Power in International Relations

The Concept of Power in International Politics

Power is the organising concept of the discipline and its least settled one. Every school of theory must say what power is before it can say anything else, and no two agree on where it resides or how it is recognised. The decisive move is to stop treating power as a substance a state stores and start treating it as a relationship a state is in. Once that move is made, the failure of capability inventories to predict outcomes stops being a puzzle.

Power as a relationship, not a possession

• The classic definition is relational, and it comes into international relations from the study of domestic politics.

“A has power over B to the extent that he can get B to do something that B would not otherwise do.” — Robert Dahl

• Applied internationally, the formulation says that power exists only in a specific relationship, between specific actors, over a specific issue. A state does not “have” power in the abstract; it has power over particular others, in particular domains, at particular moments.
  • The counterfactual is doing the work: the test is not whether B acted, but whether B acted differently from how it would otherwise have acted.
  • This makes power hard to observe. Where B complies in anticipation, no visible act of influence occurs at all, and the most complete exercises of power leave the least evidence.
• Three properties follow directly from the relational definition, and they organise everything else in this note.
  • Relational. Power is a property of a relationship, so a state can be overwhelming in one relationship and helpless in another with no change in its capabilities.
  • Relative. Power is measured against a specific other, not against an absolute scale. A state grows more powerful by rising faster than its rival, or by its rival falling faster than itself.
  • Contextual. Capabilities are domain-specific. Aircraft carriers are decisive in sea-lane control and irrelevant to a currency crisis; a veto in the Security Council does nothing about a semiconductor shortage.

Power as a means and power as an end

• Realism treats power simultaneously as instrument and as objective, and the ambiguity is deliberate rather than sloppy.

“Whatever the ultimate aims of international politics, power is always the immediate aim.” — Hans Morgenthau

• The claim is that ends vary but the proximate requirement does not. A state may want prosperity, prestige, ideological expansion or simple survival, but it must first secure the capacity to pursue them against the resistance of others, and that capacity is power.
  • Morgenthau’s fuller position, that interest is defined in terms of power, belongs with realism’s assumptions and is developed in the companion article Realist Approach to International Relations.
  • The counter-position, which liberals and constructivists share, is that treating power as the immediate aim of all foreign policy makes the theory unfalsifiable: any behaviour can be redescribed as power-seeking after the fact.

The vocabulary of power: influence, force, coercion, authority

Power is a family of related phenomena, and the words used for them are not interchangeable. Distinguishing them matters because the mechanisms differ: what makes one work makes another fail. A state relying on authority loses it the moment it substitutes force.

Term	What it is	The mechanism	Illustration
Influence	Affecting behaviour without command, through persuasion, inducement or example	B chooses differently, but chooses	Development assistance shaping a recipient’s voting in the General Assembly
Coercion	Changing behaviour by threatening unacceptable cost	B still chooses, but under a manipulated set of options	Sanctions threatened but not imposed
Force	Physically imposing an outcome regardless of the target’s will	B’s choice is bypassed altogether	Occupation of territory
Authority	Compliance because the exercise is regarded as rightful	B accepts an obligation to comply	The Security Council’s Chapter VII decisions binding all members

• Force is the crudest and least efficient form, because it must be reapplied continuously and consumes resources at every application.
• Authority is the cheapest, because the compliance is voluntary and self-policing — which is why hegemonic powers work so hard to convert preponderance into legitimacy.
• The table is a rough ladder of cost, and states move down it only when the cheaper instruments have failed.

Capability is not outcome

• The oldest error in the study of power is inferring outcomes from inventories — assuming that the side with more tanks, more money and more people will get its way.
• Vietnam is the standing refutation. The United States held overwhelming superiority in every measurable index of material capability and still failed to secure its political objective, because the objective required breaking an adversary’s will rather than destroying its forces.
  • The asymmetry of stakes matters more than the asymmetry of capability: the weaker side was fighting for what it regarded as national existence, the stronger for a peripheral strategic interest.
• Afghanistan repeated the lesson twice, for two different superpowers. The Soviet Union withdrew in 1989 and the United States and its coalition in 2021, in each case after roughly a decade, and in each case the adversary was materially negligible by any conventional measure.
• The general point is that power is a probabilistic claim, not an arithmetic one. Capabilities alter the odds; they do not determine the result.
  • Intervening variables include the credibility of a threat, the resolve of the target, the terrain, alliance behaviour, domestic tolerance for cost and simple time horizons.

Inventories of capability describe what a state could in principle do; they say nothing about what it will actually be able to make others do.

Elements of National Power

Elements of national power are the underlying capabilities from which usable power is generated. They are conventionally divided into tangible elements, which can be counted, and intangible elements, which cannot — and the intangible ones repeatedly turn out to be the decisive variables. Each element below is a potential, not a power: it becomes power only when a state can convert it, and conversion is where most states fail.

Tangible elements

• Geography is the most durable element, because it changes least and constrains most.
  • Location determines who a state’s neighbours are and therefore what threats it must plan against. It is said of the United States that it is what it is not because of who Americans are but because of where they are — two oceans, two weak neighbours and no continental peer.
  • Size provides strategic depth, the capacity to trade space for time, which saved Russia against both Napoleon and Hitler.
  • Topography shapes defensibility: the Himalayas, the English Channel, the Alps and the Sahara have all functioned as strategic assets no budget could buy.
  • Access to the sea distinguishes maritime powers from landlocked ones, whose trade costs and strategic options are permanently constrained.
• Population and demography supply the manpower for armies, factories and taxation, but the relationship is not linear.
  • Size alone is not power: Nigeria and Bangladesh are among the world’s most populous states without corresponding influence.
  • Age structure matters more than headcount. A demographic dividend — a large working-age share relative to dependants — raises savings and output; an ageing structure reverses both, which is the central constraint on Japanese, European, Russian and increasingly Chinese power.
  • Human capital — literacy, health, skills — converts numbers into productivity, which is why smaller educated populations outperform larger uneducated ones.
• Natural resources provide both the inputs to production and, when concentrated, a direct instrument of leverage.
  • Energy is the classic case. Oil made Gulf monarchies strategically significant far beyond their size, and Russian gas gave Moscow a decade of political leverage over European capitals.
  • Critical minerals are the contemporary equivalent, and the concentration is extreme: China leads refining for nineteen of twenty strategically important minerals, at an average market share of about 70%, accounts for roughly 91% of rare-earth separation and refining, and manufactures around 94% of the world’s permanent magnets.
  • Food and water determine autonomy under blockade or sanction, and a state dependent on imported calories has a vulnerability no army can close.
• Economic and industrial capacity is the element from which most other elements are purchased.
  • A strong economy funds defence, diplomacy and development simultaneously, and it does so without the political strain that forces trade-offs among them. Germany’s economic weight, not its army, gives it decisive influence within the European Union.
  • The composition matters as much as the size. Industrial and technological depth allows a state to build what it needs; an economy of comparable size resting on commodity exports cannot.
  • Manufacturing capacity is a war-fighting asset. The war in Ukraine reduced the question of artillery ammunition to one of industrial base, and exposed how far the West’s had atrophied.
  • Market size is itself an instrument. A large domestic market lets a state set regulatory terms others must meet to sell into it.
• Military capability is the element realists treat as ultimate, because it is the only one that can settle a dispute by force.
  • Its components are manpower, equipment, doctrine, training, logistics, command-and-control and the ability to sustain operations at distance.
  • Global reach distinguishes great powers from regional ones: bases, carriers, strategic airlift and refuelling convert an army into an instrument usable beyond the neighbourhood.
  • World military expenditure reached $2,887 billion in 2025, 2.5% of world GDP; the five largest spenders were the United States at $954 billion, China at $336 billion, Russia at $190 billion, Germany at $114 billion and India at $92.1 billion. The top three alone accounted for 51% of the world total.
  • Spending is an input, not an output. Russia’s military burden of 7.5% of GDP buys quantity at a cost to every other element of its power.
• Technology increasingly determines what every other element is worth.
  • It multiplies military capability — precision, surveillance, stealth, autonomy — and it does so unevenly enough to make numerical comparisons of platforms nearly meaningless.
  • It has reduced but not eliminated the importance of geography, since missiles, satellites and cyber operations cross terrain that armies cannot.
  • Leadership in artificial intelligence, semiconductors, quantum computing and space is now treated by states as a first-order security interest, not an economic one.
• Intelligence capability is a distinct element, not a subdivision of the military, and it disproportionately rewards small states.
  • It provides warning, targeting, verification and the ability to act covertly where open action would be politically impossible.
  • The strength of states such as the United States, Russia, Israel and the United Kingdom rests substantially on highly professional intelligence services — Israel’s standing in West Asia is unintelligible without Mossad, and the American reach without the CIA and the National Security Agency.
  • Intelligence is also the element most prone to catastrophic failure, and its failures — the 1962 Sino-Indian war, Yom Kippur in 1973, Iraqi weapons of mass destruction in 2003 — are more consequential than most military defeats.

Intangible elements

• Political stability and the quality of government determine whether the tangible elements can be mobilised at all.
  • A state that cannot make and keep decisions cannot use what it owns. Effective government converts resources into policy; ineffective government leaves them latent.
  • Continuity of foreign policy across changes of administration is itself an asset, because it makes commitments credible to others. China’s consistency of direction over decades, and India’s stable governments across the last decade, both illustrate the point.
  • Extractive capacity — the ability to tax, conscript and regulate — is the underrated core of state strength. A state with a large economy and a weak revenue system is poorer, strategically, than the figures suggest.
• Leadership is discounted by structural realism, which holds that individuals are interchangeable under systemic pressure, and yet it repeatedly proves decisive.
  • The strong ruler converts weak elements of the state into elements of strength, which is precisely the claim of the oldest indigenous theory of the elements of state power.
  • Kautilya’s saptanga theory identifies seven limbs of the state — swami (the ruler), amatya (ministers and bureaucracy), janapada (territory and population), durga (the fortified capital), kosha (the treasury), danda (the army) and mitra (allies) — and treats the ruler as the nabhi, the navel or hub through which the other six are coordinated.
  • The saptanga framework anticipates the modern tangible-intangible division by more than two millennia, and it is more integrated than most modern lists, because it treats allies as an element of the state’s own strength rather than as an external variable.
  • Contemporary evidence for the leadership variable is abundant: the redirection of United States foreign policy after 2017, and again after 2025, changed no material capability and changed almost every outcome.
• National morale and social cohesion determine how much cost a society will bear before it insists on stopping.
  • A cohesive society can sustain losses that a divided one cannot, which is the mechanism behind most asymmetric outcomes, Vietnam included.
  • Social composition constrains foreign policy directly. A homogeneous society permits decisions that a heterogeneous one must negotiate; India’s inability for decades to normalise relations with Israel openly reflected domestic sentiment rather than any strategic assessment.
  • The point cuts both ways. Diversity is a soft-power asset and a democratic strength even where it complicates a particular policy, and cohesion bought by repression is brittle rather than deep.
• Ideology supplies both a motive for others to follow and a claim to legitimacy.
  • An ideology with external appeal recruits allies who are not being paid, which is why the Cold War was fought partly through competing universalisms.
  • It also constrains the state that holds it: a professed commitment is a standard against which others will measure the state’s conduct, and hypocrisy is a reputational cost.
• Diplomacy is the element that converts all the others into outcomes, and it is where states most often waste what they have.
  • Diplomacy is the quality of a state’s statecraft: the ability to read others’ intentions, build coalitions, time initiatives, frame issues and extract concessions without conceding equivalently.
  • A state with modest capability and excellent diplomacy consistently outperforms — Singapore and the Gulf states are the standing examples — while great powers routinely underperform their inventories through clumsy statecraft.
  • The instruments of diplomacy as a technique of statecraft are developed in the companion article National Interest.
• Reputation and credibility are the accumulated record on which others base their expectations.
  • A state that has kept commitments can obtain compliance by promise; one that has not must pay in advance. Credibility is the interest a state earns on its own past behaviour.

Limitations of National Power

A state’s power is bounded from several directions at once — by the internal difficulty of converting assets into outcomes, by the structure of the system it operates in, and by the character of the adversaries it faces. These limitations are not exceptions to the concept of power. They are what the relational definition predicts.

• The non-convertibility problem is the deepest limitation. Resources in one domain do not exchange freely into results in another.
  • There is no reliable exchange rate between economic weight, military capability and political outcome. Japan’s economic size has never purchased strategic influence proportionate to it; Russia’s military weight has never purchased economic security.
  • Conversion requires a mechanism — a doctrine, an institution, an alliance, a diplomatic channel — and states that have not built the mechanism hold assets they cannot spend.
• The paradox of unrealised power describes states whose measurable capability persistently exceeds their observed influence.
  • Latent power becomes actual power only through political will, institutional capacity and strategy. Germany and Japan hold immense resources and deliberately decline to convert them into military instruments.
  • India is the standing case of the gap in the other direction, discussed below.
• Geography’s weight has declined without vanishing. Missiles, aircraft, satellites, submarine cables and cyber operations all cross terrain, and the two oceans that once made the United States invulnerable no longer do.
  • But geography still dictates who a state must live beside. India cannot relocate away from a contested Himalayan frontier, and no technology has made the Strait of Malacca or the Strait of Hormuz less consequential.
• Interdependence constrains the use of power by making coercion expensive to the coercer.
  • Sanctions damage the sanctioning economy as well as the target, and the deeper the trade and investment relationship, the more symmetric the damage.
  • Supply chains create hostages in both directions: a state that controls a chokepoint also depends on the customers it might close it against.
• Nuclear weapons are the clearest case of power that cancels itself.
  • Their utility lies entirely in not using them. A weapon whose use would be catastrophic for the user cannot be employed for ordinary political ends, so it deters existential threats and is irrelevant to everything below that threshold.
  • Nuclear-armed states have repeatedly lost wars to non-nuclear opponents — the United States in Vietnam, the Soviet Union and then the United States in Afghanistan — because the decisive weapon was unusable.
  • Deterrence theory in full, including the stability-instability paradox that this generates, is developed in the companion article Deterrence.
• Domestic and social constraints limit what governments can spend, risk and sustain.
  • Democracies face a casualty constraint and an electoral clock that authoritarian states face less acutely, which shortens their tolerable time horizons in protracted conflicts.
  • Budgets are finite and contested: every unit of defence spending is contested by health, education and welfare claims, and the trade-off is sharper in poorer states.
  • Federal structures, coalition politics, courts and free media all add veto points between a decision and its execution.
• International law, institutions and norms raise the cost of naked power, even where they cannot prevent its exercise.
  • The prohibition on the use of force in Article 2(4) of the UN Charter does not stop wars, but it forces states to argue that their wars are exceptions — self-defence, authorisation, humanitarian necessity — and that argument is itself a constraint on how and when force is used.
  • Reputational costs are real and cumulative. A state seen to act lawlessly finds coalition-building harder, and its assurances discounted, for years afterwards.
• The asymmetry problem is the most contemporary limitation: conventional power is poorly matched against non-state adversaries.
  • Insurgents, terrorist networks and criminal organisations present no territory to seize, no capital to threaten and no army to defeat, so the instruments that decide interstate wars do not apply.
  • They operate from the territory of third states, often friendly ones, which makes retaliation a diplomatic problem rather than a military one.
  • The stronger side must win continuously; the weaker side wins by not losing, which reverses the usual relationship between capability and expected outcome.
  • The nature of the actors involved is developed in the companion article Transnational Actors.

Power fails less often because a state lacks capability than because it cannot convert the capability it has into the outcome it wants.

Measurement of Power

Why power resists measurement

• Elements supply potential; actual power depends on how effectively a state uses them, and effectiveness is not a quantity anyone can observe in advance.
• States systematically misrepresent their own strength. Defence budgets conceal expenditure, capabilities are exaggerated for deterrence and concealed for surprise, and the most important assets — intelligence penetration, cyber capability, warhead readiness — are secret by design.
• The relational definition makes a single number incoherent. If power exists only in a relationship over an issue, then any aggregate score is an average across relationships in which the state’s actual leverage varies enormously.
• Intangibles cannot be weighted. Any index must either omit morale, leadership and cohesion — losing the variables that explain most anomalies — or assign them numbers that are essentially arbitrary.

Composite indices and what they miss

Index	What it aggregates	Principal weakness
Composite Index of National Capability (CINC), from the Correlates of War project	Six indicators — total population, urban population, iron and steel production, primary energy consumption, military expenditure and military personnel — as a share of the world total	Purely material and industrial-era. It weights steel and manpower in an economy where semiconductors and software decide outcomes, and it contains no intangible element at all
Lowy Institute Asia Power Index	Eight measures across resources and influence, including economic and military capability, resilience, future resources, diplomatic and defence networks, economic relationships and cultural influence	Regional in scope, and the influence measures rest on judgement-based coding
Comprehensive National Power (CNP), developed in Chinese strategic scholarship	A weighted aggregate of material, institutional and normative components intended to give a single national score	The weights are chosen by the analyst, so the output largely reflects the assumptions of the index designer

• In the 2025 edition of the Asia Power Index the United States scored 80.5 and China 73.7, both classified as superpowers, with India third at 40.0, ahead of Japan at 38.8 and Russia at 32.1.
  • India’s strongest showings were in economic capability and future resources; its weakest was in defence networks, where it ranked eleventh.
  • The gap in the numbers is the analytically interesting part: the distance between the top two and the third is roughly the distance between the third and the twenty-seventh.
• Indices are useful for tracking direction and useless for predicting outcomes. They tell you whether a state’s material base is thickening or thinning; they cannot tell you who will prevail in a specific contest.

Classification of states by power

Because exact measurement is impossible, states are conventionally sorted into broad categories defined by what they can do rather than by what they hold. The categories are ordinal rather than cardinal, and the thresholds between them are matters of judgement.

Category	Defining capacity	Illustrations
Superpower	Global reach across every domain; cannot be defeated even by a combination of others	The United States; the Soviet Union until 1991
Great power	Able to defend its own vital interests unaided and to shape outcomes regionally and in some global domains	China; Russia; the permanent members of the Security Council
Middle power	Able to defend itself in coalition, and to exert influence through institutions, coalitions and niche capabilities rather than unilaterally	Japan; Brazil; Indonesia; Australia; Canada
Small power	Limited or no capacity to affect outcomes beyond its own borders; a rule-taker	Most of the membership of the United Nations

• Supplementary categories fill the gaps: regional powers dominate a neighbourhood without global reach, emerging powers are defined by trajectory rather than position, and pivotal states matter disproportionately because of location or resources.
• India’s placement is genuinely contested, and the contest is informative.
  • By material aggregates — population, the fifth-largest military budget, one of the five largest economies, an independent nuclear deterrent and space programme — India exceeds most definitions of a middle power.
  • By demonstrated influence — the inability to determine outcomes even in its own neighbourhood, and the absence of a permanent Security Council seat — it falls short of great-power status.
  • The Asia Power Index resolves this by placing India in a “major power” band above middle powers and below the two superpowers, which is a useful description of an intermediate position that the older four-fold classification cannot accommodate.
  • India’s own self-description has moved from “leading power” to a claim of being a distinct pole in a multipolar order, which is a claim about trajectory as much as about present capability.

Polarity: the distribution of power

• Polarity denotes the number of centres around which the international distribution of power is organised, and the concept comes from structural realism, principally Kenneth Waltz and John Mearsheimer.
  • Unipolarity: one state possesses preponderant capability, with no other able to balance it — the international system after 1991.
  • Bipolarity: two states hold capabilities far above all others, and lesser states array around them — the Cold War system.
  • Multipolarity: three or more comparable centres, with shifting alignments among them — Europe before 1914, and the configuration many observers now describe.
• Whether any of these arrangements is more stable than the others is a separate and much larger argument, and it is developed in the companion article Balance of Power, which also covers balancing, alliances and the mechanisms by which distributions are maintained.

Types of Power

Hard power

• Hard power is the use of coercive means — military force and economic pressure — to change another state’s behaviour against its preference.
  • Its instruments are military intervention, forward deployment, blockade, sanctions, tariffs, export controls and the withdrawal of aid.
  • It rests on tangible resources and works through the mechanism of cost imposition: the target complies because non-compliance has been made more expensive than compliance.
• Realists give hard power priority on the argument that in an anarchic system without enforcement, only the capacity to hurt is finally decisive.

Soft power

• Soft power is the ability to obtain preferred outcomes through attraction rather than coercion or payment — getting others to want what you want. The term is Joseph Nye’s.
  • Its resources are culture, political values and foreign policy, and they work only where others find them legitimate or admirable.
  • Its distinguishing feature is that it is largely not owned by governments. Universities, film industries, diasporas, civil-society organisations and private firms generate it, and states can encourage but not command it.
• Nye’s account of the sources of soft power and its mechanisms is developed at length in the companion article Introduction to International Relations. What matters here is soft power’s place in the working set of types.

	Hard power	Soft power
Basis	Materialistic and resource-intensive — armies, money, sanctions	Non-materialistic — culture, values, example
Visibility	Concrete and observable; deployments and tariffs are countable	Invisible; its operation is diffuse and its effect indirect
Mechanism	Compelling — changes behaviour against preference	Relies on consent — changes preference itself
Ownership	At the disposal of the state, which commands its instruments	Dispersed among the institutions of civil society, which the state does not control
Availability	The state can use it at will, when and where it chooses	Being dispersed, the state may not be able to deploy it when required
Time to build	Can be acquired in a short time with money and decision	Takes a long time to build and accumulates only through consistency
Failure mode	Provokes balancing, resentment and counter-coercion	Evaporates when conduct contradicts the professed values

Smart power

• Nye’s later position was that soft power alone does not work, and that the two forms have to be combined by design rather than chosen between.
• Smart power is the strategy of combining hard and soft resources according to context — Nye’s formulation underscores the necessity of strong military power alongside heavy investment in institutions, partnerships and alliances. The term was popularised in American policy by Hillary Clinton.
  • The diagnosis behind it was that excessive reliance on hard power was itself a cause of the erosion of American hegemony — that coercion consumed the legitimacy on which cheap compliance depended.

Sharp power

• Sharp power names the influence operations of authoritarian states that use the openness of democracies against them. The term comes from work published by the National Endowment for Democracy, principally by Christopher Walker and Jessica Ludwig.
  • Its instruments are disinformation, covert funding of political parties and media, pressure on universities and publishers, control of diaspora organisations, and the manipulation of information environments.
  • It is distinguished from soft power by intent rather than by instrument: soft power seeks to attract, sharp power seeks to distract, divide and delegitimise.
  • Nye’s own concern is that sharp power is not a subtype of soft power at all but a form of covert coercion wearing soft power’s clothes, and that democracies weaken themselves if they respond by closing the openness that is their own attraction.
• The Chinese and Russian information campaigns of the last decade are the standard illustrations, and the response — foreign-agent registration, platform regulation, investment screening — carries an obvious cost to the openness being defended.

Structural power

• Susan Strange argued that the power to shape the frameworks within which others operate matters more than the power to make a specific other do a specific thing.
  • Relational power decides who wins a bargain; structural power decides the terms on which the bargaining happens, who is at the table and what is on the agenda.
• Strange identified four interlocking structures, and a state’s structural power is its combined position across them.
  • The security structure — who provides protection to whom, and on what terms. The provider acquires influence over the protected far beyond the security domain.
  • The production structure — what is produced, where, by whom and by what arrangements. Control over the organisation of global production shapes every participant’s development options.
  • The financial structure — who creates credit and on what conditions. This is the structure Strange thought most underrated, since the power to determine access to capital determines what any other actor can attempt.
  • The knowledge structure — what is believed, what counts as expertise, who is credentialled and who controls the channels through which knowledge travels.
• The four structures reinforce one another, which is why structural power is cumulative and hard to dislodge: a state that sets the security terms also shapes production, which shapes finance, which funds the knowledge institutions.
• The contemporary application is direct. Dollar clearing, the standards bodies of the internet, the rating agencies, the reserve-currency system and the architecture of the cloud are structural power in Strange’s precise sense — none of them coerces anyone, and all of them determine what everyone else can do.

Fast power

• A further contemporary claim is that speed has itself become a determinant of power.
  • The claim is that what matters is not heft but agility — the ability to decide, adapt and deploy faster than events and faster than rivals. The rule of the present system, on this account, is not the survival of the fittest but the survival of the fastest.
  • The evidence is the compression of decision cycles by real-time information, the tempo of cyber operations and drone warfare, and the ability of small, well-organised states to shape a crisis before larger ones have convened a meeting.
  • It reframes bureaucratic capacity as a power resource. A state whose decision-making is slow converts its assets late, and late conversion is often equivalent to no conversion.

The Barnett and Duvall taxonomy

Michael Barnett and Raymond Duvall offer the most systematic classification, generated by crossing two questions: whether power works through interaction between already-constituted actors or through the constitution of what those actors are; and whether it works directly and specifically or indirectly and diffusely.

Type	How it works	Illustration in international politics
Compulsory power	Direct control by one actor over another’s circumstances or actions, through the deployment of resources despite resistance	Military coercion; sanctions; conditional lending
Institutional power	Indirect control through the rules and procedures of institutions, affecting actors who are socially and spatially distant	Weighted voting at the IMF and World Bank; the Security Council veto; WTO dispute rules
Structural power	Direct constitution of actors’ capacities and interests through their positions in a structure, so each position defines the other	The core-periphery relation; creditor and debtor; patron and client
Productive power	Diffuse constitution of subjects and meanings through systems of knowledge and discourse	Categories such as “developed” and “developing”, “failed state”, “rogue state”, “civilised nation”

• The taxonomy stops the argument about which conception of power is correct. Compulsory power is the realist conception, institutional power the liberal one, and productive power the constructivist and post-structural one; all four are operating simultaneously in any real relationship.
• It also shows why measurement is hopeless. Only compulsory power is countable; the other three exist in rules, positions and meanings that no index registers.

Faces and Dimensions of Power in International Politics

The debate over the “faces” of power began in the study of domestic community politics and transfers cleanly to international relations, because the same question arises: is power only what you can see happening?

• The first dimension — decision-making power. Power is exercised when A prevails over B in an observable conflict over a decision.
  • Internationally, this is the visible bargaining of summits, treaty negotiations and votes, and it is what realist analysis measures.
  • Its limitation is that it registers only conflicts that reach the point of open decision.
• The second dimension — agenda-setting power, drawn from Peter Bachrach and Morton Baratz, who called it the “mobilisation of bias”: power is also the ability to prevent issues from reaching the agenda at all.
  • The Bretton Woods institutions are the clearest international case. The IMF and the World Bank operate on weighted voting tied to quota, and the effective American veto over major decisions, together with the convention that a European heads the Fund and an American the Bank, decides what is discussable before any discussion begins.
  • The Security Council’s permanent membership works the same way: the veto’s most important use is the draft never tabled because a veto is certain.
  • Non-decision is as much an exercise of power as decision, and it leaves no record, which is why the first dimension systematically underestimates the power of the strong.
• The third dimension — preference-shaping power, from Steven Lukes: the most effective power secures compliance by shaping what others want, so that no conflict arises to be observed.
  • Internationally this operates through global norms, development models, professional training and the standard vocabulary of policy. A state that adopts a structural adjustment programme because its officials were trained to regard it as sound economics is not being coerced.
• The constructivist extension is that power includes the power to constitute what counts as normal.
  • Categories are consequential. Which states are “responsible” nuclear powers and which are proliferators, which conflicts are “civil wars” and which are “terrorism”, what counts as “development” and who is authorised to certify it — these classifications distribute rights and obligations before any negotiation opens.
  • Alexander Wendt’s point that anarchy is what states make of it is the general form of the claim: the material structure does not determine its own meaning, and whoever fixes the meaning has exercised power.

Dimension	Power consists in	Where the conflict is	International example
First	Winning open contests	Observable and overt	Treaty negotiations; UNGA votes
Second	Keeping issues off the agenda	Suppressed, latent but real	Quota-weighted voting; the anticipated veto
Third	Shaping what others want	Invisible; no conflict occurs	Norm diffusion; policy orthodoxy
Constitutive	Defining what counts as normal	Prior to conflict altogether	Categories of statehood, development, legitimacy

The most complete exercises of power produce no visible event, which is exactly why the states that hold most of it prefer to be measured by visible events.

Power as Zero-Sum or Variable-Sum

Whether power is a fixed quantity that states divide among themselves or a variable quantity that can grow for everyone at once is the single question that separates realist from liberal international relations, and it is not an abstract dispute — it determines whether cooperation is expected to be durable or merely tactical.

The zero-sum position

• In a zero-sum conception, one state’s gain is by definition another’s loss, because power is relative and the total is fixed.
  • The realist argument rests on relative gains. Under anarchy, with no authority to enforce agreements, a state must ask not only whether it benefits from a bargain but whether it benefits less than its partner — because a partner who gains more has gained the capacity to coerce it later.
  • Joseph Grieco’s formulation is the sharpest: states are defensive positionalists. They are concerned with achieving and maintaining relative position, so even a mutually profitable agreement can be rejected if the distribution of the profit shifts the balance adversely.
  • This is why gains from cooperation do not automatically produce more cooperation. The fear is not of cheating alone, which institutions can address, but of the honest and asymmetric distribution of the benefit, which they cannot.

The variable-sum position

• In a variable-sum conception, power and welfare can expand for several states simultaneously, so cooperation is not a temporary truce but a rational standing strategy.
  • The liberal reply turns on absolute gains. States care principally whether they are better off than before, not whether they are better off than a rival, and on that criterion trade, investment and technology transfer are positive-sum.
  • Robert Keohane and Joseph Nye add the institutional mechanism: institutions reduce transaction costs, supply information about compliance, extend the shadow of the future and make defection costly, which flattens the relative-gains anxiety without abolishing it.
  • Some goods are inherently non-rival. Pandemic surveillance, climate mitigation, freedom of navigation and financial stability benefit every state including those that contributed nothing, which is what makes them public goods and also what makes them under-supplied.
  • Empirically, the deepest interstate relationships are variable-sum: Franco-German reconciliation converted a relationship that had produced three wars in seventy years into one in which relative gains ceased to be a security question at all.

	Zero-sum conception	Variable-sum conception
Core assumption	Power is a fixed quantity, relative by nature	Power and welfare are expandable
Which gains matter	Relative gains — how much more did the other side get?	Absolute gains — am I better off than before?
View of cooperation	Tactical, contingent, and reversed when the distribution turns adverse	Rational, self-reinforcing and institutionally stabilised
Where it holds best	Security, territory, status, veto positions	Trade, finance, technology, public goods
Theoretical home	Realism; Grieco’s defensive positionalism	Liberalism; neoliberal institutionalism
Policy implication	Balance, hedge, retain autonomy	Institutionalise, integrate, expand issue linkage

Why the same relationship contains both

• The two conceptions describe different domains of a single relationship rather than rival descriptions of the whole of it. States are simultaneously in zero-sum contests over position and variable-sum arrangements over welfare, with the same partner.
• A purely zero-sum reading cannot explain the present system. If power were strictly zero-sum, adversaries would not sustain trade at all, would not sit in the same institutions, and would not co-author agreements — and they do all three continuously.
• A purely variable-sum reading cannot explain it either, since the cooperation is repeatedly subordinated to positional concerns the moment a technology or a chokepoint becomes strategically significant.

The contemporary illustration

• United States–China relations are the working case, and they are conducted in both registers at once.
  • The zero-sum register: American export controls on advanced semiconductors and the equipment to make them; Chinese export controls on rare earths, tightened through 2025 to cover components and battery supply chains; investment screening; and a naval balance in the western Pacific in which each side’s gain is read as the other’s loss.
  • The variable-sum register: bilateral trade that neither economy can replace at tolerable cost; joint dependence on financial stability; and continued engagement on climate, where neither state can achieve its own objective without the other’s emissions falling.
  • The pandemic showed both registers simultaneously — scientific data shared across the rivalry while vaccines and equipment were used as instruments of positional advantage.
• The analytical conclusion is that the zero-sum frame alone cannot account for the observed pattern. It explains the competition and predicts none of the cooperation; the variable-sum frame explains the cooperation and cannot explain why it stops precisely where strategic technology begins.
• What explains the mixture is issue structure. Where a good is positional — status, territory, a chokepoint, a veto — the logic is zero-sum; where it is non-positional — output, health, climate stability — it is variable-sum.
  • States sort their behaviour by issue, so the same two governments look like adversaries on one file and partners on another without either posture being insincere.

Conflict and cooperation are not phases of a relationship that alternate; they are simultaneous properties of different issues within it.

Hegemony and Hegemonic Stability Theory

The theory and its authors

• Hegemonic stability theory holds that an open, rule-governed international order requires a single preponderant state to create and maintain it, and becomes unstable when no such state exists or is willing to act.
  • Charles Kindleberger originated the argument in explaining the depth and duration of the Great Depression: the interwar collapse happened because Britain could no longer stabilise the world economy and the United States would not.
  • Robert Gilpin developed it into a general theory of international political change, connecting the rise and decline of hegemonic orders to differential rates of economic growth.
• The underlying assumption is that peace and openness require a “global policeman” — an actor able to establish norms, supply public goods and punish those who defect from the order.
  • The public goods in question are an open trading system, a stable reserve currency and lender of last resort, security guarantees, and freedom of the seas.
  • The order it produces is a liberal international order, which is why hegemonic stability theory is the unusual case of a realist theory that endorses a liberal outcome.

The two qualities of a hegemon

• A hegemon requires both capacity and willingness, and the theory’s explanatory force comes from separating them.
  • Capacity is the material ability to supply the public goods and absorb the cost of enforcement.
  • Willingness is the political decision to bear that cost rather than free-ride on an order others maintain.
• The interwar United States is the case of capacity without willingness. It emerged from the First World War as the largest economy and principal creditor and declined the role — refusing to join the League, raising tariffs, withdrawing from currency stabilisation.
  • The result, on Kindleberger’s reading, was not a different order but no order at all.
• The converse case is willingness without capacity, which describes Britain after 1918: still committed to the system it had built, no longer able to underwrite it.

The bargain between hegemon and followers

• Why lesser states accept a hegemon: it supplies order, security and predictability that they would otherwise have to purchase individually and could not afford.
  • Followers also acquire voice — a hegemon that wants cheap compliance must offer consultation, and institutions are the currency of that offer.
• Why a hegemon accepts the role: because it writes the rules, and rules written by the strongest party favour the strongest party.
  • The costs of provision are real but are outweighed by the returns from an order structured around the hegemon’s currency, firms, language, standards and strategic priorities.
  • Hegemony in this sense denotes leadership rather than domination — an order others participate in because it works for them, not merely one they are compelled into.

The life-cycle: overstretch, free-riding and decline

• A hegemonic order is conventionally given a span of roughly a hundred years, after which the hegemon’s relative position erodes.
• Overstretch is the first mechanism: the commitments accumulate faster than the economic base that funds them, so an increasing share of national resource is spent on maintaining position rather than on growth.
• Free-riding is the second: because the public goods are non-excludable, other states consume the security and openness without contributing, and grow faster than the provider precisely because they are not paying.
• Differential growth is the third: the challengers industrialise from a lower base and therefore grow faster, so the gap closes even if the hegemon makes no mistakes.
• Megalomania is the fourth, and it is self-inflicted. A state without a peer competitor comes to believe there are no limits on its power and acts unilaterally against the institutions it built.
  • The invasion of Iraq in 2003 without Security Council authorisation, and the paralysis of the WTO’s Appellate Body, unstaffed since December 2019, are the standard examples of a hegemon undermining its own order.

Pax Britannica and Pax Americana

• The two historical hegemonies are Britain from roughly 1815 to 1914 and the United States from 1945, and the label attached to each carries an argument.
• Pax means a dictated peace, not an agreed one — the peace of the dominant, on the dominant’s terms — and the Roman ancestry of the phrase is exact.
• Their instruments differed. Britain’s rested on naval supremacy, the gold standard, free trade and formal empire; America’s on nuclear preponderance, the dollar, alliance networks and multilateral institutions rather than territorial possession.

Realist and Gramscian conceptions of hegemony

	Realist hegemony	Gramscian hegemony
What hegemony rests on	Military and economic preponderance	Ideology, culture and consent
How compliance is obtained	Deterrence, inducement, enforcement	Followers internalise the order as natural and legitimate
Who the actors are	States, principally the leading state	Transnational social forces and classes, with the state as one vehicle
What produces decline	Loss of material preponderance	Loss of ideological credibility and the emergence of a counter-hegemonic bloc
Attitude to the order	Explanatory: hegemony stabilises	Critical: hegemony conceals domination as common sense

• Antonio Gramsci’s original insight was that durable rule combines coercion with consent, and that consent is manufactured through culture, education and the institutions of civil society. The ruled comply because they accept the order’s account of itself.
• Robert Cox transposed this to the international level: world hegemony is a fit among material capabilities, ideas and institutions, and it is exercised through transnational social forces rather than by a state alone.

“Theory is always for someone and for some purpose.” — Robert Cox

• Cox’s distinction between problem-solving theory, which takes the existing order as given and makes it function better, and critical theory, which asks how the order came about and whether it might be otherwise, is itself an application of the argument: the theories a hegemonic order generates about itself are part of how it rules.
• The practical difference between the two conceptions is a prediction. The realist expects hegemony to end when material preponderance ends; the Gramscian expects it to survive material decline as long as the ideological framework holds — and to be vulnerable to ideological challenge even while the material base is intact.

Criticisms of hegemonic stability theory

• The historical fit is poor. The nineteenth-century order was maintained by the Concert of Europe and a balance among several powers as much as by British preponderance, and the freest trading period of the century came after Britain’s relative economic lead had begun to narrow.
• Order has outlived hegemony. Robert Keohane’s central argument is that institutions created under hegemony persist after it, because states continue to value the information, transparency and reduced transaction costs they supply — cooperation after hegemony is possible.
• The theory is ideologically loaded. It restates the interest of the dominant state as a general interest, which is exactly the move E. H. Carr identified in the doctrine of the harmony of interests.
• Hegemons are also destabilising. They fight peripheral wars, sponsor coups, and act outside the rules they enforce on others, so the record includes disorder caused by the stabiliser.
• It specifies no threshold distinguishing a hegemon from a very strong great power, which makes it difficult to falsify.
• It underweights the challenger’s agency, treating the order as the hegemon’s product when rising states shape it too — which is the gap power transition theory fills.

Power transition theory and the Thucydides Trap

• A. F. K. Organski, later with Jacek Kugler, argued that the international system is hierarchical rather than anarchic, with a dominant power at the top and a rule-set that reflects its preferences.
  • War becomes most likely not when the dominant power is strongest but when a dissatisfied challenger approaches parity with it. The two variables are relative power and satisfaction with the existing order.
  • A rising but satisfied state does not fight — the American overtaking of Britain transferred leadership without war, because the challenger accepted the order’s terms.
  • A rising and dissatisfied state does — Germany before 1914 is the canonical case, a challenger whose growth outran the status and access the order allowed it.
  • The theory inverts the balance-of-power expectation: parity is dangerous, and preponderance is stabilising.
• The Thucydides Trap is Graham Allison’s coinage, not Thucydides’s. Thucydides wrote that what made war inevitable was the growth of Athenian power and the fear this caused in Sparta; the phrase, the framework and the case set are Allison’s.
  • Allison’s study examines sixteen historical cases in which a rising power challenged a ruling one, and finds that twelve ended in war.
  • The four exceptions are the analytically important cases, since they show the outcome is not determined, and that avoidance required deliberate accommodation on both sides.
  • The framework’s weaknesses are that case selection and coding drive the ratio, that “war” covers events of wildly different character, and that nuclear weapons make the twentieth-century and twenty-first-century cases structurally different from the earlier ones.
• The contemporary application is the United States and China, and the value of the frame is diagnostic rather than predictive: it identifies the combination — rapid relative growth plus dissatisfaction plus mutual fear — that historically precedes great-power war, and thereby identifies what has to be managed.

The Contemporary Anatomy of Power

Technological and informational power

• The decisive capabilities are shifting from mass to computation. Artificial intelligence, semiconductors, cyber capability and access to space now determine the value of every conventional element.
  • Compute is being treated as a strategic resource in the way oil once was, with states restricting who may buy the most advanced processors and where the data centres may be built.
  • Space is a contested military domain rather than a support function: navigation, communications and surveillance satellites are single points of failure for modern forces.
  • Cyber capability narrows the gap for weaker states, because entry costs are low, attribution is slow and effects are deniable.

Weaponised interdependence

• Henry Farrell and Abraham Newman argue that the networks of globalisation are not flat: they have hubs, and the states in which those hubs sit can convert connection into coercion.
• Two effects follow. The panopticon effect is the ability to see what flows through a hub, giving the host state intelligence on transactions it is not party to. The chokepoint effect is the ability to deny access to the hub altogether.
• The dollar clearing system and SWIFT messaging are the paradigm hubs. The dollar is 57.1% of allocated reserves and about 48% of SWIFT messaging, so exclusion from dollar clearing is exclusion from most of world trade.
  • The 2022 sanctions on Russia were the largest demonstration to date — asset freezes on a central bank’s reserves, exclusion of major banks from SWIFT, and export controls with extraterritorial reach.
  • Export controls are the same logic applied to technology. United States and allied firms hold roughly 90% of global semiconductor manufacturing equipment and about 92% of supply-chain value, which converts a commercial position into a denial capability.
• Weaponisation is self-limiting, which is the analytically important point. Each use gives targets a reason to build alternatives — non-dollar settlement, indigenous fabrication, parallel messaging — which erode the hub over time, so structural power is spent as much as exercised.

Energy, minerals and data as leverage

• Resource leverage has migrated from fuels to processing. The scarcity is not in the ground but in refining, and refining is where concentration is extreme.
  • China’s October 2025 export controls, effective from November, extended beyond raw rare earths to components containing Chinese-origin material, to processing equipment and to lithium-ion battery supply chains.
  • Mineral leverage decays faster than technological leverage, because price spikes trigger substitution and state-backed diversification.
• Data and platform power is a genuinely new category, because the entities holding it are firms rather than states — search, cloud, social platforms and payment rails decide what information circulates and on what terms.
  • The regulatory response is itself a form of power: a large market that sets rules for data, competition and content exports those rules to everyone who wishes to sell into it.

Diffusion as well as redistribution

• Two distinct shifts are occurring simultaneously, and conflating them causes most of the confusion in the decline debate.
  • Power transition is the redistribution of power among states, principally from the Atlantic to Asia.
  • Power diffusion is the movement of power away from states altogether — to corporations, networks, criminal organisations, non-governmental organisations and individuals with a connection.
• Nye’s argument is that the second is the more consequential. The strongest state in a world of diffused power still cannot achieve its objectives alone, because the actors that matter for pandemics, finance, climate and information cannot be deterred, invaded or outspent.
• The result is not the end of state power but a change in what it is for. The capacity to convene, to set standards, to regulate and to build coalitions is displacing the capacity to compel as the currency of influence.

India’s Power Profile

• India is the clearest contemporary case of the gap between capability and influence, which makes it the best test of every argument above.
• The material base is substantial and improving.
  • The fifth-largest military expenditure in the world at $92.1 billion in 2025, up 8.9%, with an independent nuclear deterrent, a completing triad and a domestic space programme.
  • Among the five largest economies by nominal output and the third largest by purchasing power parity, with the demographic advantage that the ageing of China, Japan and Europe removes from them.
  • Third in the 2025 Asia Power Index at 40.0, above Japan and Russia, with its strongest scores in economic capability and future resources.
• The conversion problem is equally clear.
  • Defence networks are India’s weakest measured dimension, eleventh in the region, reflecting a deliberate refusal of alliance commitments.
  • Influence over outcomes in India’s own neighbourhood remains limited, and the claim to a permanent Security Council seat through the G4 has produced no institutional movement.
• Strategic autonomy and multi-alignment are the strategy for converting an intermediate position into leverage.
  • Rather than choosing a bloc, India holds simultaneous memberships in the Quad, BRICS, the Shanghai Cooperation Organisation and the I2U2 grouping, and trades and buys weapons across the rivalry.
  • The value of the position is that several parties must court India, and it is a structural-power play rather than a relational one: the aim is to sit at the intersection of frameworks rather than to prevail in any single bargain.
  • Its cost is that a state committed to no one is trusted fully by no one, which is visible in the friction of the 2025 tariff dispute with the United States and the subsequent reset.
• Soft power resources are unusually rich and unusually hard to convert.
  • The instruments are real: a diaspora of over thirty million, a film industry with genuine transnational reach, yoga and the International Day of Yoga, Buddhist heritage diplomacy across East and Southeast Asia, English-language higher education, and vaccine and disaster-relief assistance across the Indian Ocean.
  • The limits are equally real. India ranked 32nd, with a score of 48.0, in the 2026 Global Soft Power Index — well below its material standing — because soft power is generated by civil society and constrained by the perceived gap between professed values and domestic practice.
  • Soft power also cannot substitute for the hard kind. It has not altered Chinese behaviour on the frontier and did not shape the outcome of any crisis with Pakistan.
• Kautilya’s saptanga remains the most useful indigenous framework for reading this profile, because it treats the treasury, the army, the territory, the bureaucracy, the ruler and the allies as a single interdependent system in which weakness in one limb disables the others — which is precisely the conversion problem restated.