# When deterrence fails: NATO, Russia and the political economy of a major-power war

**19 October 2026 | Global security-system research | Evidence cutoff: 18 October 2026**

A direct war between NATO and Russia would not be simply an expanded version of the fighting in Ukraine. It would collide two nuclear-armed security systems, integrate military and civilian infrastructure into the contest, and reorganize economic relations far beyond Europe. The immediate destruction might be concentrated around particular frontiers and maritime corridors, while effects on energy prices, public finance, trade, information, migration, technology and political institutions would travel through global networks. The longer the conflict, the greater the difference between winning a battle and preserving a functioning society. [1][3][5][7][13]

The risk has become more serious in identifiable ways without becoming certain. NATO has expanded its eastern-flank readiness, Russia has demonstrated persistence and adaptation in Ukraine, European governments have committed to a long defence-industrial expansion, and U.S.–Russian nuclear arms control has weakened. Yet preparation for conflict is not evidence that any state has decided to launch a full-scale NATO–Russia war. The Danish Defence Intelligence Service's **24 September 2026** assessment judges the risk of a **limited Russian military attack on one or several neighbouring NATO states to be low but growing**. It separately describes an outright invasion as **highly unlikely**, with no observed indications of such a decision. Estonia's foreign intelligence service assessed in **February 2026** that Russia did not intend to attack a NATO member during the following year. These assessments differ in scope, timing and analytic emphasis; neither can be converted into a numerical probability of world war. [1][2]

The Danish service's earlier **2025 conditional planning horizons** also deserve close reading: approximately **six months for a local war, two years for a Baltic regional war and five years for a major war in Europe** *if* the war in Ukraine ends or freezes, Russian capacity keeps expanding without comparable Western preparation, and Moscow judges that the United States will not support European allies. Those were capability-development estimates under a bundle of assumptions, **not dates for attacks, predictions of war, or expected durations of fighting**. The September 2026 assessment separately describes a low but increasing risk of limited attack despite continued constraints on a wider campaign. [1]

The central danger is a confrontation in which leaders intend to remain below the threshold of general war but cannot be certain how their opponent will interpret an attack, a cyber disruption, an aircraft incident or the loss of critical infrastructure. Credibility of commitments, attribution, speed of consultation and the ability to stop retaliation matter as much as declared intentions. A nuclear-armed military system can be sufficiently strong to deter deliberate invasion yet still remain exposed to miscalculation. Conversely, the very buildup intended to strengthen deterrence increases the number of aircraft, ships, sensors and weapons operating close to one another. [1][5][14]

## The evidence points to escalation risk, not a scheduled world war

The **2025 Hague NATO summit** committed members to spending **5% of GDP on defence and security-related requirements by 2035**, with **at least 3.5% for core defence** and up to **1.5% for resilience and related purposes**. The **July 2026 Ankara declaration** reported more than **US$50 billion of new procurements** and a stronger collective manufacturing commitment. A pledge is not delivered inventory; an order is not a trained unit; and procurement outlays are not an immediate measure of the equipment's availability or operational effectiveness. Nonetheless, political commitments can influence factory investments, long-term fiscal obligations and the calculations of other governments. [3][4]

SIPRI estimates world military expenditure at **US$2.887 trillion in 2025**, up **2.9% in real terms** from the previous year. Expenditure by NATO members totalled **US$1.581 trillion** on SIPRI's accounting basis, while Russia's spending was estimated at **US$190 billion**, or **7.5% of Russian GDP**. Monetary spending does not translate one-for-one into usable military power: relative wages, prices, purchasing-power differences, existing equipment, supply chains, training and readiness all matter. The NATO total includes the U.S. budget; it must not be added to U.S. expenditure as if the two were separate pools. [6]

| Observable development | Dated evidence | What can be concluded | What remains unproven |
|---|---|---|---|
| Limited NATO-border incident | Danish intelligence, September 2026 | Low but rising assessed risk | Decision to attack or predetermined trigger |
| Large conventional invasion | Danish assessment, September 2026 | Highly unlikely at that date; no indications of preparation | Absolute impossibility or a stable long-term forecast |
| Estonian one-year outlook | February 2026 | No Russian intent to militarily attack NATO in coming year, in that service's assessment | Applicability to all later incidents or every form of coercion |
| NATO spending commitment | Hague 2025 and Ankara 2026 | Long-term defence and industrial mobilisation | Immediate availability of all planned capabilities |
| New START expiration | February 2026 | Bilateral strategic nuclear limitation treaty ended without successor | Immediate deployment of every technically possible additional warhead |
| Airspace incidents and allied air defence | NATO statements 2025–2026 | Documented security consultations and enhanced defensive activity | Attribution of every later anonymous drone, cyber or sabotage incident |

The table summarizes official assessments rather than averaging them. Intelligence reports describe judgments, not direct access to every decision-maker's intent. [1][2][3][4][8][27]

```chart
type: bar
title: Military expenditure by selected states, 2025
unit: US$ billion at market exchange rates; SIPRI estimates, states are separate; NATO aggregate is omitted to avoid overlapping US spending
United States | 954
China | 336
Russia | 190
Germany | 114
India | 92.1
```

The United States, China, Russia, Germany and India were SIPRI's five largest military spenders in 2025. That ordering says nothing by itself about superiority in a particular theatre. The geography of deployment, logistic distance, actual stocks, integration with allies, the sustainability of production and political control of escalation can overturn comparisons based on gross budget figures. Moreover, spending totals reflect previous choices and national accounting conventions, not future mobilization output. [6]

## Why Europe became a system of frontiers, commitments and contested security arrangements

This confrontation has historical layers, but no single historical grievance determines today's choices. After the Cold War, NATO enlargement and the sovereign right of European states to choose alignments developed alongside attempts to construct cooperative relations with Russia. The **1997 NATO–Russia Founding Act** institutionalized cooperation without transferring a veto over accession. NATO's **2008 Bucharest declaration** stated that Ukraine and Georgia would become members without providing a membership timetable. Russian leaders increasingly treated enlargement and neighbouring alignments as threats to their security and status; many states in Eastern Europe saw NATO membership as protection against coercion. Ukraine's choices are those of an independent state, not merely a bargaining object between Moscow and Washington. [5][13][30][31]

Russia's annexation of Crimea and the outbreak of war in eastern Ukraine in 2014 destroyed much of the credibility of the earlier security settlement. The full-scale invasion in 2022 turned that contested order into protracted interstate war. Explaining the sequence of threat perceptions is not equivalent to assigning symmetrical legal responsibility. The UN Charter prohibits the use of force against states' territorial integrity and preserves individual and collective self-defence when an armed attack occurs. The institutional problem for a future settlement is how to provide verifiable security without making neighbouring states' sovereignty depend entirely on a stronger power's unilateral interpretation of its own safety. [13]

NATO itself is not one state. Poland, the Baltic states, Germany, France, the United Kingdom, the United States and Türkiye have distinct geography, supply chains, political constraints, nuclear roles and opinions about the distribution of defence costs. **Article 4** provides for allied consultation when security is threatened. **Article 5** treats an armed attack against one as an attack against all but requires each ally to take such action as it **deems necessary**, including armed force. This is a binding collective-defence obligation, not an automatic algorithm sending identical numbers of soldiers from every member. Russia must also weigh the response of multiple governments rather than assume that a single political statement guarantees their future behavior. [5][27]

## Geography determines how escalation spreads

The Baltic Sea combines short military distances with dense civilian activity: ports, maritime traffic, energy facilities, cables, aviation and population centres. Unidentified drones, aircraft interceptions and incidents at sea become dangerous when identification and command decisions must occur rapidly. In 2025, Russian drone incursions into Polish airspace and an incursion by Russian jets into Estonian airspace led to NATO Article 4 consultations and strengthened air defence. Their documentation confirms tension and responses; it does not establish that each incident was intended to trigger general war. [27]

The Black Sea ties Ukrainian agricultural exports, ports, Romanian and Bulgarian security, Turkish maritime interests and shipping insurance. Interruptions can increase the cost of moving grain or energy before they reduce total agricultural or hydrocarbon production. The Arctic and North Atlantic matter for reinforcement routes, submarine activity, maritime communications and military access. The Mediterranean and Middle East could transmit concurrent energy and maritime risks, but a major European war would not mechanically bring every regional power into the fighting. [6][14][18][19]

```map
title: Geographic transmission corridors in a hypothetical wider European war
Baltic Sea and eastern NATO flank | Short military response times, civilian airspace and maritime infrastructure | Incidents can spread politically before their cause is independently verified
Black Sea and Turkish Straits | Grain, ports, shipping access and coastal states | Maritime trade disruption need not imply universal closure of all routes
North Atlantic and Arctic | Reinforcement, navigation, communications and access | Weather, distance and allied basing alter capacity
Central and Western Europe | Rail, electricity, hospitals, factories and host communities | Mobilisation competes with civilian service continuity
Middle East and global energy routes | Hydrocarbon supply, pricing and insurance | European war would interact with but not predetermine other regional conflicts
Americas, Asia and Africa | Credit, trade, fertiliser, food and technology imports | Exposure differs by reserves, contracts, production and import dependence
```

These are spatial transmission mechanisms, not simulated routes of attack. Civilian links are relevant because disruption affects households and economic activity, not because all civilian assets become lawful military targets.

## Conventional force does not equal headline numbers of tanks and aircraft

A large-power confrontation would combine ground, air, maritime, electronic, cyber and space-related capabilities. Land forces still need protected movement, engineering support, ammunition, medical evacuation, fuel, repair shops and trained replacements. A nominally large formation unable to sustain equipment or transport fuel has different operational value from an integrated formation with support. Air power needs warning networks, maintenance, munitions and suitable airfields, while air and missile defence must detect, classify and respond under pressure. Naval forces face ship repair, logistics and maritime-awareness constraints. [9][10][28]

In Ukraine, inexpensive drones, longer-range uncrewed aircraft and precision-guided munitions have changed the exposure of soldiers, vehicles, artillery and rear-area infrastructure. The **SIPRI Yearbook 2026** describes rapidly proliferating missiles and armed drones, increasing use in different strike roles, and a parallel expansion of counter-drone systems. But a drone shown in a demonstration is not a proven wartime replacement for aircraft, heavy industry or logistics. Performance depends on production, communications, electronic interference, weather, tactical adaptation and the availability of human operators. [10]

Electronic warfare, sensing, satellites and resilient communications form a contest over the reliability of information. Interference can degrade navigation or links; automated processing may shorten analysis time but can introduce error and opaque decisions. SIPRI's 2026 work on military AI procurement identifies the pressure to acquire capabilities quickly and the corresponding need for testing, evaluation and accountability. Model performance in a demonstration says little about robustness across complex civilian-military environments, false identification or adversarial disruption. The human and legal responsibility for an attack is not eliminated because software helped generate a recommendation. [11][21]

Chemical and biological hazards also require separate legal and public-health analysis, not sensational predictions. The **Chemical Weapons Convention** prohibits the development, production, stockpiling and use of chemical weapons and establishes an international verification framework through the OPCW. Biological weapons are likewise prohibited under a distinct convention, but public reporting of a suspected chemical or biological incident would require careful investigation of exposure, cause and attribution before being treated as an established attack. Industrial capacity for protective equipment, clinical diagnosis and medical response matters regardless of any claim about an adversary's intentions. [32]

The strategic industrial issue is endurance. Manufacturing munitions and interceptors requires labour, production lines, specialized components, testing, energy and long orders. A country can purchase expensive platforms during peacetime but still lack affordable defensive supplies for persistent attacks. Conversely, a workforce and factory network that scales production must obtain raw materials, electronics, explosives inputs and maintenance capacity. European Commission plans through **Readiness 2030** specifically emphasize industrial output, cross-border military mobility, air defence, drones and space capabilities. Targets and proposed coalitions remain distinct from measured annual production and deliveries. [9][10]

## Nuclear weapons change the decision problem even when none is used

SIPRI estimates that in January 2026 nine nuclear-armed states together possessed **12,187 nuclear warheads**, including **9,745 in military stockpiles** and about **4,012 deployed**. The U.S. military stockpile was estimated at **3,700** and Russia's at **4,400**; each country's total inventory, which also counts retired warheads awaiting dismantlement, is larger. Stockpile estimates have uncertainty and should not be confused with the number ready for immediate launch. [7]

```chart
type: bar
title: Estimated military nuclear warhead stockpiles, January 2026
unit: number of warheads in military stockpiles; excludes retired warheads awaiting dismantlement; SIPRI estimates
Russia | 4400
United States | 3700
China | 620
France | 290
United Kingdom | 225
```

**New START expired in February 2026**, weakening bilateral limits and verification of U.S. and Russian strategic forces. The expiry does not by itself mean a state has already increased the number of deployed weapons, but it reduces the institutional framework for observing, limiting and discussing future changes. Meanwhile, systems capable of carrying either conventional or nuclear payloads can complicate interpretation during a crisis. A conventional strike may be perceived as a threat to a nuclear command system even when its initiator believes the operation to be limited. [7][8]

Escalation should be understood as a ladder of political and operational risks rather than an inevitable sequence. A conventional clash might end with rapid disengagement. It might instead produce attacks against military infrastructure, threats involving nuclear readiness, or pressure to demonstrate resolve. Any nuclear detonation would introduce mass casualties, long-lived health and environmental effects, overwhelmed health systems and pressure for further retaliation. The ICRC warns that a large nuclear exchange would have catastrophic consequences beyond the belligerents, including damage to the environment and food production. Quantitative climate and mortality models depend strongly on assumptions about which weapons are used, the fires produced and agricultural responses; they are not a forecast of a specific future exchange. [7][12]

A war also poses distinct nuclear safety hazards at civilian power plants. Loss of external power, workforce pressure, damaged cooling and emergency logistics can create danger without any nuclear weapon being detonated. Civilian nuclear facilities and hospitals remain protected by the law of armed conflict; descriptions of vulnerabilities are not permissions to attack them. [12][21]

## Cyber and space capabilities connect military operations to ordinary life

Positioning, communications, cloud computing, mapping, electronic payments and transport schedules depend on networks whose components are often privately owned and shared by civilian and defence users. Loss of a satellite service could impede both military coordination and civilian logistics; a payments disruption could damage commerce without changing who legally owns a single bank deposit. The International Committee of the Red Cross warns that cyber operations can impair hospitals, electricity, water, rail and financial services and that the protection of civilian infrastructure remains applicable in cyberspace. Attribution can be difficult, especially at the beginning of an incident. [21]

The escalation danger is double. A government may read a network interruption as preparation for armed attack; a defender may overestimate the intention or effects of an intrusion; and technical components could fail for reasons unrelated to hostile activity. Consequently, credible public analysis must distinguish a reported outage, technical cause, independently supported attribution and verified material consequences. Labelling every unexplained fault as sabotage would be analytically unsound and socially destabilizing.

The global economy would be exposed through dependencies it may not perceive as military: undersea cables, satellite navigation, chips, cloud regions, specialized cybersecurity personnel, replacement transformers, data centres, logistics platforms and power suppliers. Where resilience exists, private and public operators may restore services through redundancy. Where shared providers are concentrated, a small technical failure may have outsized consequences. These possibilities require service-specific evidence rather than an invented list of targets or instructions for disruption. [9][21][23]

## A major-power war reallocates labour, capital and productive capacity before it creates new prosperity

The first financial shock would likely be repricing and liquidity stress: investors would reassess counterparty risk, energy prices, currency exposure, government debt and the expected duration of interruption. Yet **a price increase is not a transfer of ownership**. Higher yields immediately change the market value of existing fixed-income securities; fiscal interest costs on fixed-rate debt rise gradually as new debt is issued and old obligations mature. Fund redemptions, margin calls and changes in collateral could accelerate market selling before underlying goods shortages become visible. The IMF's **April 2026 Global Financial Stability Report** documents this kind of amplification risk from leveraged nonbank finance and emerging-market cross-border flows. [23]

```flow
Direct fighting → damaged infrastructure and labour absences → lower output and fiscal revenue → higher medical, security and reconstruction needs → financing pressure
Defence procurement → orders for materials and equipment → production if workers, inputs and power are available → possible civilian crowding-out or later technical spillovers
Energy and shipping disruption → delivered import costs and insurance → business margins, inflation and currency conditions → household real income and political pressure
Market fear → securities repricing and funding withdrawals → collateral and refinancing stress → weaker private investment even outside the war zone
```

More military spending can raise measured GDP in the short run. But the goods purchased may be used or destroyed rather than adding to civilian housing, health, transport or future consumer productivity. It is also possible for some defence technologies to generate civilian applications. The net result depends on financing, spare industrial capacity, what would otherwise have been produced, where purchases are imported from and whether a future peace allows reuse. It is neither automatically an economic stimulus nor automatically an equal decline in civilian output. The IMF's 2026 work on conflict-related macroeconomic scarring finds persistent output losses in directly affected economies and slow, uneven recovery. [20]

Public budgets could come under simultaneous pressure from defence, interest, hospitals, housing support, refugee reception and infrastructure. Taxation, inflation, bond issuance, spending cuts and exceptional financing distribute costs differently across taxpayers, creditors, wage earners and future generations. In countries with high existing debt, a sustained war premium in borrowing costs can crowd out unrelated investment. In countries with effective tax systems and domestic production, some procurement spending can circulate locally; in import-dependent states, equivalent expenditure may weaken the trade balance. There is no uniform war multiplier.

Russia's fiscal and social trade-off is especially evident in the SIPRI accounts: the organization estimated that military expenditure absorbed **7.5% of GDP and approximately 20% of Russian government spending in 2025**, alongside reduced budget transparency. Those figures concern the Russian war economy in 2025, not a forecast of a future direct NATO war. Europe's planned expenditure requires its own comparison of borrowing conditions, industrial capability, ageing populations and civilian commitments. [6]

## Energy, food and resources would carry the shock to countries outside Europe

Energy price effects begin with geography and physical logistics. Crude oil production, actual export volume, gas pipeline or LNG availability, refining, shipping insurance, inventories and final retail prices are not interchangeable measures. New sanctions could reroute exports or alter discounts without reducing total global production by the same amount. A disruption involving European ports or the Black Sea could raise freight and insurance even if alternative routes remain open. Other regional wars and climate-related supply disruptions could intensify the same shock, but their consequences must not be attributed mechanically to a NATO–Russia conflict. [20][23]

Food security depends on more than grain stocks. Fertilizer feedstocks, fuel, working capital, crop insurance, port operations, export licensing and local purchasing power determine how an international price change affects families. Some agricultural exporters could receive higher prices for available surplus; net food importers and low-income households would face greater vulnerability. Export controls might temporarily protect domestic markets yet increase international scarcity and encourage neighbouring restrictions. Existing food insecurity would condition harm, so a future assessment should examine calories and diet diversity separately from headline global crop supply. [20]

```map
title: Unequal global economic exposures to a major-power confrontation
European frontline economies | Direct security risk, electricity, housing and mobility | Severe effects depend on actual geography and intensity
Western Europe | Industry, energy contracts, public borrowing and migration | Import substitution takes time and may raise costs
Russia and neighbouring Eurasian economies | War mobilisation, sanctions, labour and commodity exports | Nominal receipts differ from delivered volumes and civilian welfare
East and Southeast Asia | Semiconductors, shipping, machinery and markets | Each state's exposure varies by trade, neutrality and supply-chain role
Middle East and North Africa | Hydrocarbons, wheat, fertiliser and maritime insurance | Energy exporters and food importers have different net positions
Latin America and sub-Saharan Africa | Food, fertiliser, metals, exchange rates and global credit | Distribution depends on import dependence, fiscal capacity and reserves
```

Technological and resource rivalry would extend to aerospace components, advanced semiconductors, optics, electricity equipment, industrial machinery, metals and satellite-related services. Scarcity is material only where production, refining, qualification or logistics cannot be replaced within the necessary time and cost. Describing an input as “strategic” does not prove there is a worldwide physical shortage. Sanctions require separate analysis of legal scope, enforcement, substitution routes and observed volumes. [9][10][23]

For Brazil, a war's primary mechanism would not necessarily be direct military participation. Exposure could arrive through diesel and fertilizer prices, crop markets, freight and insurance, exchange-rate fluctuations, international interest rates, industrial parts and demand for exports. Commodity exporters may benefit in some contracts and lose through higher imported inputs or transport. Households with lower income typically devote a greater share to essential goods and face more immediate welfare pressure when food and transport costs rise. None of this implies Brazil would automatically join the fighting or that every price rise would improve its trade position.

## Health, displacement, demography and communities become part of the war's duration

War is not only the number killed on the battlefield. It changes household structures, the availability of medical staff, fertility decisions, school progression, intergenerational wealth and the ability of older people to remain in their homes. The World Bank, European Commission, UN and Ukrainian authorities estimated almost **US$588 billion in Ukrainian recovery and reconstruction needs over the following decade**, based on damage and needs assessed through **31 December 2025**. This is a forward-looking cost estimate, not an annual budget, not the cash already spent and not a forecast of what an expanded conflict would cost. [18]

The WHO's February 2026 Ukraine emergency appeal estimated **4.1 million people requiring health assistance** and stated that a mental-health crisis affected an estimated **10 million** people in the country. Such needs include anxiety, trauma, grief, chronic disease, disabilities and disrupted access to care, but the figures are planning estimates and do not establish 10 million psychiatric diagnoses. The existence of severe stress does not make an entire nationality psychologically homogeneous. [19]

Displacement also transfers the burden to host societies: classrooms, housing, municipal transport, health services and local labour markets must adapt. At the same time, migrants can contribute employment, skills and tax revenues; displacement is not solely an economic burden. Return depends on security, destroyed homes, legal status, family reunification and the availability of jobs, not just the signing of a ceasefire. Children may grow up in a different language and education system from the one their parents left, making demographic consequences persist across generations. [18][19]

Long wars redistribute political voice. Veterans and families of the missing acquire claims on pensions, medical care and public recognition. People employed in defence production may have different interests from workers displaced from civilian industries. Shortages and prolonged separation alter care responsibilities and can affect marriage, birth decisions and internal migration. These are plausible channels requiring population-level evidence; their magnitude in a hypothetical global war cannot responsibly be calculated from Ukraine alone.

## Psychological mobilisation can sustain defence and corrode civil life

Collective fear can initially generate solidarity and voluntary assistance. It can also generate exhaustion, panic buying, suspicion of outsiders, conspiracy narratives and conflicts over unequal sacrifice. The direction depends on trust, perceived fairness of mobilisation, casualty communication, the protection of families and credible prospects for safety. Psychological outcomes among civilians cannot be inferred simply from military spending or patriotic public statements.

A prolonged climate of alarms and uncertainty can disrupt sleep, children's development, concentration and expectations about the future even in locations outside actual battles. Social reactions also vary by prior trauma, material security and age. Education and health services may come under pressure when staff are mobilised or relocated. Institutions that provide reliable information, continuity of basic goods, mental-health care and safe public discussion can limit additional injury. Conversely, deliberately humiliating public communication or unsubstantiated threat claims can undermine legitimacy and damage collective judgment.

The state must also manage internal disagreement. Citizens can support national defence while opposing a particular military measure, executive restriction or spending allocation. In Russia, Ukraine and NATO countries, media access and political institutions differ considerably, and polling conducted during wartime can suffer nonresponse, social-desirability bias and risks to respondents. It is unsound to equate publicly reported approval of an operation with unlimited willingness to accept personal mobilisation, or to assign a fixed “national mentality” to entire populations.

## Information control, media restrictions and digital platforms would be contested terrain

Modern wars affect the production and validation of information. Governments issue security instructions and battlefield claims; militaries restrict disclosure of some operational information; platforms moderate dangerous or manipulated content; intelligence services identify influence operations; journalists face physical and legal risks; and civilians rely on private messaging services for safety information. These activities are not one phenomenon. **A government claim, an independently verified event, a deliberately manipulated message and a private platform's moderation action must be distinguished.** [22][24][25]

The EU's External Action Service issued its **fourth annual report on foreign information manipulation and interference in March 2026**. It documents the institutional view that organized campaigns can operate across channels and languages. Its report is not independent proof that every dissenting voice is foreign-controlled. European legal acts have also restricted broadcasting by certain Russian state-linked outlets; the **2022 EU regulation concerning RT and Sputnik** stated security and public-order grounds and was contested on freedom-of-expression grounds before EU courts. Those measures must be described with their specific scope rather than equated with total censorship of all Russian-language journalism. [24][25]

The **EU Digital Services Act**, Article 36, establishes a mechanism through which the European Commission, under specified crisis conditions, can require very large platforms and search services to assess and mitigate serious risks. The text includes requirements to consider fundamental rights and the proportionality of action. It does not mean any government can legally suppress all criticism in an emergency. Different jurisdictions may adopt more intrusive restrictions, however, and emergency measures can persist beyond the original justification if there is no sunset rule, independent review or public reporting. [26]

AI-generated audio, video and large-scale automated content can increase the speed of misinformation; they can also support translation, verification and emergency communication. False attribution is especially dangerous during a short military crisis. A forged video or misleading account of a civilian incident might influence public expectations faster than forensic verification can occur. Information integrity therefore requires authenticated official communications, independent journalism, access to original evidence, correction mechanisms and protection against unlawful surveillance. The ICRC also warns that civilians who participate directly in cyberhostilities can face significant risks and that harmful information can amplify physical and psychological injury. [21][22]

```mindmap
Information and social response during prolonged conflict
- Public authority
  - emergency communications and evidence updates
  - legal restrictions with judicial and legislative review
- Independent information
  - journalists, researchers and civil society
  - provenance, correction and protected access
- Digital infrastructure
  - platforms, messaging, satellite and telecommunications
  - cyber disruption and data privacy
- Society
  - solidarity and collective protection
  - misinformation, polarisation, fear and protest
- Legitimacy
  - proportionality, equal treatment and time limits
  - accountability after exceptional powers expire
```

Media freedom is not irrelevant to security. Independent reporting can expose corruption in procurement, avoidable civilian deaths and failures of public services. Restrictions that suppress such evidence can make a military system less able to learn from mistakes. At the same time, publishing identifiable information that creates immediate risks to protected persons can itself be harmful. A rights-respecting response needs clear legal definitions, necessity, proportionate scope, review and remedies, not slogans about complete freedom or complete control.

## How long could such a conflict last?

No credible source can specify the duration of an unrealized NATO–Russia war. Even the phrase “world war” contains different situations: one incident among nuclear-armed powers, a geographically limited campaign, a sustained Europe-wide conventional war, simultaneous conflicts on more than one continent, and direct nuclear exchange are different systems. Duration depends on military objectives, political ceilings, industrial replacement, casualties, third-party decisions, diplomatic channels and the capacity of societies to bear costs.

| Conditional pathway | Plausible analytic horizon, not forecast | Mechanism extending or limiting duration | Principal civilian consequence |
|---|---|---|---|
| Incident followed by managed disengagement | Hours to weeks | Rapid attribution, hotlines, consultations and credible reciprocal restraint | Short-lived fear, service disruption and repricing |
| Limited conventional confrontation | Weeks to months, potentially longer | Narrow objectives, credible ceasefire and boundaries to retaliation | Local casualties and wider transport and market shock |
| Sustained regional conventional war | Months to years | Mobilisation, replenishment, logistics, contested objectives and allies' support | Destruction, displacement, fiscal deterioration and civilian rationing |
| Persistent coercion below open war | Years | Cyber incidents, sabotage allegations, sanctions and unresolved security disputes | Chronic insecurity and higher infrastructure costs |
| Nuclear use or uncontrolled multi-theatre escalation | No defensible standard duration | Command decisions and retaliation dynamics can change in hours | Potential catastrophic damage with consequences lasting generations |

The horizons are conceptual, not probabilities. A short armed clash can yield decades of political consequences; a long period of cyber coercion need not create a conventional war. Planning for both rapid crisis containment and multi-year societal resilience is more defensible than choosing a single median duration.

Military exhaustion alone does not guarantee peace. States may continue fighting despite severe losses if leaders believe withdrawal endangers regime survival or national security, if external supplies continue, or if no credible implementation mechanism exists for a settlement. Equally, even strong forces may limit their goals when nuclear escalation risks outweigh possible territorial or political gains. The most dangerous assumption would be that one side can safely determine the other's escalation threshold in advance.

## How wars stop: distinct diplomatic outcomes and their trade-offs

A **unilateral halt** may reduce casualties without establishing mutual protection. A **ceasefire** suspends specified hostilities and needs a definition of prohibited acts, geography, timing and verification. An **armistice** provides more durable military arrangements without necessarily resolving territorial and political claims. A **peace treaty** addresses a wider set of legal obligations, while lasting stability requires implementation, political legitimacy and the absence of incentives for renewed force. These are separate legal and social outcomes. [13][14]

A narrow emergency arrangement might begin with direct military communication, independent verification of incidents, local disengagement and protection of essential civilian services. A broader settlement could require rules on forward deployments and exercises, air and maritime incident prevention, verification or observation missions, treatment and exchange of detainees, return and protection of displaced people, cross-border infrastructure repair, humanitarian access and arrangements for contested claims. Monitoring is a capability, not merely a name: it needs mandate, access, staffing, safety, reporting rights and consequences for verified breaches. [14]

The **OSCE Vienna Document** offers examples of notification, information exchange and inspection mechanisms designed to reduce misunderstandings. New arrangements would confront severe trust deficits, different threat perceptions and unresolved questions of sovereignty. A framework cannot credibly presume that Ukraine is absent from decisions directly affecting its future. Nor should a ceasefire automatically be presented as legal recognition of territorial changes achieved by force. The UN Charter's rules on the use of force and self-defence remain relevant throughout negotiations. [13][14]

Nuclear stabilization would require parallel work even if the conventional war were controlled. Recreating channels for strategic stability talks, notification, verified limits and incident management could reduce mistaken interpretation of military activity. Sanctions relief, frozen assets, compensation, security assurances and reconstruction finance are distinct instruments. Sanctions can be lifted, suspended or conditioned; their effects depend on enforcement, financial intermediaries and actual trade, not simply diplomatic announcements. Arrangements that cannot be audited risk leaving all sides believing their opponent is cheating.

The possible endings are therefore plural. A **rapid negotiated de-escalation** might end combat while leaving a long-term competitive military frontier. A **protracted armistice** could freeze certain deployments and permit reconstruction without resolving core claims. A **negotiated wider security framework** could lower incident risk but would require verifiable compromises and political consent. An **exhaustion-driven outcome** might produce a ceasefire under acute fiscal or social pressure yet remain fragile. A **nuclear escalation** would destroy the ordinary assumptions under which economic reconstruction, bargaining and civilian protection are planned. None is a forecast or recommendation to impose an arrangement on the affected population.

```flow
Crisis containment → verified communication and limits on incidents → local disengagement → monitored ceasefire → humanitarian access and emergency repair
Monitored ceasefire → force separation and inspection arrangements → verified implementation → wider security dialogue and sustainable reconstruction
Failure of verification or renewed coercion → erosion of confidence → remobilisation and recurrent escalation risk
```

An important distinction is between **the end of shooting** and **the end of the conflict system**. Weapons production, alliance changes, debt, damaged habitats, missing persons, traumatised families, sanctions and migration patterns can persist for decades. Peace can be materially improved while political competition remains intense; institutional trust may recover more slowly than roads and buildings.

## The world after a ceasefire could remain more militarised and less integrated

Ten or fifteen years after a major war, the distribution of industrial capacity may have shifted toward defence, energy security and supply-chain redundancy. Companies can relocate plants, governments can hold larger reserves, banks can demand political-risk premiums, and households can maintain higher precautionary savings. Yet decoupling is rarely total: trade may be rerouted through third countries, new intermediaries may gain profits and suppliers may change without underlying technological dependence disappearing. The existence of an alternative payments platform does not necessarily replace the liquidity and legal network effects of major currencies. [20][23]

The ecological inheritance could include damaged soil, toxic residues, unexploded ordnance, damaged power plants and interrupted environmental monitoring. Reconstruction would require years of engineering, land surveys, health services and safe clearance. Demographic change could be even more persistent: births deferred, young workers leaving, communities aging and children educated abroad. The World Bank's Ukraine reconstruction assessment and the WHO health appeal illustrate how physical assets and human capabilities are jointly damaged, but they cannot serve as a numerical template for a future continental conflict. [18][19]

Postwar democracy and institutional legitimacy depend on how exceptional powers end. Surveillance practices, military contracting, restrictions on access to information and special fiscal arrangements may survive after their initial security purposes disappear. Legislative oversight, judicial remedies, public procurement audits and preservation of independent journalism affect whether exceptional powers are rolled back. Other societies may experience pressure for centralisation or political fragmentation. The outcome depends on domestic institutions, electoral choices and evidence of abuses; a war does not impose one political model on every participant.

The longer-run structure might thus be neither a restored pre-2022 European order nor universal confrontation. Multiple states could sustain deterrence and limited economic exchange simultaneously. Durable reduction of risk would be evidenced by lower military proximity, functioning communication and verification, fewer serious incidents, credible civilian infrastructure protection and a political settlement that affected populations can live with.

## Observable warning signs, counterevidence and the limits of current knowledge

The immediate indicators worth tracking are serious air or maritime incidents with independently verified attribution; evidence of major changes in force posture rather than announcements alone; allied consultations and agreed responses; repeated damage to critical civilian services; verified changes in nuclear readiness or arms-control arrangements; defence production and deliveries relative to procurement commitments; and the availability of military-to-military communication. In finance, watch credit spreads, sovereign refinancing costs, capital flows, energy delivery and shipping insurance rather than one-day equity moves. In society, watch actual displacements, school attendance, medical continuity, labour mobilisation and credible longitudinal measures of trust.

There is material counterevidence to a deterministic countdown. Estonian intelligence found no intention to attack NATO in its February 2026 horizon; Danish intelligence in September still judged outright invasion highly unlikely; Article 5 collective defence and nuclear deterrence raise the prospective cost of a deliberate large war; and NATO's new investments may reduce rather than increase the incentive for an attack. These factors do not nullify the rising risk of a limited incident. They indicate that **more tension, more spending and a higher risk of miscalculation do not prove a decision to start global war**. [1][2][3][7]

Open-source estimates of forces, classified decision-making and public opinion are incomplete. SIPRI figures are estimates; intelligence judgments may change; reported incidents require independent attribution; models of economic loss rely on uncertain duration and geographic extent; and nuclear consequences become highly sensitive to the scale of use. A rigorous assessment therefore avoids single-number forecasts of fatalities, the date of war, market collapse or the hour at which nuclear weapons would be used.

A major-power war could transform the international system not only through territorial control but through the loss of people, institutions and productive capacity that societies take generations to rebuild. The most consequential question in October 2026 is not whether war has become conceivable — it has — but whether governments can simultaneously improve credible defence, reduce ambiguity, maintain communication and preserve civilian conditions that make prolonged peace politically sustainable.

### Sources

1. Danish Defence Intelligence Service, *Assessment of the threat from Russia*, 24 September 2026: https://www.fe-ddis.dk/globalassets/fe/dokumenter/2026/trusselsvurderinger/-assessment-of-the-threat-from-russia-.pdf
2. Estonian Foreign Intelligence Service, *International Security and Estonia 2026*, 10 February 2026: https://valisluureamet.ee/doc/raport/2026-en.pdf
3. NATO, *Ankara Summit Declaration*, 8 July 2026: https://www.nato.int/en/about-us/official-texts-and-resources/official-texts/2026/07/08/the-ankara-summit-declaration
4. NATO, *The Hague Summit Declaration*, 25 June 2025: https://www.nato.int/en/about-us/official-texts-and-resources/official-texts/2025/06/25/the-hague-summit-declaration
5. NATO, *North Atlantic Treaty*, Articles 3–5: https://www.nato.int/docu/basictxt/treaty.htm
6. SIPRI, *Trends in World Military Expenditure, 2025*, April 2026: https://www.sipri.org/publications/2026/sipri-fact-sheets/trends-world-military-expenditure-2025
7. SIPRI, *SIPRI Yearbook 2026: World Nuclear Forces*, June 2026: https://www.sipri.org/media/press-release/2026/increasing-focus-nuclear-weapons-amid-heightened-escalation-risks-new-sipri-yearbook-out-now
8. US Department of State, *New START Treaty*, treaty duration and extension: https://2021-2025.state.gov/new-start/
9. European Commission, *Readiness Roadmap 2030*: https://defence-industry-space.ec.europa.eu/eu-defence-industry/readiness-roadmap-2030_en
10. SIPRI, *Yearbook 2026: Proliferation and use of missiles and armed uncrewed aerial vehicles*: https://www.sipri.org/yearbook/2026/09
11. SIPRI, *Responsible Procurement of Military Artificial Intelligence*, 2026: https://www.sipri.org/publications/2026/other-publications/responsible-procurement-military-artificial-intelligence
12. ICRC, *Frequently asked questions: Nuclear weapons*, 6 August 2026: https://www.icrc.org/en/article/faq-nuclear-weapons
13. United Nations, *Charter of the United Nations*, Articles 2 and 51: https://www.un.org/en/about-us/un-charter/full-text
14. OSCE, *Confidence- and Security-Building Measures / Vienna Document*: https://www.osce.org/secretariat/107484
15. NATO, *NATO Integrated Air and Missile Defence*, 7 October 2026: https://www.nato.int/en/what-we-do/deterrence-and-defence/nato-integrated-air-and-missile-defence
16. NATO, *Strengthening NATO's Eastern Flank*, 29 September 2026: https://www.nato.int/en/what-we-do/deterrence-and-defence/strengthening-natos-eastern-flank
17. UNHCR, *Ukraine Refugee Situation, Operational Data Portal*, accessed October 2026: https://data.unhcr.org/en/situations/ukraine
18. World Bank, Government of Ukraine, EC and UN, *RDNA5*, 23 February 2026: https://www.worldbank.org/en/news/press-release/2026/02/23/updated-ukraine-recovery-and-reconstruction-needs-assessment-released
19. WHO, *Ukraine Health Emergency Appeal 2026*, 3 February 2026: https://www.who.int/publications/m/item/ukraine--who-health-emergency-appeal-2026
20. IMF, *World Economic Outlook — Global Economy in the Shadow of War*, April 2026: https://www.imf.org/en/publications/weo/issues/2026/04/14/world-economic-outlook-april-2026
21. ICRC, *International humanitarian law and cyber operations during armed conflicts*: https://international-review.icrc.org/articles/ihl-and-cyber-operations-during-armed-conflicts-913
22. ICRC, *Cyber operations and harmful information*: https://www.icrc.org/en/law-and-policy/cyber-and-information-operations
23. IMF, *Global Financial Stability Report*, April 2026: https://www.imf.org/en/publications/gfsr/issues/2026/04/14/global-financial-stability-report-april-2026
24. European External Action Service, *Fourth Annual Report on Foreign Information Manipulation and Interference*, 12 March 2026: https://www.eeas.europa.eu/eeas/4th-eeas-annual-report-foreign-information-manipulation-and-interference-threats_en
25. EU Council, *Regulation 2022/350* (broadcasting restrictions): https://eur-lex.europa.eu/legal-content/en/ALL/?uri=CELEX%3A32022R0350
26. European Union, *Digital Services Act, Regulation 2022/2065*, Article 36: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A32022R2065
27. NATO, *The consultation process and Article 4*, 2025: https://www.nato.int/en/what-we-do/introduction-to-nato/the-consultation-process-and-article-4
28. NATO, *Deterrence and Defence*, 2026: https://www.nato.int/en/what-we-do/deterrence-and-defence/deterrence-and-defence
29. UN Human Rights Monitoring Mission in Ukraine, *Protection of Civilians — September 2026*, published 15 October: https://ukraine.ohchr.org/en/reports/protection-of-civilians

30. NATO, *Founding Act on Mutual Relations, Cooperation and Security between NATO and the Russian Federation*, 27 May 1997: https://nato.int/en/about-us/official-texts-and-resources/official-texts/1997/05/27/founding-act
31. NATO, *Bucharest Summit Declaration*, 3 April 2008: https://nato.int/en/about-us/official-texts-and-resources/official-texts/2008/04/03/bucharest-summit-declaration
32. Organisation for the Prohibition of Chemical Weapons, *Chemical Weapons Convention*: https://www.opcw.org/chemical-weapons-convention