For decades, the conventional wisdom held that economic interdependence promotes peace. Countries that trade together have too much to lose from conflict; global supply chains create shared interests; integration binds nations into a community of mutual benefit. This “commercial peace” thesis, with roots stretching back to Montesquieu and Kant, guided the post-Cold War era’s enthusiasm for globalization. When China joined the World Trade Organization in 2001, proponents argued that integration would liberalize not just China’s economy but its politics. Trade flows between the United States and China grew from approximately $5 billion in 1980 to over $750 billion by 2022. European dependence on Russian gas was presented as a feature, not a bug: “Wandel durch Handel” (change through trade) would moderate Moscow’s behavior by creating mutual stakes in cooperation.
The logic seemed irrefutable: why would states risk war when commerce makes them rich?
But interdependence has a shadow side that the optimists underestimated. The same networks that enable trade and communication also create vulnerabilities. States that occupy central positions in these networks—controlling financial chokepoints, dominating technological standards, or hosting critical infrastructure—can weaponize these positions against rivals. The 2022 Russian invasion of Ukraine and the subsequent Western response demonstrated this logic with brutal clarity: the same integration that was supposed to constrain Russian aggression became the vector for the most comprehensive sanctions regime in history, while Europe’s energy dependence gave Moscow coercive leverage of its own.
This is weaponized interdependence: the strategic exploitation of global networks for coercive purposes. The concept, developed by political scientists Henry Farrell and Abraham Newman in their influential 2019 International Security article, fundamentally reframes how we understand economic globalization—not as an inherently pacifying force, but as a new arena of great power competition.
The Structure of Network Power¶
Global systems are not flat; they have architecture. Some nodes matter more than others, and the distribution of centrality determines who can weaponize interdependence and who is vulnerable to such weaponization:
Hub-and-spoke structures concentrate flows through central points in ways that emerged from efficiency considerations but created strategic vulnerabilities. The international payments system routes through a handful of institutions: approximately 11,000 financial institutions in over 200 countries use SWIFT messaging to coordinate transactions, but the correspondent banking system that actually moves money concentrates flows through a much smaller number of major banks, predominantly in New York and London. Internet traffic passes through specific exchange points—the DE-CIX in Frankfurt handles peak traffic exceeding 14 terabits per second; submarine cables concentrate at a few landing points; over 90% of intercontinental data flows through undersea cables vulnerable to surveillance or sabotage. Semiconductor supply chains depend on a few critical manufacturers: TSMC in Taiwan produces approximately 90% of the world’s most advanced chips; ASML in the Netherlands is the sole producer of extreme ultraviolet lithography equipment essential for cutting-edge manufacturing. These hubs enable efficiency—but also create chokepoints that states can exploit.
Asymmetric access means some states and firms occupy more central positions than others, conferring leverage that market share alone does not capture. The United States dominates dollar-based finance: the dollar is used in approximately 88% of all foreign exchange transactions; U.S. banks are essential counterparties for global dollar clearing; the Federal Reserve functions as the world’s de facto lender of last resort (its dollar swap lines reached $449 billion during the 2008 crisis and $444 billion during COVID). American technology platforms intermediate global communications: Google handles over 90% of global search traffic; Amazon Web Services and Microsoft Azure together host approximately 50% of cloud computing; Apple and Google’s mobile operating systems run on 99% of smartphones. This centrality confers power that dispersed participation does not: controlling a network hub means controlling who can use the network and on what terms.
Lock-in effects make switching from dominant networks prohibitively costly, even when alternatives exist. Once institutions, firms, and individuals depend on particular networks, abandoning them requires not just individual decisions but coordinated collective action. SWIFT for international payments has been operating since 1973 and has built decades of trust, standardization, and integration with banking systems worldwide. The GPS constellation provides free, reliable positioning that billions of devices depend upon; Russia’s GLONASS and China’s BeiDou offer alternatives, but achieving comparable adoption requires overcoming vast installed-base advantages. English has become the global language of scientific publication, business, and aviation—not because of inherent superiority, but because network effects make joining the dominant network more valuable than maintaining alternatives. Alternatives exist but adoption requires sustained effort, coordination among many actors, and acceptance of transition costs that individual defectors often refuse to bear.
Mechanisms of Weaponization¶
States exploit network centrality through two primary mechanisms that operate through different logics but often reinforce each other:
Panopticon effects enable surveillance. Central positions provide visibility into flows passing through network hubs, creating intelligence advantages that would be impossible to achieve through traditional espionage. The United States can monitor dollar-denominated transactions through the Treasury Department’s access to SWIFT data (revealed in the 2006 “Terrorist Finance Tracking Program” disclosure) and through the Bank Secrecy Act’s reporting requirements on U.S. financial institutions. The NSA’s PRISM program, revealed by Edward Snowden in 2013, demonstrated how American technology companies’ global market share translated into surveillance capabilities: if data flows through American servers or software, American intelligence agencies can potentially access it. A state controlling internet exchange points can observe unencrypted communications traversing its territory. Platforms collecting data know what users search, buy, and say with a granularity that previous generations of intelligence officials could only dream of—Google processes over 8.5 billion searches daily; Meta’s platforms have approximately 3 billion daily active users. This visibility provides intelligence advantages for counterterrorism, sanctions enforcement, and competitive intelligence, while enabling the targeted action that chokepoint effects make possible.
Chokepoint effects enable coercion by allowing states to block, degrade, or condition access to networks. Exclusion from SWIFT prevents international payments: when Iranian banks lost SWIFT access in 2012, the country could no longer process most international transactions, contributing to a 50% currency collapse and severe economic contraction. Denial of semiconductor manufacturing equipment cuts off chip supplies: American export controls prevent ASML from selling its most advanced lithography equipment to China, potentially setting back Chinese chip manufacturing by years. Removal from app stores limits software distribution: when the United States banned TikTok from government devices and considered a broader ban, it threatened to cut ByteDance off from 150+ million American users. When Russia was partially excluded from SWIFT in 2022 and approximately $300 billion in central bank reserves were frozen, these chokepoint effects imposed immediate, severe economic costs. Threatening or executing such exclusion becomes a tool of statecraft with coercive power that can exceed military force in some contexts.
The scholars Henry Farrell and Abraham Newman, who developed the concept, describe it as “the ability of some states to leverage their privileged positions in networks of interdependence to extract compliance or punish noncompliance from other actors.” This represents a fundamental revision of how we understand economic integration: not as mutual vulnerability that promotes cooperation, but as asymmetric vulnerability that enables coercion by those who control the chokepoints.
Weaponization in Practice¶
Multiple domains demonstrate weaponized interdependence with increasing sophistication and intensity:
Financial networks present the most developed case, with decades of accumulated experience in leveraging American centrality. Dollar dominance means most international trade settles in dollars, flowing through American-regulated correspondent banks that must comply with U.S. law regardless of where they are headquartered. The dollar’s role in approximately 88% of foreign exchange transactions creates a chokepoint: banks that want to clear dollars must maintain correspondent relationships with U.S. institutions, subjecting them to U.S. jurisdiction. The SWIFT messaging system, though headquartered in Belgium and governed by Belgian law, operates under effective American influence—the U.S. Treasury can and has demanded that SWIFT exclude targeted entities under threat of sanctions against SWIFT itself. When the United States imposes sanctions, it can exclude targets from the global financial system with devastating effect because there is no alternative infrastructure of comparable scale and liquidity.
Iran’s experience illustrates the power with stark clarity. Following nuclear-related sanctions, Iranian banks lost SWIFT access in 2012; the country could not receive payments for oil exports or pay for imports through normal channels. Oil exports fell from approximately 2.5 million barrels per day to under 1 million; the rial lost half its value; inflation reached 50%; GDP contracted sharply. The economic damage helped bring Tehran to the negotiating table for the JCPOA nuclear deal in 2015. When the Trump administration reimposed sanctions in 2018, the pattern repeated despite European efforts to maintain trade through the INSTEX mechanism—which processed negligible transactions because European companies feared American secondary sanctions more than they valued Iranian business.
Technology platforms intermediate information flows globally, creating vulnerabilities that extend from commerce to surveillance to social cohesion. American firms—Google, Apple, Meta, Amazon, Microsoft—dominate cloud computing (AWS and Azure together hold approximately 50% global market share), mobile operating systems (Android and iOS run on 99% of smartphones), search (Google handles 90%+ globally), and social media (Meta’s platforms reach 3+ billion users). European dependence on these platforms prompted “digital sovereignty” initiatives: the General Data Protection Regulation (GDPR), proposed Digital Markets Act, and discussions of European cloud infrastructure (Gaia-X) all reflect concern about American platform power. Chinese exclusion (both self-imposed through the Great Firewall and externally applied through U.S. restrictions on Huawei and other firms) created a separate digital ecosystem with domestic alternatives (Baidu for search, WeChat for social media, Alibaba Cloud for cloud computing) that insulate China from American platform leverage while creating Chinese chokepoints over the domestic market.
Semiconductor supply chains concentrate critical production in a few locations with a degree of geographic concentration that astonishes policymakers who examine it closely. Taiwan Semiconductor Manufacturing Company (TSMC) fabricates approximately 90% of the world’s most advanced chips (those at 7nm and below); a single company on an island 100 miles from mainland China controls technology essential for smartphones, data centers, AI development, and advanced weapons systems. ASML in the Netherlands is the sole producer of extreme ultraviolet (EUV) lithography equipment essential for cutting-edge manufacturing—each machine costs approximately $150 million, takes 40 freight containers to ship, and requires years to install and calibrate. American firms design key chips (Nvidia, AMD, Qualcomm) and develop crucial software tools (Cadence, Synopsys for chip design; Applied Materials for manufacturing equipment). Export controls can deny adversaries access to essential technology—as American restrictions on China demonstrate, cutting Huawei off from advanced chips and blocking ASML sales of EUV equipment to Chinese fabs.
Cloud infrastructure hosts critical data and computing services with implications for sovereignty, surveillance, and economic competitiveness. When data resides on foreign servers subject to foreign law, fundamental questions arise about who can access it. The American CLOUD Act (2018) enables U.S. authorities to compel data disclosure from American firms regardless of where data is stored physically—if Microsoft stores European customer data in an Irish data center, American warrants can still demand access. This creates conflicts with European privacy law (particularly GDPR) and has prompted data localization debates: should sensitive data be required to stay within national borders? Russia, China, India, and others have enacted or proposed data localization requirements. The result may be fragmentation of the global cloud into national or regional silos—reducing efficiency while enhancing sovereignty.
Strategic Responses¶
States facing weaponized interdependence pursue several strategies, often in combination, to reduce vulnerability and restore freedom of action:
Diversification reduces dependence on any single network or provider by cultivating multiple relationships. Europe’s efforts to develop alternative financial messaging for Iran trade (INSTEX, established 2019, though it processed negligible transactions) demonstrated the difficulty but also the conceptual commitment. China’s creation of CIPS for yuan-denominated payments—processing approximately $14 trillion in 2023 with connections to over 100 countries—provides a backup to SWIFT-based transactions. Various states pursue domestic cloud infrastructure: the EU’s Gaia-X project, India’s data localization requirements, and China’s domestic cloud ecosystem all reduce dependence on American hyperscalers. The European Union’s effort to source critical materials from multiple suppliers (the Critical Raw Materials Act targets reducing any single-country dependence to below 65%) reflects diversification logic applied to supply chains.
Duplication builds parallel systems that can substitute for networks subject to weaponization, accepting some efficiency loss in exchange for strategic autonomy. Russia’s SPFS payment system connected over 500 institutions and processed increasing volumes even before 2022 sanctions made it essential—though its effectiveness proved limited when the largest Russian banks were cut off from international correspondents. China’s BeiDou navigation constellation (completed 2020) provides an alternative to GPS with 35 satellites offering global coverage and sub-meter accuracy for military users; it reduces dependence on American-controlled positioning while giving China its own dual-use capability. Huawei’s HarmonyOS, developed after American sanctions cut access to Google’s Android services, attempts to create options outside the Google-dominated mobile ecosystem—though adoption outside China remains limited.
Obstruction resists compliance with weaponization demands through legal and regulatory mechanisms. The EU’s blocking statute (originally 1996, updated 2018) forbids European firms from complying with certain extraterritorial U.S. sanctions—though in practice, most firms choose American market access over protected trade with sanctioned countries. Data localization laws in Russia, China, India, and elsewhere prevent information from reaching foreign jurisdictions where it might be subject to surveillance or compulsion. Encryption and anonymization technologies reduce surveillance vulnerability at the technical level: end-to-end encrypted messaging, VPNs, and Tor make panopticon effects harder to achieve. The legal battles over European-American data transfers (Safe Harbor struck down 2015; Privacy Shield struck down 2020; the Trans-Atlantic Data Privacy Framework remains contested) reflect ongoing obstruction of American data access.
Resilience building accepts continued network participation while reducing vulnerability to disruption through preparations that would kick in if access were denied. Stockpiling critical components (Taiwan maintains several months of semiconductor inventory; the U.S. Strategic Petroleum Reserve holds approximately 350 million barrels), developing domestic production capacity for essential goods (the U.S. CHIPS Act provides $52 billion for domestic semiconductor manufacturing), and maintaining alternative communication channels all enhance resilience. The goal is not to exit networks but to reduce the costs of potential exclusion, thereby reducing the coercive leverage that network centrality provides.
Retaliation threats may deter weaponization by promising reciprocal costs that the weaponizer would prefer to avoid. A state with its own network centrality can threaten counter-exploitation: China’s control over rare earth processing (approximately 60% of global production, 90% of processing) provides leverage against American technology firms. Even states without such centrality may possess other leverage that deters action: Russia’s nuclear arsenal, China’s holdings of U.S. Treasury securities (approximately $800 billion), or simply the threat of market access denial can impose costs that make weaponization less attractive. Deterrence works only if threats are credible; states must demonstrate willingness to bear the costs of retaliation.
Consequences and Risks¶
Weaponized interdependence carries significant implications:
Network fragmentation may result as states seek to reduce vulnerability. The multipolar digital and financial order that emerges could be less efficient than integrated systems but may be more resilient to coercion—and may better reflect diverse values and interests.
Erosion of trust undermines the benefits interdependence provides. If economic relationships are potential weapons, states will invest less in them. The efficiency gains from globalization depend on confidence that connections will not become constraints.
Escalation dynamics may emerge when states respond to weaponization with counter-measures. Sanctions invite retaliation; technology denial prompts indigenous development; financial exclusion accelerates de-dollarization. Spirals of mutual exclusion could fragment the global economy.
Normative ambiguity surrounds the legitimacy of weaponized interdependence. When the United States uses financial power to punish nuclear proliferation, many see legitimate enforcement of international norms. When China uses economic pressure against Australia for demanding a COVID inquiry, many see illegitimate bullying. The line between justified use of leverage and improper coercion remains contested.
Institutional strain affects the organizations managing global networks. SWIFT, once a purely technical utility, has become a geopolitical instrument. Internet governance bodies face pressure from competing sovereignties. The WTO struggles to address “national security” exemptions that justify weaponized measures.
The Future of Interdependence¶
Several trends will shape how weaponized interdependence evolves in the coming decades:
Great power competition between the United States and China ensures continued attempts to leverage network positions and continued efforts to reduce vulnerability. Geoeconomic competition through networks has become a permanent feature of their rivalry, with annual escalations: American semiconductor export controls (October 2022, expanded October 2023), Chinese critical mineral export restrictions (gallium and germanium, July 2023; graphite, October 2023), competing cloud infrastructure investments, and battles over telecommunications standards (5G, 6G). Neither side can achieve complete autonomy—the global economy is too integrated—but both sides are investing billions to reduce their most critical dependencies.
Middle power maneuvering will intensify as states seek to avoid becoming collateral damage in great power contests. India maintains defense relationships with both the United States and Russia while pursuing “Make in India” manufacturing; Indonesia balances Chinese investment against security ties with Australia and the U.S.; Brazil, Turkey, and Saudi Arabia all pursue multi-alignment strategies that preserve access to all major networks while avoiding the dependencies that enable weaponization. The expanded BRICS grouping (adding six new members in 2024) partly reflects this logic: a forum for coordinating responses to Western network power without fully aligning with China.
Technology evolution may create new chokepoints or reduce existing ones—the structure of future networks remains contested terrain. Quantum computing, if achieved at scale, could break current encryption standards, creating new vulnerabilities while potentially enabling unhackable communications. Advanced AI concentrates in a few firms (OpenAI, Google DeepMind, Anthropic, a handful of Chinese companies) with implications for economic competitiveness and military capability. But technology can also reduce centralization: decentralized finance (DeFi) and cryptocurrency offer alternatives to traditional payment networks, though they face adoption barriers and regulatory uncertainty; mesh networks and satellite internet (Starlink has over 2 million subscribers) reduce dependence on centralized infrastructure; open-source alternatives to proprietary platforms exist even if they lack the scale of dominant providers.
Normative development could theoretically establish limits on weaponization, distinguishing acceptable uses (sanctions against human rights violators or weapons proliferators) from illegitimate coercion (weaponizing trade relationships for political disputes unrelated to international peace and security). International law addresses some aspects of economic coercion, and the WTO provides dispute resolution for trade measures. But agreement on clear norms requires consensus that great powers currently lack: what the United States calls sanctions enforcement, others call economic warfare; what some see as legitimate national security measures, others see as protectionism by another name. The normative vacuum means weaponization continues largely unconstrained by agreed rules.
The insight of weaponized interdependence is that connection creates not only opportunity but also vulnerability. The integrated global economy that produced prosperity—global GDP grew from approximately $30 trillion in 1990 to over $100 trillion by 2023, with trade and investment flows binding nations together as never before—also produced dependencies that can be exploited by states occupying central network positions. Managing this reality—maintaining the benefits of interdependence while limiting its weaponization—will be a central challenge of twenty-first-century statecraft. The alternative, fragmentation into separate economic and technological blocs, would sacrifice the efficiency gains of integration while not necessarily eliminating the vulnerabilities of more limited interdependence within each bloc.
Sources & Further Reading¶
-
Weaponized Interdependence: How Global Economic Networks Shape State Coercion by Henry Farrell and Abraham Newman — The foundational 2019 International Security article that introduced the concept and its theoretical framework.
-
Underground Empire: How America Weaponized the World Economy by Henry Farrell and Abraham Newman — The authors expand their argument into a book-length treatment of how network power operates and its implications for global order.
-
The Uses and Abuses of Weaponized Interdependence edited by Daniel W. Drezner, Henry Farrell, and Abraham Newman — A collection examining how weaponized interdependence operates across different domains and the strategic responses it generates.
-
Commerce and Coalitions by Ronald Rogowski — Classic political economy work that provides theoretical background on how trade and economic integration reshape domestic politics and international relations.