative outcomes in several regards. They result in larger imports of arms and high-tech components, mainly from the US; they lead to lower availability of public resources for environmental and social priorities; and they have a significantly lower effect in terms of domestic growth of output and employment, compared to other potential destinations of public expenditures.

According to our estimates, investment in environmental protection has greater multiplier effects in terms of domestic output than arms acquisition; in terms of domestic employment effects, investing in environmental protection, education, and health activities generate two to four times as many jobs as investing in arms acquisition. These estimates allowed us to compare the number of jobs created by the cumulative increase in arms expenditure in the period 2013–2023 with the potential employment gains if the same amounts of resources had been spent on non-military alternatives. Our calculations show that missed job creation opportunities range between 53,000 and 206,000 jobs/year in Germany, between 100,000 and 131,000 jobs/year in Italy, and between 44,000 and 92,000 jobs/year in Spain.

In quantitative terms, considering the performances of economic and employment growth, our findings show that increased military expenditures – associated with the prospect of a stronger European ‘military-industrial complex’ – may slow down Europe’s development, compared to trajectories based on increased environmental and social expenditures. In terms of the quality of Europe’s development, more expenditures on education, health and the environment bring improvements in wellbeing and sustainability that are even more important than the quantitative gains we may estimate.