Industrial Policy, Defence Spending and Financialization: State Capitalism Reinterpreted

After four decades in which supply-chain globalisation, deregulation and private capital allocation defined the dominant model of economic governance, major economies have reintroduced explicit industrial strategy at scale. The instruments of this return are varied — semiconductor manufacturing grants, green energy investment packages, reshoring incentive frameworks, dual-use technology programmes — but the underlying logic is consistent: states are reasserting their role as strategic coordinators of productive investment in sectors deemed essential to national security, technological competitiveness and economic sovereignty. This is not a marginal adjustment to the liberal market model. It represents a structural shift in the relationship between state authority and market mechanism.

The revival coincides with the most sustained increase in global military expenditure since the Cold War. SIPRI’s Military Expenditure Database records total world defence spending of $2,443 billion in 2023, representing an increase of 6.8 per cent in real terms over 2022 and the ninth consecutive annual increase globally. The United States alone spent $916 billion on defence in 2023, more than the next ten largest defence budgets combined. China allocated an estimated $296.4 billion; Russia $109.5 billion in the context of its ongoing military campaign in Ukraine. Twenty-two of thirty-two NATO members met the 2 per cent of GDP spending target in 2024, with European NATO members increasing defence expenditure by 19.4 per cent in real terms in 2024 alone.

This simultaneous intensification of industrial ambition and defence expenditure unfolds within a financial environment that previous iterations of state capitalism did not face. Global public debt reached 93 per cent of world GDP in 2023 (IMF Fiscal Monitor, October 2023). Advanced economies carry an average sovereign debt-to-GDP ratio of approximately 112 per cent. The G7 average exceeds 128 per cent of GDP, elevated substantially by Japan’s extreme debt position. Institutional investors managing approximately $112 trillion in assets, operating through bond markets with approximately $133 trillion in outstanding securities, exercise continuous and immediate discipline over sovereign fiscal decisions. Industrial policy no longer unfolds within insulated national financial systems. It unfolds under the continuous surveillance of globalised capital markets.

This paper develops three analytical arguments. First, that the defence-industrial nexus has evolved beyond its Cold War configuration into a structure in which financial market beneficiaries — institutional investors, publicly listed defence contractors, asset managers with equity stakes in strategic sector firms — are embedded within the political economy of strategic competition in ways that shape both the scale and direction of state intervention. Second, that the financialization of sovereign balance sheets creates structural asymmetries between reserve-currency-issuing states, which can finance industrial ambition at relatively contained borrowing costs, and emerging economies, which cannot — widening rather than narrowing the gap between monetary core and periphery. Third, that the distributional consequences of current industrial programmes, if uncorrected by design, risk channelling the fiscal benefits of state intervention toward equity holders and high-skilled workers rather than the broad labour force in whose name industrial policy is typically justified.

The contemporary wave of industrial strategy is distinguished from its mid-20th-century antecedents by the institutional channels through which states intervene. Ownership nationalisation, centralised planning and price controls characterised earlier forms of state capitalism. The contemporary model operates primarily through subsidies, tax incentives, procurement preferences, public-private partnerships and credit guarantees. Ownership remains formally private; strategic direction is publicly provided through fiscal instruments and regulatory architecture. This distinction matters both analytically and politically: it generates different distributional consequences, different interactions with financial markets, and different accountability structures.

The quantitative scale of current commitments is without recent precedent outside wartime. The United States passed the CHIPS and Science Act and the Inflation Reduction Act in 2022, with the IRA’s climate and energy provisions scored at $369 billion over the FY2022–2031 window by the Penn Wharton Budget Model using Congressional Budget Office scoring conventions — a figure that independent analyses suggest substantially understates actual credit uptake under open-ended tax credit structures. The European Commission mobilised a €43 billion public and private investment target through the EU Chips Act framework. Japan committed subsidies of up to JPY 476 billion, approximately $3.5 billion, for the first TSMC Kumamoto fabrication facility alone. South Korea announced a 26 trillion won ($19 billion) semiconductor support package. India approved Rs 76,000 crore, approximately $9.95 billion, in semiconductor Production Linked Incentive scheme commitments. China’s National Integrated Circuit Industry Investment Fund — the “Big Fund” — totalled RMB 138.7 billion under its first phase and RMB 204 billion under its second, with a third fund reportedly under preparation.

The semiconductor sector illustrates the convergence of industrial and security logics with particular clarity. Global semiconductor sales reached $526.8 billion in 2023 (SIA/WSTS, February 2025). At least twenty major economies announced semiconductor industrial strategies between 2020 and 2023. The sector simultaneously serves as the material infrastructure of civilian digital economies and the enabling technology of modern military systems, from precision-guided munitions to electronic warfare platforms to autonomous systems. The industrial policy competition over semiconductor manufacturing capacity is therefore simultaneously a competition over civilian technological leadership and military-technological primacy — a distinction that policymakers in Washington, Brussels, Beijing and Tokyo have formally dissolved.

Military expenditure has historically served as the primary mechanism through which states directed long-horizon, high-risk technological investment that private capital markets would not fund at equivalent scale or on equivalent timelines. The canonical examples are well established: aerospace engineering, semiconductor research, the internet, GPS navigation, and nuclear energy each originated in defence-funded research programmes that subsequently generated civilian technology platforms of enormous economic value. The developmental state and the fiscal-military state are not analytically distinct institutions. They are the same institution observed in different sectors.

The contemporary defence-industrial relationship has, however, evolved in ways that alter its political economy substantially. The largest defence contractors are publicly listed corporations subject to shareholder primacy, capital markets discipline and quarterly earnings expectations. Lockheed Martin reported total revenues of $67.6 billion in FY2023; RTX Corporation $68.9 billion; Northrop Grumman $39.3 billion. The combined revenues of these three firms alone exceed $175 billion annually — comparable in scale to the entire defence budgets of France, Germany or the United Kingdom. Large asset managers and pension funds hold substantial equity stakes in all three. Defence sector equities demonstrated precisely the financial market logic of their current structure during 2022: MSCI World Aerospace and Defence index delivered an approximate total return of negative 3 per cent in 2022, against the MSCI World’s approximate negative 18 per cent return — a differential of approximately 15 percentage points at the point of maximum geopolitical volatility. Defence sector equity therefore functioned as a portfolio hedge against the very geopolitical instability that defence spending is intended to resolve. The investors who benefit from geopolitical tension through portfolio gains and the states that seek to manage that tension through defence expenditure are not in a principal-agent relationship. They are aligned stakeholders.

US Department of Defense procurement appropriations for FY2023 totalled $170.9 billion, excluding research, development, testing and evaluation funding. This represents a sustained and contractually committed demand base for the defence industrial ecosystem — a form of countercyclical industrial stimulus that economists rarely incorporate into their assessments of US manufacturing capacity, but which in practice sustains domestic fabrication capacity, technical workforce depth, and supply chain resilience in ways that no equivalent private sector demand programme achieves. The defence procurement budget, in this light, functions as the most durable industrial policy instrument in the US economy, predating and substantially exceeding any of the recent civilian industrial initiatives.

Financialization describes the growing influence of financial markets, financial institutions and financial logics over economic activity, policy formation and corporate behaviour. It is not a recent phenomenon but a structural transformation of the global economy that accelerated from the 1980s through deregulation, the removal of capital controls and the growth of institutional investment intermediation. Its quantitative dimensions are now sufficiently large to constitute a structural environment that constrains all other policy domains. Global financial assets — equities, debt securities and loans combined — represent approximately 3.5 times world GDP, according to McKinsey Global Institute analysis (2022). Global listed equity market capitalisation reached approximately $111 trillion at end-2023 (World Federation of Exchanges). The global bond market outstanding stood at approximately $133 trillion in 2023 (SIFMA Capital Markets Fact Book, 2023), of which government bonds alone account for approximately $88 trillion (BIS Debt Securities Statistics, 2023). Industrial policy that requires sovereign borrowing to finance it therefore immediately enters a market in which its marginal fiscal impact is continuously priced.

The institutional structure of this financial environment concentrates an extraordinary degree of influence in a small number of organisations. BlackRock, Vanguard and State Street collectively manage approximately $23 to $25 trillion in assets — a combined portfolio that, at year-end 2023, approaches the entire US GDP. Global institutional investors across pension funds, insurance companies and investment funds manage approximately $112 trillion in assets (OECD Institutional Investors Statistics, 2023). Passive index funds hold an estimated 50 per cent of S&P 500 equity (Morningstar research, 2023), meaning that a large proportion of the largest publicly listed corporations in the world’s largest economy are owned by funds whose investment decisions are determined by index composition rather than active assessment of individual corporate performance. When states subsidise specific sectors and firms, they are therefore subsidising firms whose ownership structure is dominated by passive institutional vehicles whose returns are broadly distributed across pension and insurance fund beneficiaries — but whose governance influence over the subsidised firms is structurally limited.

The concept of the military-industrial complex, popularised by President Eisenhower’s 1961 farewell address, described a structural relationship between defence contractors and procurement bureaucracies that risked distorting public resource allocation in the direction of corporate and institutional preferences rather than genuine security needs. The analytical framework remains valid, but the nexus it describes has expanded substantially in both scope and complexity since the Cold War period Eisenhower was characterising. The transformation has three dimensions.

First, the principal defence contractors are now publicly listed corporations integrated into the financial market ecosystem in ways that were not characteristic of the Cold War period. The equity of Lockheed Martin, RTX and Northrop Grumman is held substantially by institutional investors — the same asset managers and pension funds that hold the equities of technology firms, energy companies, and financial institutions across the broader market. When geopolitical tension increases defence sector valuations, the portfolio gains accrue not to a narrow constituency of military industrialists but to pension beneficiaries, index fund holders and insurance policyholders who may be largely unaware of the defence sector’s weight within their portfolios. The political economy of this diffusion is complex: it both broadens the social base of defence sector financial interest and depoliticises the financial relationship between geopolitical competition and asset returns.

Second, the boundary between defence and civilian production has become progressively less distinct. Dual-use technology — in artificial intelligence, autonomous systems, satellite communications, advanced materials and quantum computing — means that industrial subsidies directed ostensibly at civilian technological leadership simultaneously expand military-relevant capabilities. The CHIPS Act subsidises semiconductor manufacturing capacity that serves both the civilian digital economy and the precision weapons systems that depend on advanced integrated circuits. Green energy infrastructure that employs battery storage reduces dependence on adversary supply chains for critical minerals and enhances grid resilience against both natural disasters and deliberate attacks. Industrial policy, defence policy, and energy security policy have converged into a unified strategic logic that, while coherent from a national security perspective, creates significant analytical and governance challenges.

Third, sovereign wealth funds have increasingly been deployed as instruments of strategic industrial policy alongside traditional state procurement and subsidy mechanisms. Global sovereign wealth funds manage approximately $10 to $11 trillion in assets (SWF Institute, 2024). Gulf state funds, Norwegian pension capital and Asian government investment vehicles have taken strategic equity positions in technology, energy and infrastructure assets. This instrument, deployed by states with the financial capacity to use it, constitutes a form of market-aligned industrial policy that bypasses the legislative and bond-market scrutiny that formal subsidy programmes attract — with corresponding reductions in transparency and democratic accountability.

The energy transition has become the primary site of green industrial policy competition among major economies, and the sector that most clearly illustrates both the ambition and the structural constraints of contemporary state capitalism. Global clean energy investment exceeded $1.7 trillion in 2023, according to IEA World Energy Investment 2023 analysis — exceeding fossil fuel investment for the first time in history and reflecting a structural shift in the direction of capital formation in the energy sector. Electric vehicles represented 18 per cent of all new car sales globally in 2023 (IEA Global EV Outlook, 2024), with China’s domestic EV market accounting for the majority of this share. The European Commission’s Green Deal mobilisation target, at a minimum of €1 trillion over the 2021–2030 decade, represents the largest programmatic fiscal commitment in the bloc’s history.

These investment flows operate within an industrial geography that has not yet been meaningfully reshaped by subsidy programmes. As the Meridian Research Supplement R1 documented, approximately 70 to 90 per cent of renewable energy equipment installed in the Global South is imported rather than locally manufactured — a figure that reflects the concentration of solar photovoltaic, wind turbine and battery manufacturing capacity in China and a small number of other advanced producers. The green industrial competition between the United States and China is not merely a competition over market share in clean energy products. It is a competition over whether the industrial geography of the energy transition will be determined by the existing technology and manufacturing advantages of early movers, or whether new entrants can establish competitive positions through subsidy and policy support. The IRA’s domestic content requirements, the EU’s Net-Zero Industry Act, and national battery and EV manufacturing programmes are all attempts to alter the industrial geography of the transition in favour of domestic producers — with the distributional implication that the global cost of the transition may be higher than under open trade, as less efficient domestic production substitutes for lower-cost imports.

The capacity of states to deploy industrial strategy at scale is not uniformly distributed. It is structured by the same monetary hierarchy that conditions fiscal multipliers and balance-of-payments constraints in the Keynesian framework examined in Research Supplement R1. Reserve-currency-issuing states — most importantly the United States, but also the European Central Bank zone, Japan and the United Kingdom — borrow in their own currencies, face contained sovereign spreads, and have access to central bank liquidity backstops that provide, in practice, unconstrained short-run financing capacity for fiscal commitments. Emerging economies face higher borrowing costs, currency risk on external debt, and the continuous threat of sudden stop dynamics when capital flow reversals compress fiscal space simultaneously with the onset of the economic conditions that motivate industrial intervention.

The US federal debt held by the public reached $26.2 trillion in FY2024 (US Treasury Bureau of Fiscal Service). The G7 average sovereign debt-to-GDP ratio stands at approximately 128 per cent. Despite these fiscal positions, advanced economy governments retain the capacity to launch industrial programmes at the scale described in the previous section because their borrowing costs, while elevated relative to pre-pandemic levels, remain contained by market confidence in their monetary sovereignty and by the institutional architecture of reserve currency management. The asymmetry was starkest during the COVID-19 fiscal response: advanced economies accounted for approximately 85 to 90 per cent of total announced global fiscal support in 2020–2021 despite representing a substantially smaller share of global population — a disproportion that reflected not merely wealth differentials but the structural advantages of monetary hierarchy in determining which states could borrow without triggering market discipline.

For emerging economies, the implication is clear and documented in the Meridian’s prior analysis: the revival of industrial policy in major advanced economies does not create equivalent space for industrial policy in the Global South. It may narrow it, through the reallocation of global investment capital toward subsidised advanced economy industrial programmes, the reshoring of manufacturing that previously generated export earnings for emerging producers, and the competitive pressure on emerging economy export sectors from the subsidised firms that industrial programmes support. The green transition illustrates this asymmetry with particular force: the renewable energy and electric vehicle supply chains that advanced economies are attempting to build domestically through subsidy were, under previous trade arrangements, significant sources of manufacturing employment and export earnings for emerging producers. Industrial policy in the monetary core is not neutral with respect to industrial development in the periphery.

Industrial policy rhetoric consistently emphasises employment creation, wage growth and the revival of middle-class manufacturing livelihoods. The empirical record of distributional outcomes from financialized industrial programmes is more ambiguous. The structural shift in the US labour market over the past five decades is illustrative: manufacturing employment peaked at approximately 22 per cent of total US employment in 1979 and had declined to approximately 8.2 per cent of total employment by 2023, according to Bureau of Labor Statistics Current Employment Statistics data. The absolute peak in manufacturing jobs was 19.4 million in 1979; by 2023, the figure was approximately 13 million in an economy of 160 million employed. Industrial programmes do not reverse this structural trend; they slow specific components of it in targeted sectors.

The labour share of GDP in advanced economies has declined by approximately 3 to 5 percentage points since the early 1980s, according to ILO World Employment and Social Outlook 2023 analysis, with partial stabilisation observed post-2015 but no structural reversal. Over the same period, corporate share buybacks in S&P 500 firms have become the primary mechanism for distributing corporate earnings to shareholders. S&P 500 buybacks reached $922 billion in 2022 and approximately $795 billion in 2023 (S&P Dow Jones Indices Buybacks Report, 2023). Cumulative buybacks over the 2018–2022 period totalled approximately $4.8 trillion, compared with capital expenditure by the same firms of approximately $3.9 to $4.0 trillion over the same period — a ratio of approximately 1.2 times buybacks to capital investment. Industrial policy that increases corporate profitability in subsidised sectors without distributional conditions on capital allocation may, in the absence of explicit conditionality, accelerate buyback programmes rather than productive investment.

The employment projections attached to major industrial programmes should be read against this backdrop. The CHIPS Act is projected to generate approximately 40,000 direct semiconductor manufacturing jobs (US Department of Commerce CHIPS Program projections, 2024). The IRA’s clean energy provisions are projected to support approximately 1.5 million additional clean energy jobs by 2030 under US Department of Energy and Princeton REPEAT modelling (2023). Both are modelled projections, not achieved outcomes; both face considerable execution risk from supply chain constraints, skilled labour availability and technology adoption timelines. The broader point is structural: even if both programmes fully achieve their employment projections, the combined figure represents approximately 1 per cent of the US labour force — meaningful but not transformative without complementary policies on wages, training, union density and regional development. The top 1 per cent of wealth holders accounted for approximately 46 per cent of global household wealth in 2023 (UBS Global Wealth Report, 2023). Industrial programmes that increase equity valuations before generating broad employment will, in the absence of redistributive mechanisms, compound this concentration.

The concept of state capitalism carries significant analytical baggage that obscures more than it illuminates when applied to contemporary industrial policy. The China model — in which state-owned enterprises account for approximately 40 per cent of total industrial assets (OECD Business and Finance Outlook, 2023) and strategic sectors are directed through a combination of Party authority, SOE management and state development bank financing — is structurally distinct from the subsidy-based, procurement-driven model that the United States, European Union and Japan deploy. Both involve state actors directing investment toward strategic ends; neither involves the market-replacing central planning of the mid-20th-century socialist economies. But the accountability mechanisms, the distributional structures, the role of private capital and the relationship between state direction and market pricing differ substantially.

The proliferation of formal industrial policy frameworks is nonetheless striking. The OECD’s Going for Growth 2023 and associated industrial policy analysis indicate that more than 75 per cent of OECD member states have active industrial or strategic sector policy frameworks, a proportion that represents a structural departure from the liberal consensus of the 1990s and early 2000s. The legitimising logic of this departure has shifted from the supply-side arguments of the Washington Consensus to a security-inflected framework in which industrial strategy is justified as a national security necessity rather than as market failure correction. This rhetorical shift matters: it reduces the analytical scrutiny applied to industrial programme design, weakens the case for performance conditionality and sunset provisions, and makes it politically easier to sustain programmes that underperform on productivity criteria as long as security rationales remain salient.

The accountability deficit in contemporary industrial strategy is compounded by the financial structure through which intervention occurs. Subsidies delivered through tax credit mechanisms — as with much of the IRA — are claimed against private investment decisions that the state does not directly observe until they materialise in tax returns. The strategic direction is public but the execution is private and the fiscal cost is uncertain in magnitude and timing. This structure provides flexibility and reduces execution risk for the state, but it also reduces the leverage that states have over the distributional outcomes of the programmes they are financing. A manufacturing subsidy conditional on wage floors, domestic content requirements and union recognition generates different outcomes than a tax credit available to any eligible investor regardless of labour practices. The difference between these two designs is the difference between industrial policy as distributional instrument and industrial policy as asset-support mechanism.

The analytical framework of this paper would be incomplete without explicit attention to the asymmetric structure of state capitalism across the international monetary hierarchy. The discussion of financialization in Section IV and the fiscal constraint analysis in Section VII establish the relevant parameters: reserve-currency states can finance industrial ambition at contained costs; emerging economies cannot. But the asymmetry extends beyond financing costs to the capacity to sustain industrial programmes through the inevitable cycle of political challenge, implementation difficulty and near-term fiscal cost that precedes long-run industrial development outcomes.

The EMBI Global sovereign spread of approximately 380 basis points in 2024, against a pre-pandemic baseline of approximately 290 basis points (JPMorgan, 2024), represents the structural premium that emerging economy sovereigns pay for financing conditions that remain systematically less favourable than those available to reserve-currency issuers. Sub-Saharan African economies, with an average fiscal deficit of 5.3 per cent of GDP and interest payments consuming 12 per cent of government revenue (IMF Regional Economic Outlook SSA, 2024), have neither the fiscal space to deploy industrial subsidies at meaningful scale nor the financial system depth to absorb sovereign borrowing for that purpose without crowding out private credit. Only two of fifty-four Sub-Saharan African countries hold investment-grade sovereign ratings.

The consequence is that the current wave of industrial policy competition primarily intensifies the gap between the monetary core and periphery. Advanced economies subsidise reshoring of manufacturing activities that previously generated employment and export earnings in emerging markets. They compete for critical mineral supply chains that emerging producers supply but do not process. They attract skilled workers trained in emerging economy universities into high-skill programmes supported by advanced economy industrial subsidies. And they do this while the international financial architecture — the currency hierarchy, the distribution of IMF voting shares, the allocation of permanent Federal Reserve swap lines to five advanced economy central banks with no emerging economy counterpart — remains structurally unchanged. Industrial policy is not self-evidently a Global South policy instrument under current conditions. It requires the conditions that industrial policy is intended to create before it can be effectively deployed.