The Price of Abandonment: A Scenario Analysis of U.S. Withdrawal from NATO

A United States departure from NATO would not end American exposure to European insecurity. It would dissolve the institutional machinery that has contained that insecurity, weaken the credibility of American commitments elsewhere, impose substantial economic costs, and increase the probability that a future European crisis would cross the nuclear threshold.

On a January morning, an American president transmits notice that the United States intends to leave the North Atlantic Treaty. Article 13 permits withdrawal one year after notice. Congress immediately invokes 22 U.S.C. § 1928f, which purports to require either an act of Congress or the consent of two-thirds of the Senate. Litigation begins. The administration does not wait for its resolution. It orders the removal of American forces and pre-positioned stocks, ends participation in NATO defense planning, recalls officers from the integrated command structure, suspends nuclear-sharing arrangements, and declares that the United States will no longer treat Article 5 as a binding political commitment.

The legal question would remain unsettled; the strategic question would not. Expectation is the linchpin of deterrence. Moscow, Beijing, European capitals, financial markets, and allied defense ministries would update their calculations as soon as Washington demonstrated that it no longer intended to perform the treaty. The formal one-year clock would therefore obscure the immediate rupture.

This scenario examines the consequences of a rapid and politically hostile withdrawal in which the United States offers no equivalent bilateral guarantees and Europe has not yet constructed an autonomous command, intelligence, logistics, and nuclear-deterrence system.

NATO is often described in American politics as a protection service supplied to insufficiently grateful Europeans. That interpretation omits the bargain’s reciprocal character. The alliance deters territorial revision in one of the world’s other principal industrial centers; suppresses incentives for allied nuclear proliferation; provides the United States with bases, overflight rights, intelligence, interoperable forces, and political support; and gives Washington substantial influence over European defense policy. Leaving NATO would exchange those benefits for an uncertain reduction in American costs and obligation. The United States would emerge less constrained in the narrow legal sense, but less capable of shaping the environment in which it must operate.

The First Shock: American Commitments Become Contingent

The first effect would be a credibility shock, but it is not likely to be a cascading one unless further actions precipitate those concerns. Daryl Press’s critique of reputation-based deterrence remains important: governments generally assess whether another state will fight by examining its present interests and capabilities, not merely its conduct in an unrelated past crisis (Press 2005). Japan would not conclude mechanically that the United States would abandon it because Washington left NATO. The strategic geography, force posture, domestic constituencies, and balance of interests in East Asia are different.

Withdrawal would nevertheless create a genuine commitment and confidence problem. NATO is the oldest, most institutionalized, and most operationally integrated American alliance. If Washington repudiated it despite congressional opposition, extensive sunk costs, and the absence of an allied attack on American interests, other governments would learn that an American security guarantee can be reversed by a single election before the beneficiary has time to replace it.

Future agreements would therefore require a hedge. Allies would demand more visible and less reversible demonstrations of commitment: statutory guarantees, forward-deployed forces, pre-positioned weapons, shared command arrangements, automatic consultation mechanisms, or access to sensitive technologies. Where those signals were unavailable, states would diversify. Some would accommodate nearby powers; others would create regional coalitions; still others would develop latent military capabilities that reduced dependence on Washington. Recent research on retrenchment finds that allies commonly respond to fears of abandonment by attempting to bind the United States, increasing self-help, or seeking alternative alignments (Priebe et al. 2025a). The loss is not simply goodwill. It is the higher price the United States would have to pay to make every subsequent promise believable.

Thomas Schelling observed that “the power to hurt is bargaining power” (Schelling 1966, 2). Alliance commitments convert that latent power into a predictable deterrent. Their credibility rests partly on forces and plans, but also on the institutional routines that make intervention less discretionary at the moment of crisis. NATO’s integrated command, common standards, exercises, and pre-delegated planning reduce the number of political and military decisions required after an attack. Withdrawal would restore discretion by dismantling those routines. That would make American nonintervention easier, but it would also make threats of intervention less credible.

The effect would travel beyond Europe through allied expectations rather than through a metaphysical national reputation. Japan, South Korea, Australia, and the Philippines would reassess the durability of agreements exposed to American electoral volatility. Beijing could rationally conclude that Washington’s alliance policy had become less predictable and that allied access, coalition formation, and sanctions coordination might be less reliable. In a crisis, small revisions to those estimates could be consequential.

South Korea presents the clearest cross-theater test. Its 1953 Mutual Defense Treaty with the United States is legally and operationally distinct from NATO, and Washington could retain its forces and nuclear guarantee on the peninsula. Yet a deliberate repudiation of the United States’ oldest and most integrated multilateral alliance would supply direct evidence that even deeply institutionalized commitments can be easily reversed. The historical precedent is uncomfortable: reductions in American forces and doubts about extended deterrence helped prompt Park Chung Hee’s covert nuclear weapons effort during the 1970s (Jang 2016). The contemporary political base for renewed hedging is already substantial. The 2026 Asan poll found that 80 percent of South Koreans supported acquiring an indigenous nuclear capability even as confidence in U.S. extended nuclear deterrence rose to 59 percent (Asan Institute 2026). Kim and Jeong find that support for nuclear armament is driven partly by pessimism about the future durability of American power and commitments rather than by assessments of present assurance alone (Kim and Jeong 2025). A NATO withdrawal would convert a hypothetical commitment problem into observed behavior. Seoul would have stronger incentives to demand less reversible guarantees, expand nuclear latency, or reopen an indigenous weapons option.

The European Transition Gap

Europe is wealthy enough to defend itself. The central difficulty is turning fiscal capacity into trained personnel, munitions, platforms, logistics, intelligence, command systems, and coherent operational plans before an adversary can exploit the transition.

European allies and Canada spent more than \$574 billion on defense in 2025 in constant 2021 dollars, equivalent to 2.3 percent of their combined GDP; every ally met or exceeded the previous 2 percent benchmark (NATO 2026a). Yet aggregate expenditure conceals the military functions supplied disproportionately by the United States: theater-wide intelligence, surveillance, and reconnaissance; space-based collection and fusion; strategic lift; aerial refueling; integrated air and missile defense; long-range precision strike; command and control; and the nuclear backstop.

Three major modeling exercises reach compatible conclusions despite different assumptions. The International Institute for Strategic Studies estimates that replacing the conventional American contribution would require Europe to compensate for approximately 128,000 U.S. personnel and undertake roughly \$1 trillion in procurement over a twenty-five-year equipment life cycle. Its central warning is temporal: “Europe’s window of vulnerability would open quickly” (Barry et al. 2025, 4). Burilkov and Wolff estimate that credible autonomous defense could require approximately 300,000 additional European troops and at least €250 billion in additional annual spending in the short term, including about fifty new brigades (Burilkov and Wolff 2025). RAND reaches no single price estimate, but concludes that a rapid drawdown would risk “immediate capability gaps and suboptimal defense investments” even if it generated greater spending and cooperation (Priebe, Burdette, and Behmer 2025b, vii).

These figures use different baselines and measure different combinations of forces, procurement, readiness, and recurrent costs. But their conclusion is the same. Europe could substitute for the United States over time, but the substitution would be expensive, organizationally demanding, and slowest in precisely the enabling capabilities that bind national forces into a theater-wide military system.

The political response would likely be fragmented. Poland, the Baltic and Nordic states, Germany, France, and the United Kingdom would not perceive the same threat, accept the same fiscal burdens, or agree immediately on command authority. States closest to Russia would prioritize land forces, air defense, mines, and long-range fires. France would seek greater strategic autonomy and a larger role for its nuclear force. Britain would attempt to preserve a transatlantic bridge even after the bridge had been politically demolished. Southern European governments would face pressure to divert resources from the Mediterranean, migration, and domestic priorities. Procurement decisions made under emergency conditions would duplicate systems, reinforce national industrial interests, and produce less combat power per dollar than coordinated planning inside NATO.

A sudden exit would therefore create a classic preventive-opportunity problem. European capability would be rising, but its near-term coherence would be falling. Russian leaders might conclude that waiting would allow Europe to close the gap, creating an incentive to test the new order early. European leaders, aware of the same logic, might adopt more forward-leaning deployments or declaratory policies to compensate for uncertainty. Both sides could interpret defensive preparations as evidence of offensive intent.

Revisionist Tests and the Escalation Ladder

The most likely Russian response would not be an immediate armored assault on Warsaw. It would be a campaign of calibrated testing below the threshold of general war, but with substantially greater intensity than today and likely at a higher tempo: sabotage, cyber operations, attacks on undersea infrastructure, border incidents, air and maritime probes, election interference, manufactured unrest, coercion against Moldova, and pressure on the Baltic states. Moscow would be testing whether Europe could attribute attacks, agree on a response, mobilize forces, and accept the risk of escalation without American leadership. Russia already conducts many of these operations, and the European reaction has been far from reassuring. Should the United States withdraw from NATO, however, Europe may no longer have the luxury of responding cautiously or incrementally.

The Estonian Foreign Intelligence Service assessed in early 2026 that Russia had no intention of attacking Estonia or another NATO state in the immediate term. That finding is reassuring but conditional. The same assessment emphasized that preparedness and deterrence help keep Russian intentions constrained, while Moscow’s military reforms are designed to increase future capacity (Estonian Foreign Intelligence Service 2026). Present restraint under NATO cannot be treated as evidence of future restraint after NATO’s central military power has departed.

A plausible crisis would begin ambiguously. Russian or Belarusian units might create a localized incident near the Suwałki corridor or Estonia’s eastern border such as “accidentally” crossing into Polish or Lithuanian territory during a military exercise. Cyberattacks could disrupt communications and anonymous actors could sabotage transport infrastructure. Moscow could disclaim escalation while mobilizing forces in the Leningrad Military District. European governments would confront the central problem that NATO currently mitigates: whether to treat an uncertain probe as a collective military contingency before the adversary has created a fait accompli.

Russia’s concept of strategic deterrence (strategicheskoe sderzhivanie) combined with active defense is much broader than the Western nuclear-deterrence concept: it integrates military and nonmilitary instruments across peace, crisis, and war in order to shape an adversary’s decisions and control escalation. Active defense, as articulated by Gerasimov, adds anticipatory action, surprise, maintenance of the initiative, and what Russian theorists call the “preemptive neutralization” of threats. CNA’s reconstruction of Russian military strategy describes precisely this combination of coercive signaling, high readiness, nonmilitary measures, and potentially limited force during a “threatened period” (Kofman et al. 2021).

The risk of catastrophic escalation would arise from miscalculation. Moscow might underestimate European cohesion after observing political division. A European coalition might respond more aggressively precisely because it lacked confidence in its conventional ability to sustain a prolonged conflict. Washington, having withdrawn to avoid entanglement, could face pressure to reenter after violence disrupted markets, threatened remaining bases, endangered American citizens, or appeared likely to produce Russian hegemony over Europe. Delayed intervention would occur with fewer pre-positioned forces, less access, no integrated command, and a higher risk that Russia would interpret reinforcement as preparation for regime-threatening war.

The Balkans would constitute a secondary front. NATO’s KFOR presence in Kosovo and the European Union’s Operation Althea in Bosnia and Herzegovina demonstrate that the post-Yugoslav peace remains institutionally supported. American withdrawal would not recreate the political and military conditions of the 1990s. It could, however, weaken Western coordination, encourage separatist or nationalist brinkmanship, and create opportunities for Russian political interference while European forces were concentrated on the eastern flank.

The Economic Repricing of Insecurity

The economic effects would begin before any military clash. Security is embedded in asset prices, insurance contracts, investment horizons, energy markets, and supply-chain decisions. A disorderly American withdrawal would force firms and governments to price a permanently higher probability of coercion and war in Europe.

The scale of exposure is substantial. U.S.-European Union goods and services trade was approximately \$1.5 trillion in 2024, and goods trade alone reached about \$1.05 trillion in 2025 (USTR 2026). A security shock affecting Europe would therefore transmit directly into American exports, financial institutions, pension funds, multinational earnings, energy prices, and supply chains.

The immediate European response would include a defense-spending surge financed through some combination of debt, taxation, cuts to civilian expenditure, and common borrowing. Defense production can raise measured GDP and industrial output, particularly when economies possess unused capacity. It is not costless wealth creation. Resources devoted to recreating command systems, stockpiles, air defense, strategic lift, and nuclear infrastructure are resources unavailable for civilian investment, consumption, social insurance, or debt reduction. The IISS estimates that Europe would struggle particularly with long-lead aerospace and maritime capabilities even under aggressive investment (Barry et al. 2025).

A recent CSIS gravity-model analysis estimates that NATO membership raises bilateral trade among members by 12 to 27 percent in the long run and that U.S. exports to NATO partners could fall by roughly 16 percent after withdrawal (Luck 2026). Such estimates are sensitive to model specification and cannot isolate every channel of a historically unique rupture. They are, however, consistent with a broader empirical observation: security alignment lowers political risk, supports compatible standards, and encourages trade and investment.

The global effect would depend on whether withdrawal remained a European event or initiated wider geopolitical fragmentation. IMF surveys of trade-fragmentation models place long-run global output losses between 0.2 and 7 percent under increasingly severe scenarios, with larger losses when adjustment is rapid and technological decoupling accompanies trade restrictions (Aiyar et al. 2023). NATO withdrawal alone would not generate the upper bound. It would, however, make bloc formation, sanctions conflict, defense protectionism, and strategic supply-chain duplication more likely. Investors would face a world in which the principal provider of security public goods had deliberately reduced the predictability of its own commitments.

The United States would also sacrifice economic influence over European rearmament. In the short term, emergency procurement could benefit American defense firms. Over the longer term, political abandonment would strengthen European incentives to exclude U.S. suppliers, localize production, build autonomous standards, and reduce dependence on American technology and supply chains. Washington might save part of the cost of European deployments while forfeiting a large share of the industrial and regulatory leverage those deployments helped sustain.

Nuclear Hedging After the Umbrella

The nuclear consequences are the most dangerous second-order effect. American extended deterrence has altered the cost-benefit calculation of acquiring an independent arsenal. Bleek and Lorber’s cross-national analysis concludes that “security guarantees significantly reduce proliferation proclivity among their recipients” (Bleek and Lorber 2014, 429). The historical record of West Germany, Japan, South Korea, Sweden, and others shows that technically capable states can remain nonnuclear when the combination of security guarantees, alliance influence, and nonproliferation pressure is credible.

Withdrawal from NATO would not produce a slew of European bomb programs the following morning. The initial response would be hedging. Governments would expand civil-nuclear expertise, protect enrichment or reprocessing options, develop dual-capable delivery systems, debate participation in French or British deterrence arrangements, and reduce the time required for a future weapons decision. Strategic latency would offer insurance without immediately incurring the diplomatic and legal costs of leaving the Nuclear Non-Proliferation Treaty.

The political debate is already moving in that direction. A 2026 CSIS study based on forty-two interviews found that forty respondents encountered “friendly proliferation” frequently in expert discussion, while forty-one regarded allied hedging against regional threats and abandonment as already occurring or imminent (Chevreuil 2026). Of the forty-two interviewees, twenty-one continued to oppose proliferation. The report does establish, however, that the intellectual taboo has weakened as confidence in American guarantees has declined.

Saudi Arabia illustrates how the credibility shock could extend beyond formal treaty allies. Riyadh has no Article 5-style guarantee, but it has sought stronger American security commitments while insisting on the option to develop a domestic nuclear fuel cycle. Crown Prince Mohammed bin Salman has explicitly linked Saudi nuclear acquisition to an Iranian bomb, and Saudi officials have announced an intention to pursue the "entire nuclear fuel cycle," including uranium enrichment (Davenport 2023). Haggag describes the broader regional trajectory as one of "creeping nuclearization," in which latency and hedging can spread without an immediate decision to manufacture a weapon (Haggag 2024). After a NATO withdrawal, Saudi leaders would have stronger reasons to regard any executive assurance as temporary and to preserve enrichment, missile, and supplier-diversification options as insurance. Turkey, Egypt, and the United Arab Emirates would then assess not only Iran’s program but Saudi latency, creating the conditions for a regional cascade before any state formally left the Nuclear Non-Proliferation Treaty.

France and the United Kingdom would be the obvious candidates to provide a European substitute. Their nuclear forces are formidable and survivable, but neither currently supplies the full combination of scale, forward deployment, consultation mechanisms, theater planning, missile warning, and political integration associated with the American umbrella. Extending deterrence would require decisions about command authority, financing, targeting, basing, and whether Paris or London would accept existential risk for every European partner. Separate national arsenals might complicate Russian planning, but they would also multiply decision centers and create uncertainty about escalation control.

The classic proliferation debate remains relevant. Kenneth Waltz argued that nuclear weapons can induce caution by making conquest prohibitively dangerous; Scott Sagan emphasized organizational failure, accidents, unauthorized use, and preventive-war incentives (Sagan and Waltz 2013). A post-NATO Europe would present the least reassuring version of the Waltzian case: new or expanded arsenals emerging during institutional breakdown, under intense time pressure, with contested command arrangements and adversaries capable of striking early-warning and nuclear infrastructure.

The principal danger is not simply a larger warhead count. It is a greater number of crisis dyads, doctrines, warning systems, custody arrangements, and political authorities capable of initiating escalation. European states uncertain of conventional reinforcement might rely earlier on nuclear signaling. Russia, uncertain which European commitments were credible, might disperse forces or raise readiness sooner. Shorter decision times and ambiguous command relationships would increase the probability that a conventional probe, false warning, or limited strike produced irreversible escalation.

What Washington Would Surrender

Advocates of withdrawal properly identify a real opportunity cost: forces and resources assigned to Europe cannot be assigned simultaneously to the Indo-Pacific. The strongest restraint argument therefore favors transferring a larger conventional burden to Europe so that the United States can concentrate scarce naval, air, space, and munitions capacity against China. It does not follow that dismantling NATO is the efficient means of accomplishing that objective.

NATO supplies assets that cannot be reproduced cheaply from the continental United States. European allies provide sovereign access to bases, ports, airspace, logistics networks, communications infrastructure, and intelligence sites. Austin Carson, Rachel Metz, and Paul Poast show that such access has been indispensable to American military operations even when allied combat contributions were modest (Carson, Metz, and Poast 2025). The 2025 U.S. European Command posture statement described European access, basing, and overflight agreements as enabling the United States to “intercept threats before they approach the United States” (Cavoli 2025).

The alliance also aggregates advanced economies behind American initiatives. European states contribute intelligence, sanctions enforcement, cyber capabilities, naval forces, air defense, specialized units, and diplomatic legitimacy. NATO standardization shapes defense markets and makes coalition warfare possible without improvising interoperability during a crisis. The alliance’s common budgets, worth up to €5.3 billion in 2026, support command, communications, infrastructure, and operations that would be far more expensive to reconstruct through bilateral arrangements (NATO 2026a).

A Chinese contingency would illustrate the cost of treating alliances as separable regional accounts. European forces would not determine the naval balance around Taiwan, but European bases, intelligence, sanctions, export controls, industrial capacity, and political support could affect the duration and economic sustainability of a conflict. After an American withdrawal from NATO, European governments would have weaker incentives to incur economic costs on Washington’s behalf. Some would hedge commercially toward China; others would reserve military capacity for their own continent. An exit intended to concentrate American power in Asia could therefore reduce the coalition resources available there.

Nor would withdrawal guarantee that the United States remained outside a European war. Offshore balancing assumes that Washington can intervene after a regional hegemon begins to emerge. That is logically possible, but intervention after deterrence fails is generally more expensive than sustaining the arrangements that made failure less likely. The United States entered both world wars after attempting to remain outside their initial European phases. Historical analogy does not prove recurrence, but it exposes the central gamble: withdrawal sacrifices influence over the prewar balance while preserving the possibility that American interests will later compel intervention.

The Restraint Objection

The serious case against NATO rests on three claims. Europe possesses sufficient wealth to balance Russia; American protection encourages free riding; and forward commitments create risks of entrapment in conflicts peripheral to U.S. security (Posen 2014). Beckley’s examination of American alliances also finds less evidence of entanglement than both critics and defenders often assume, suggesting that alliance costs should be evaluated empirically rather than rhetorically (Beckley 2015).

The first claim is correct in the long run. Europe can defend itself. The second is partly correct: decades of American predominance allowed some European governments to defer necessary investment. The third describes a genuine risk inherent in any commitment. None establishes that abrupt institutional withdrawal is superior to negotiated burden transfer within the alliance.

The relevant comparison is not between NATO and a costless Europe capable of immediate autonomous defense. It is between reforming an existing system and destroying it before its replacement is operational. A coordinated transition could shift conventional responsibilities, strengthen a European pillar, preserve intelligence and command integration, and retain the American nuclear guarantee while reducing permanent U.S. force requirements. A hostile withdrawal would maximize the transition gap, forfeit American leverage over allied choices, and encourage the very proliferation and military duplication that restraint seeks to avoid.

Alliances can restrain as well as entangle. American influence helped contain allied nuclear ambitions, standardize military planning, and channel national security competition through common institutions. Glenn Snyder’s alliance-security dilemma consists of both abandonment and entrapment risks; eliminating the alliance resolves one only by radically increasing the other (Snyder 1997). Europe would gain autonomy, but Washington would lose the ability to shape how that autonomy was exercised.

Conclusion: Abandonment Is Not Insulation

A United States withdrawal from NATO would not mechanically cause a Russian invasion, global recession, or nuclear war. It would produce a simultaneous strategic update across several systems that have rarely been tested without American participation. Russia would confront a temporary opportunity to probe a rearming but incompletely integrated Europe. European governments would divert large fiscal and industrial resources toward replacing American capabilities. Markets would price a higher risk of conflict and fragmentation in one of the world’s central economic regions. American allies and security partners elsewhere, particularly South Korea and Saudi Arabia, would seek more expensive guarantees, preserve sensitive fuel-cycle options, or develop strategic alternatives. Nuclear hedging would become rational for states that had previously outsourced the most destructive element of their defense.

These effects would reinforce one another. Military uncertainty would raise economic costs; economic strain would complicate rearmament; conventional weakness would increase reliance on nuclear threats; and doubts about American reliability would reduce the willingness of allies to support U.S. policy in other theaters. The outcome need not be a single dramatic collapse. A poorer and more dangerous equilibrium could emerge through cumulative hedging, duplicated military expenditure, weakened investment, proliferating arsenals, and repeated coercive crises.

NATO did not single-handedly prevent another world war. Nuclear deterrence, European integration, American alliances in Asia, international economic institutions, and the distribution of power after 1945 all contributed. NATO supplied the hard-security foundation of the Euro-Atlantic settlement: it deterred territorial revision, embedded the United States in Europe, reassured states that might otherwise have pursued independent military solutions, and made American power usable through access and interoperability.

Leaving would not free the United States from Europe’s strategic consequences. It would surrender the institutions through which those consequences have been contained. Washington would retain exposure to European war, economic disorder, Russian expansion, and nuclear escalation while possessing fewer allies, fewer bases, less intelligence, and less credibility with which to manage them. The United States would not have escaped an obligation. It would have discarded an asset.

References

Aiyar, Shekhar, Jiaqian Chen, Christian Ebeke, Roberto Garcia-Saltos, Tryggvi Gudmundsson, Anna Ilyina, Alvar Kangur, Tansaya Kunaratskul, Sergio Rodriguez, Michele Ruta, Tatjana Schulze, Gabriel Soderberg, and Juan Pedro Trevino. 2023. “Geoeconomic Fragmentation and the Future of Multilateralism.” Staff Discussion Note 2023/001. Washington, DC: International Monetary Fund. https://www.imf.org/-/media/files/publications/sdn/2023/english/sdnea2023001.pdf.

Asan Institute for Policy Studies. 2026. South Koreans and Their Neighbors 2026. Seoul: Asan Institute for Policy Studies. https://asaninst.org/bbs/board.php?bo_table=s1_6_1_eng&wr_id=25.

Barry, Ben, Douglas Barrie, Henry Boyd, Nick Childs, Michael Gjerstad, James Hackett, Fenella McGerty, Ben Schreer, and Tom Waldwyn. 2025. Defending Europe Without the United States: Costs and Consequences. London: International Institute for Strategic Studies. https://www.iiss.org/research-paper/2025/05/defending-europe-without-/-the-united-states-costs-and-consequences/.

Beckley, Michael. 2015. “The Myth of Entangling Alliances: Reassessing the Security Risks of U.S. Defense Pacts.” International Security 39 (4): 7-48.

Bleek, Philipp C., and Eric B. Lorber. 2014. “Security Guarantees and Allied Nuclear Proliferation.” Journal of Conflict Resolution 58 (3): 429-454. https://doi.org/10.1177/0022002713509050.

Brooks, Stephen G., and William C. Wohlforth. 2016. America Abroad: The United States’ Global Role in the 21st Century. New York: Oxford University Press.

Burilkov, Alexandr, and Guntram B. Wolff. 2025. “Defending Europe Without the US: First Estimates of What Is Needed.” Kiel Policy Brief 183. Kiel: Kiel Institute for the World Economy; Brussels: Bruegel. https://www.bruegel.org/analysis/defending-europe-without-us-first-estimates-what-needed.

Carson, Austin, Rachel Metz, and Paul Poast. 2025. “Allies and Access: Implications of an American Turn Away from Alliances.” International Organization 79 (S1): S103-S116.

Cavoli, Christopher G. 2025. “United States European Command Posture Statement.” Testimony before the House Armed Services Committee, April 8. Stuttgart: U.S. European Command. https://www.eucom.mil/document/43164/2025-useucom-posture-statement–hascpdf.

Chevreuil, Astrid. 2026. Friendly Proliferation: Assessing U.S. Perceptions on Proliferation Among Allies and Partners. Washington, DC: Center for Strategic and International Studies. https://www.csis.org/analysis/friendly-proliferation.

Davenport, Kelsey. 2023. “Saudi Push for Enrichment Raises Concerns.” Arms Control Today, November. https://www.armscontrol.org/act/2023-11/news/saudi-push-enrichment-raises-concerns.

Estonian Foreign Intelligence Service. 2026. International Security and Estonia 2026. Tallinn: Republic of Estonia. https://www.valisluureamet.ee/doc/raport/2026-en.pdf.

Haggag, Karim. 2024. “Seeing Through a ‘Nuclear’ Glass Darkly: Envisioning Irreversible Nuclear Disarmament in the Middle East.” Journal for Peace and Nuclear Disarmament 7 (2): 350-374. https://doi.org/10.1080/25751654.2024.2435688.

Ikenberry, G. John. 2001. After Victory: Institutions, Strategic Restraint, and the Rebuilding of Order After Major Wars. Princeton, NJ: Princeton University Press.

Jang, Se Young. 2016. “The Evolution of US Extended Deterrence and South Korea’s Nuclear Ambitions.” Journal of Strategic Studies 39 (4): 502-520. https://doi.org/10.1080/01402390.2016.1168012.

Jervis, Robert. 1976. Perception and Misperception in International Politics. Princeton, NJ: Princeton University Press.

Johnson, Jesse C., and Brett Ashley Leeds. 2011. “Defense Pacts: A Prescription for Peace?” Foreign Policy Analysis 7 (1): 45-65.

Kim, Yang Gyu, and Sangmi Jeong. 2025. “Preventive Nuclearization: Power Shifts, Anticipated Insecurity, and Public Support for Nuclear Armament in South Korea.” Journal of East Asian Studies 25 (3): 302-325. https://doi.org/10.1017/jea.2025.10012.

Kofman, Michael, Anya Fink, Dmitry Gorenburg, Mary Chesnut, Jeffrey Edmonds, and Julian Waller. 2021. Russian Military Strategy: Core Tenets and Operational Concepts. Arlington, VA: CNA, October. DRM-2021-U-029755-1Rev.

Leeds, Brett Ashley. 2003. “Do Alliances Deter Aggression? The Influence of Military Alliances on the Initiation of Militarized Interstate Disputes.” American Journal of Political Science 47 (3): 427-439.

Luck, Philip A. 2026. “NATO’s Hidden Dividend and the Avoidable Cost of U.S. Withdrawal.” Washington, DC: Center for Strategic and International Studies, April 22. https://www.csis.org/analysis/natos-hidden-dividend-and-avoidable-cost-us-withdrawal.

Monteiro, Nuno P., and Alexandre Debs. 2014. “The Strategic Logic of Nuclear Proliferation.” International Security 39 (2): 7-51.

NATO. 1949. The North Atlantic Treaty. Washington, DC, April 4. https://www.nato.int/cps/en/natohq/official_texts_17120.htm.

NATO. 2026a. “Funding NATO.” Updated April 8. https://www.nato.int/en/what-we-do/introduction-to-nato/funding-nato.

NATO. 2026b. “Defense Expenditures and NATO’s 5% Commitment.” Updated June 29. https://www.nato.int/en/what-we-do/introduction-to-nato/defence-expenditures-and-natos-5-commitment.

USTR. 2026. “European Union: Trade Summary.” Office of the United States Trade Representative, Washington, DC. https://ustr.gov/countries-regions/europe-middle-east/europe/european-union.

Posen, Barry R. 2014. Restraint: A New Foundation for U.S. Grand Strategy. Ithaca, NY: Cornell University Press.

Press, Daryl G. 2005. Calculating Credibility: How Leaders Assess Military Threats. Ithaca, NY: Cornell University Press.

Priebe, Miranda, Mina Pollmann, Naoko Aoki, Alexandra T. Evans, Kristen Gunness, Garrett Hinck, Samuel J. Leiter, Clara de Lataillade, Nathan Waechter, Nathaniel Edenfield, and Karl P. Mueller. 2025a. Anticipating Allies’ Responses to U.S. Retrenchment: Lessons from Limited Military Withdrawals During the Cold War. RR-A739-8. Santa Monica, CA: RAND Corporation. https://www.rand.org/pubs/research_reports/RRA739-8.html.

Priebe, Miranda, Zachary Burdette, and Scott Behmer. 2025b. How Might NATO Allies Respond if the United States Retrenches from Europe? RR-A2807-2. Santa Monica, CA: RAND Corporation. https://doi.org/10.7249/RRA2807-2.

Rapp-Hooper, Mira. 2020. Shields of the Republic: The Triumph and Peril of America’s Alliances. Cambridge, MA: Harvard University Press.

Sagan, Scott D. 1994. “The Perils of Proliferation: Organization Theory, Deterrence Theory, and the Spread of Nuclear Weapons.” International Security 18 (4): 66-107.

Sagan, Scott D., and Kenneth N. Waltz. 2013. The Spread of Nuclear Weapons: An Enduring Debate. 3rd ed. New York: W. W. Norton.

Schelling, Thomas C. 1966. Arms and Influence. New Haven, CT: Yale University Press.

Snyder, Glenn H. 1997. Alliance Politics. Ithaca, NY: Cornell University Press.

Stewart, Ruben. 2026. “A European Way of War Without the United States.” Survival 68 (3): 7-36. https://www.iiss.org/globalassets/media-library-/-/-content-/-migration/files/online-analysis/survival-online/2026/06/iiss_a-european-way-of-war-without-the-united-states_12062026.pdf.

United States Code. 2026. Title 22, Section 1928f, “Limitation on Withdrawal from the North Atlantic Treaty Organization.” https://uscode.house.gov/view.xhtml?req=(title%3A22+section%3A1928f+edition%3Aprelim.