Edgeworth Paradox¶

Show that Bertrand's price-equals-marginal-cost result collapses once firms face capacity constraints below total demand: no pure-strategy equilibrium exists and prices cycle endlessly between the competitive floor and monopoly ceiling.

Core Idea¶

The Edgeworth paradox is the finding that Bertrand's price-competition result — two firms with identical constant unit costs will undercut each other to marginal cost — fails to produce a stable equilibrium once each firm faces a binding capacity constraint below total market demand. Edgeworth (1925) showed that under this configuration, no pure-strategy Nash equilibrium in prices exists: at any proposed price above marginal cost, a firm that is capacity-constrained cannot serve the entire market, which gives its rival an incentive to raise price and skim the residual demand not captured by the capacity-constrained firm; but at that higher price, the rival in turn faces defection; and so on indefinitely. The consequence is that prices cycle continuously between the competitive floor and the monopoly ceiling rather than settling at any stationary level, even in markets with stable demand, stable costs, and perfectly rational firms.

The paradox reveals that the Bertrand prediction was an artefact of the assumption of unlimited supply capacity: with infinite capacity, any firm priced above marginal cost loses all its customers to the undercutting rival, so the only stable outcome is price equal to marginal cost. Once capacity is finite and binding, each firm retains residual demand at elevated prices because its rival cannot serve the whole market, discontinuing the best-response function that the Bertrand logic requires. The structural consequence is that pure-strategy equilibrium existence in price competition depends critically on the relationship between each firm's capacity and total demand — a finding that motivates mixed-strategy equilibria as the only consistent solution in the Edgeworth case, and that Kreps and Scheinkman later reconciled with Cournot by showing that capacity-choice followed by price-competition recovers the Cournot quantity outcome.

Structural Signature¶

Sig role-phrases:

the price-setting sellers — two (or more) firms with identical constant unit cost competing on a continuous price action
the shared demand curve — a single market demand the sellers split, against which residual demand is reckoned
the binding capacity constraint — each firm's output ceiling sits below total market demand, so no firm can serve the whole market alone
the residual-demand skim — at any price above marginal cost, a firm's rival cannot capture everyone, leaving residual demand a firm can exploit by raising price
the discontinuous best-response — the Bertrand best-response function breaks at the point where capacity binds, the structural hinge of the result
the non-existence of pure-strategy equilibrium — no stationary price survives, since every proposed price invites defection upward then redefection
the perpetual price cycle — prices oscillate endlessly between the competitive floor and the monopoly ceiling as the only consistent (mixed-strategy) outcome
the capacity-to-demand existence boundary — the ratio of each firm's capacity to total demand decides which regime obtains: clean marginal-cost equilibrium versus cycling
the artefact exposure — the clean Bertrand result is revealed to have rested on the unlimited-capacity assumption

What It Is Not¶

Not a refutation of Bertrand competition. It does not show the undercutting logic is wrong; it shows the clean price-equals-marginal-cost result rested on an unstated assumption — unlimited capacity. Restore a binding capacity ceiling below total demand and the Bertrand best-response function discontinues. The paradox locates the load-bearing assumption rather than overturning the analysis built on it.
Not evidence of collusion or irrationality. Endless price cycling in a capacity-constrained oligopoly is the predicted equilibrium response to finite capacity, not a coordination signal or a behavioural defect. Reading weekly price cycles on a route as misbehaviour inverts the lesson: under binding capacity, cycling is evidence for competition, which is exactly why it shifts the antitrust evidentiary burden.
Not a result that merely raises the equilibrium price. Adding the capacity constraint does not shift a stationary equilibrium upward; it abolishes the existence of any pure-strategy equilibrium. The distinction that does the work is between equilibrium existence, uniqueness, and stability — the paradox is a failure of existence, not a change in level.
Not a claim that no equilibrium of any kind exists. A mixed-strategy equilibrium does exist and is the consistent solution; prices oscillate between the competitive floor and the monopoly ceiling as its realisation. The non-existence is specifically of a pure-strategy equilibrium — saying "there is no equilibrium" full stop overstates the result.
Not the general discontinuous-game existence problem itself. The bare shape — a binding constraint added to a continuous-action game destroys pure-strategy existence and forces mixing — recurs in matching with indivisibilities and war-of-attrition collapses. But that travels as the parent existence apparatus (the territory of the Nash, Glicksberg, and Reny theorems), not as the Edgeworth paradox, whose specific cargo is Bertrand undercutting, capacity, residual demand, and price cycling.

Scope of Application¶

The Edgeworth paradox lives within a single home discipline — oligopoly price-competition theory — restaged across its applications; its reach is bounded to settings with price-setting sellers of finite capacity facing a shared demand curve, and the deeper "a binding constraint can destroy pure-strategy equilibrium existence" shape travels to other discontinuous games only under the parent existence apparatus (the Nash, Glicksberg, and Reny theorems), not as this named paradox.

Industrial-organization theory — the home turf: the canonical example motivating mixed-strategy equilibria in price competition and the Kreps-Scheinkman two-stage capacity-then-price model that recovers the Cournot quantity.
Pricing analytics — supplies the structural reason for observed weekly price cycles among capacity-constrained sellers: airlines on a route, gasoline stations, hotel rooms in a city.
Antitrust analysis — informs whether oscillating prices in a capacity-bounded oligopoly are read as the competitive mixed-strategy equilibrium rather than as collusion, shifting the evidentiary burden.
Auction-theoretic models — where related best-response discontinuities arise as capacity is exhausted mid-allocation.

Clarity¶

Naming the Edgeworth paradox isolates which assumption was carrying the clean Bertrand result, and thereby separates three properties of equilibrium that students of price competition routinely run together: existence, uniqueness, and stability. The Bertrand reader takes price-equals-marginal-cost as the natural outcome of undercutting; the paradox shows that the conclusion depended entirely on unlimited supply, and that restoring a perfectly realistic capacity ceiling below total demand does not merely shift the equilibrium price but abolishes the existence of any pure-strategy equilibrium at all. With infinite capacity, a firm priced above cost loses every customer, so marginal-cost pricing is forced; with binding capacity, each firm keeps residual demand at an elevated price because its rival cannot serve the whole market, the Bertrand best-response function discontinues, and prices cycle endlessly between the competitive floor and the monopoly ceiling. The concept makes that dependence explicit rather than letting "competition drives price to cost" stand as a substrate-free truth.

The sharper, practically consequential question this opens is whether oscillating prices in a capacity-constrained oligopoly are evidence of misbehaviour or simply the equilibrium itself. Without the paradox, weekly price cycles in airline routes, gasoline retail, or hotel rooms invite a collusion reading; with it, the antitrust analyst recognises the cycling as the predicted mixed-strategy response to finite capacity and adjusts the evidentiary burden accordingly. The concept also reframes equilibrium existence in price competition as something that depends on the ratio of each firm's capacity to total demand — a parameter to be checked rather than assumed — which is precisely the question Kreps and Scheinkman exploited in showing that capacity choice followed by price competition recovers the Cournot quantity.

Manages Complexity¶

Oligopoly theory faces a recurring question across industries — airline routes, gasoline retail, hotel rooms, any pair of capacity-bounded sellers — of whether price competition settles or wanders, and whether observed price cycling signals collusion or competition. The Edgeworth paradox compresses this by locating the whole answer in a single comparison: the ratio of each firm's capacity to total market demand. The analyst no longer re-solves the pricing game industry by industry; they check that one ratio and read off the qualitative regime. The branch structure is binary. Where capacity comfortably exceeds demand, the Bertrand best-response function stays intact and price is driven to marginal cost — a clean stationary pure-strategy equilibrium. Where capacity binds below total demand, each firm retains residual demand at elevated prices because its rival cannot serve the whole market, the best-response function discontinues, no pure-strategy equilibrium exists, and prices oscillate endlessly between the competitive floor and the monopoly ceiling as the mixed-strategy outcome. This collapses a high-dimensional modelling problem — the full specification of demand, costs, and conjectures — to one parameter governing equilibrium existence, and it does double duty by pruning the large class of otherwise-reasonable price-competition models that assume capacity away: any such model is exposed as resting on an artefact the moment the capacity ratio is restored. The compression further sharpens the analyst's reading of data, since price cycling in a binding-capacity oligopoly is reclassified from anomaly-to-be-explained (collusion, irrationality) to the predicted equilibrium itself, shifting the antitrust evidentiary burden — and it is exactly this capacity-to-demand parameter that Kreps and Scheinkman exploited to recover the Cournot quantity from capacity-then-price competition, so the same compressed handle organises the reconciliation with Cournot rather than a separate derivation.

Abstract Reasoning¶

The Edgeworth paradox licenses a set of oligopoly inferences, all keyed to one comparison — each firm's capacity against total market demand — and to the separation of equilibrium existence from uniqueness and stability.

Boundary-drawing (the capacity-to-demand ratio decides whether a pure-strategy equilibrium exists at all). The signature move is to make equilibrium existence a parameter to check rather than a fact to assume. The analyst reasons FROM "capacity comfortably exceeds demand" TO "the Bertrand best-response function stays intact, price is driven to marginal cost, a clean stationary pure-strategy equilibrium exists"; FROM "capacity binds below total demand" TO "each firm retains residual demand at elevated prices because its rival cannot serve the whole market, the best-response function discontinues, and no pure-strategy equilibrium exists." The load-bearing inference is that adding a perfectly realistic constraint does not merely shift the equilibrium price but can abolish existence — so the analyst reasons FROM the capacity ratio TO which regime obtains before attempting to solve for a price.

Predictive (binding capacity → endless price cycling as the mixed-strategy outcome). From the binding-capacity branch the framework predicts non-stationary behavior even with stable demand, stable costs, and rational firms: prices oscillate continuously between the competitive floor and the monopoly ceiling, the only consistent (mixed-strategy) solution. The analyst reasons FROM "each firm can skim residual demand at a higher price, so any proposed price invites defection upward, then redefection" TO "prices never settle — expect cycling." The prediction is qualitative and regime-specific: a capacity-bounded oligopoly is predicted to show price cycles, not a stationary level.

Diagnostic (reclassify observed cycling from misbehaviour to equilibrium). The framework licenses reading price cycles in a capacity-constrained oligopoly as the predicted equilibrium rather than as collusion or irrationality. The analyst reasons FROM "weekly price cycles on an airline route, in gasoline retail, or in hotel rooms, where each seller's capacity is below market demand" TO "this is the mixed-strategy response to finite capacity, not a coordination signal," and accordingly shifts the antitrust evidentiary burden. The diagnostic inverts the naïve reading: cycling is evidence for competition under binding capacity, not against it.

Boundary-drawing (the artefact exposure, and the substrate edge). The paradox functions as a meta-inference about models: it exposes that the clean Bertrand result rested on the unlimited-capacity assumption, so the analyst reasons FROM "this price-competition model assumes capacity away" TO "its equilibrium is an artefact — restore a binding capacity ratio and existence may vanish." This is what motivates the mixed-strategy treatment and underwrites the Kreps-Scheinkman reconciliation, where the same capacity-to-demand handle recovers the Cournot quantity from capacity-then-price competition. The same logic marks the concept's edge: the inferences require sellers, a continuous price action, marginal cost, a demand curve, and the Nash solution concept, so the bare residue — a binding constraint added to a continuous-action game can destroy pure-strategy equilibrium existence — travels to other discontinuous games (matching with indivisibilities, war-of-attrition collapses) only as a general game-theoretic shape, while the price-cycling content is bound to the oligopoly price-competition substrate.

Knowledge Transfer¶

Within oligopoly theory the paradox transfers as mechanism: the boundary check (compare each firm's capacity to total demand to decide whether a pure-strategy equilibrium exists at all), the prediction (binding capacity yields endless price cycling between the competitive floor and the monopoly ceiling), and the diagnostic (read observed cycling as the mixed-strategy equilibrium, not as collusion) carry intact wherever there are price-setting sellers with finite capacity facing a shared demand curve. So the apparatus moves without translation across the home domain's applications: from the industrial-organization classroom (motivating mixed-strategy equilibria and the Kreps-Scheinkman two-stage capacity-then-price model that recovers the Cournot quantity), to pricing analytics (supplying the structural reason for observed weekly price cycles among capacity-constrained airlines on a route, gasoline stations, or hotel rooms in a city), to antitrust analysis (informing whether oscillating prices are read as competitive or collusive), to auction-theoretic models where related discontinuities arise as capacity is exhausted mid-allocation. The industry varies; the capacity-to-demand ratio and its binary regime read the same in each — and all four of these are the same price-competition substrate restaged, not genuinely distinct domains.

Beyond price competition the transfer splits into a portable diagnostic question and a home-bound content, and the honest characterisation of the deeper structural residue is a shared abstract game-theoretic pattern carried one level up, not the Edgeworth paradox itself. The price-cycling content — Bertrand undercutting, marginal-cost pricing, the residual-demand skim, the monopoly ceiling — is bound to the oligopoly price-competition substrate and does not travel; a "matching market that won't settle" is not exhibiting an Edgeworth paradox. What does generalise is the structural shape the paradox instantiates: a binding constraint added to a game with a clean continuous-action equilibrium can destroy pure-strategy equilibrium existence and force a mixed-strategy resolution. That shape genuinely recurs across game-theoretic substrates as co-instances — discontinuous best-response functions generally, existence failures in matching markets with indivisibilities, mixed-strategy collapses in war-of-attrition games — but it recurs as the parent pattern (the discontinuous-game existence apparatus, the territory of the Nash, Glicksberg, and Reny existence theorems for which Edgeworth is a clean counter-example), not as the named paradox. When the cross-domain lesson is needed, it should be carried by that parent, and operationalised as the genuinely transferable diagnostic the seed names: in any competitive setting with a clean continuous-action equilibrium, ask what real-world constraint — capacity, lumpiness, transaction cost, regulation — was assumed away, and whether adding it back preserves equilibrium existence. That question carries; the Edgeworth-specific price-cycling story does not. Invoking "an Edgeworth paradox" outside price competition is therefore analogy — borrowing the destabilisation-by-realistic-constraint shape while leaving behind the demand-curve, capacity, and Bertrand machinery that make the original a definite result (see Structural Core vs. Domain Accent).

Examples¶

Canonical¶

Take the textbook capacity-constrained duopoly. Market demand is Q = 100 − p, marginal cost is zero, so the monopoly price is p = 50 (Q = 50). Give each firm a capacity of 40 units, so joint capacity 80 sits below the 100 units demanded at price zero. Suppose firm B prices at marginal cost and sells its full 40. Firm A now faces residual demand of (100 − p) − 40 = 60 − p, on which it acts as a monopolist: maximizing p(60 − p) gives p = 30, q = 30, feasible since 30 < 40, earning profit 30 × 30 = 900 > 0. So pricing at marginal cost is not a best response — A defects upward to 30. But at 30, B in turn wants to shade just below and skim, and the chase never terminates: no stationary price survives, and prices wander between the competitive floor (0) and the monopoly ceiling (50).

Mapped back: The two firms with zero unit cost are the price-setting sellers; Q = 100 − p is the shared demand curve; the 40-unit ceiling below 100 is the binding capacity constraint. Residual demand 60 − p that A monopolizes for a profit of 900 is the residual-demand skim, which turns marginal-cost pricing into a losing move — the discontinuous best-response that voids the non-existence of pure-strategy equilibrium and drives the perpetual price cycle between 0 and 50. That joint capacity 80 < 100 is exactly the capacity-to-demand existence boundary, and the collapse of the Bertrand p = 0 outcome is the artefact exposure.

Applied / In Practice¶

Retail gasoline markets supply the clearest field realization, in the form of "Edgeworth cycles" formalized by Maskin and Tirole (1988) and documented empirically in Canadian and Australian retail markets — Noel's work on Toronto-area stations is a well-known case. Prices climb in a slow crawl toward a peak, then one station undercuts sharply, rivals follow within a day or two down to near cost, and the slow crawl restarts, producing an asymmetric sawtooth rather than a settled price. Because a single station's pumps and short-run throughput cannot serve the whole local market at the trough price, an undercutter cannot capture everyone, leaving residual demand that makes a price hike profitable — so the market never rests at marginal cost. Antitrust readers treat this pattern, on its own, as consistent with competition under binding capacity rather than as proof of collusion.

Mapped back: Neighboring stations are the price-setting sellers facing a shared local demand curve; each station's finite short-run throughput is the binding capacity constraint, so an undercutter's inability to serve everyone leaves the residual-demand skim that makes the upward "restore" phase profitable. The sawtooth is the perpetual price cycle and the very non-existence of pure-strategy equilibrium made observable; reading it as competition rather than collusion is the capacity-to-demand existence boundary doing diagnostic work in the field.

Structural Tensions¶

T1: Shifted level versus abolished existence (what a realistic constraint destroys). The intuitive expectation is that adding a capacity ceiling would merely raise the equilibrium price — a quantitative adjustment to the Bertrand answer. The paradox's force is that it does something categorically different: it abolishes the existence of any pure-strategy equilibrium, discontinuing the best-response function rather than moving where it crosses. This is why the result requires separating three properties normally fused — existence, uniqueness, stability — and why a "realistic" tweak can destroy the question rather than perturb the answer. The tension is that the most natural way to absorb the paradox (capacity makes prices higher) is exactly the misreading it exists to correct; the constraint does not relocate the equilibrium, it removes it. Diagnostic: Is the capacity constraint here being treated as shifting the equilibrium price, or as potentially abolishing pure-strategy equilibrium existence altogether?

T2: Cycling as competitive equilibrium versus cover for collusion (the diagnostic inversion and its risk). The paradox licenses reading price cycles in a capacity-bound oligopoly as the predicted mixed-strategy equilibrium, not as a coordination signal — a genuine correction that rightly shifts the antitrust evidentiary burden. But the inversion cuts both ways: because competitive Edgeworth cycling and collusive price coordination can produce superficially similar oscillations, the same result that exculpates honest competition can furnish cover for actual collusion dressed up as capacity-driven cycling. The tension is that the concept simultaneously prevents false positives (calling competition collusion) and enables false negatives (excusing collusion as competition), and distinguishing the two requires establishing that capacity genuinely binds below demand rather than accepting the cycle pattern as self-certifying. Diagnostic: Is capacity independently shown to bind below total demand here (competitive cycling), or is the Edgeworth reading being invoked to explain away a pattern capacity does not actually force?

T3: Artefact exposed versus benchmark retained (when the clean Bertrand result still earns its keep). The paradox exposes that price-equals-marginal-cost rested on the unlimited-capacity assumption, which tempts a wholesale dismissal of the Bertrand model as an artefact. But the idealized result retains real analytical value: as a benchmark, a teaching device, and an accurate prediction wherever capacity comfortably exceeds demand, the clean Bertrand equilibrium is not wrong so much as regime-specific. The tension is that the paradox both discredits the unlimited-capacity model (as a universal claim) and depends on it (as the baseline whose collapse it dramatizes), so the analyst must neither treat marginal-cost pricing as a substrate-free law nor discard it as a mere fiction. Which stance is right is decided entirely by the capacity-to-demand ratio the paradox foregrounds. Diagnostic: Does capacity comfortably exceed demand here (Bertrand benchmark valid) or bind below it (Bertrand result is the exposed artefact)?

T4: Pure-strategy non-existence versus mixed-strategy resolution (existence restored, interpretation strained). The paradox is often summarized as "no equilibrium exists," which overstates it: a mixed-strategy equilibrium does exist and is the consistent solution, realized as prices oscillating between floor and ceiling. Restoring existence via mixing is mathematically clean, but it trades a directly interpretable outcome (a stationary price) for a solution concept whose behavioural meaning — firms randomizing over prices — is harder to read onto real managers, even as the observed price cycles match its qualitative signature. The tension is that the resolution which rescues the game from non-existence does so at a concept whose empirical content is a distribution of prices rather than a price, so "the equilibrium" is real but not something any single period's price reveals. Diagnostic: Is the claim here that no equilibrium exists (overstated) or that no pure-strategy equilibrium exists while a mixed one does (correct, but harder to interpret)?

T5: Autonomy versus reduction (a named oligopoly paradox or an instance of discontinuous-game existence failure). The Edgeworth paradox has specific price-competition cargo — Bertrand undercutting, the residual-demand skim, the monopoly ceiling, the capacity-to-demand ratio — and within oligopoly theory it transfers as mechanism across airlines, gasoline, and hotels, all the same substrate restaged. But its deeper shape is not price-specific: a binding constraint added to a game with a clean continuous-action equilibrium can destroy pure-strategy existence and force mixing is the parent, the territory of the Nash, Glicksberg, and Reny existence theorems, recurring in matching with indivisibilities and war-of-attrition collapses. Invoking "an Edgeworth paradox" for a matching market that won't settle borrows the destabilization shape while leaving the demand-curve and capacity machinery behind. The tension is between a named result that anchors oligopoly pedagogy and the recognition that its cross-domain lesson belongs to the discontinuous-game existence apparatus. Diagnostic: Resolve toward the discontinuous-game existence parent (ask which assumed-away constraint, if restored, breaks equilibrium existence) when outside price competition; toward the Edgeworth paradox when price-setting sellers with binding capacity face a shared demand curve.

Structural–Framed Character¶

The Edgeworth paradox sits toward the structural end of the spectrum but stops short of the pole — best read as mixed-structural: a genuine game-theoretic result about equilibrium existence wearing heavy oligopoly-pricing vocabulary. On evaluative_weight it is nil — the non-existence of a pure-strategy equilibrium and the resulting price cycle praise and blame nothing; the entry is explicit that the endless cycling is the equilibrium, "not a coordination signal or a behavioural defect," so there is no defective move being convicted, only a discontinuous best-response and its consequence. Institutional_origin is slight in the relevant sense: the result is a mathematical fact about a continuous-action game once a binding constraint is imposed — a theorem in the territory of the Nash, Glicksberg, and Reny existence theorems — not an artifact of any agency, standard, or convention; Edgeworth derived a structure the game already had rather than legislating one. It is only weakly human_practice_bound: the actors are firms and the action is price-setting, so the substrate is human economic behavior, but the load-bearing content is the capacity-to-demand existence boundary and the discontinuous best-response, which hold as formal properties of the strategic form regardless of who instantiates them — the paradox does not dissolve when a particular market is removed the way a practice-constituted concept does. On import_vs_recognize it is unusually strong for a domain-specific entry: within oligopoly theory the boundary-check, the cycling prediction, and the diagnostic reclassification transfer as mechanism across airlines, gasoline, and hotels — the same result recognized intact, one substrate restaged.

Where it is pulled back from the structural pole is vocab_travels, which it fails. The operative vocabulary — Bertrand undercutting, marginal cost, residual-demand skim, monopoly ceiling, capacity constraint, mixed-strategy equilibrium over prices — is irreducibly oligopoly-pricing idiom, and none of it floats free of the price-competition substrate; a "matching market that won't settle" is not exhibiting an Edgeworth paradox, and invoking the name outside price competition is avowed analogy. The one portable structural skeleton is constraint-induced existence failure: a binding constraint added to a game with a clean continuous-action equilibrium can destroy pure-strategy equilibrium existence and force a mixed-strategy resolution. That skeleton is genuinely substrate-independent and recurs as co-instances (matching with indivisibilities, war-of-attrition collapses) — but it is precisely what the Edgeworth paradox instantiates from its umbrella, the discontinuous-game existence apparatus (the Nash/Glicksberg/Reny territory), not what makes "Edgeworth paradox" itself travel: the cross-domain reach belongs to that parent, while the demand-curve, capacity, and Bertrand machinery are exactly the parts that stay home. Its character: structural in skeleton — a real, evaluatively neutral, formally-derived constraint-induced existence failure with a mixed-strategy resolution — but stated in oligopoly price-competition vocabulary that pins it to its home domain, leaving it mixed-structural rather than a free-floating prime.

Structural Core vs. Domain Accent¶

This section pins down why the Edgeworth paradox is catalogued as a domain-specific abstraction rather than a prime — the line between the formal shape that lifts and the pricing content that stays is what the ruling turns on.

What is skeletal (could lift toward a cross-domain prime). Strip away the market and a thin relational structure survives: a game with a clean continuous-action equilibrium acquires a binding constraint that discontinues a player's best-response function, and the discontinuity abolishes the existence of any pure-strategy equilibrium, forcing a mixed-strategy resolution. The portable pieces are wholly abstract — a continuous action space, best-response functions, a constraint that binds below the level the clean result assumed away, a discontinuity at the binding point, and the categorical jump from a shifted equilibrium to an abolished one. That skeleton is genuinely substrate-portable: it recurs observer-free in matching markets with indivisibilities and in war-of-attrition collapses, which is exactly why the paradox reads as a clean instance of the discontinuous-game existence apparatus — the territory of the Nash, Glicksberg, and Reny existence theorems for which Edgeworth supplies a canonical counter-example. But that formal shape is the core it shares, not what makes it the Edgeworth paradox.

What is domain-bound. Nearly all the worked content is oligopoly-pricing furniture that does not survive extraction: Bertrand undercutting, the identical constant unit cost, the shared demand curve, the residual-demand skim a rival cannot capture, the monopoly ceiling and competitive floor between which prices cycle, the capacity-to-demand ratio that decides the regime, the Edgeworth/Maskin-Tirole price cycle, and the Kreps-Scheinkman capacity-then-price reconciliation with Cournot. These are the vocabulary, the instruments, and the empirical cases (airline routes, gasoline retail, hotel rooms) the discipline actually studies. The decisive test: remove the price-competition substrate — sellers, a marginal cost, a demand curve, a continuous price action — and there is no "paradox," only the bare fact that a constraint broke pure-strategy existence; a matching market that won't settle is not exhibiting an Edgeworth paradox. The naming vocabulary renames every component off the pricing substrate.

Why this does not clear the prime bar. A prime's vocabulary travels and its cross-domain transfer is recognition of the same mechanism, not analogy. The paradox's transfer is bimodal. Within oligopoly price competition — the IO classroom, pricing analytics, antitrust, capacity-constrained auctions — the boundary check (compare capacity to total demand), the cycling prediction, and the diagnostic reclassification of cycling as competitive equilibrium all travel intact as the same mechanism, one substrate restaged across industries. Beyond it — invoking "an Edgeworth paradox" for a matching market or a bargaining game that won't settle — the name reaches only by analogy: it borrows the destabilization-by-realistic-constraint shape while leaving the demand-curve, capacity, and Bertrand machinery behind. And when the bare structural lesson is what is needed cross-domain — ask which real constraint (capacity, lumpiness, transaction cost, regulation) the clean model assumed away, and whether restoring it preserves equilibrium existence — it is already carried, in more general form, by the discontinuous-game existence apparatus (the Nash/Glicksberg/Reny existence theorems) that the paradox instantiates. The cross-domain reach belongs to that parent; "Edgeworth paradox," as named, carries the price-cycling content that should stay home.

Relationships to Other Abstractions¶

Current abstraction Edgeworth Paradox Domain-specific

Parents (3) — more general patterns this builds on

Edgeworth Paradox presupposes Bertrand Paradox (Economics) Domain-specific

The Edgeworth Paradox presupposes the Bertrand benchmark because it is defined by restoring the binding-capacity assumption that makes Bertrand's marginal-cost equilibrium collapse.
Edgeworth Paradox is part of Mixed Strategy Equilibrium Domain-specific

A Mixed Strategy Equilibrium is the strict solution form underlying Edgeworth price cycling once no stable pure-price profile exists.
Edgeworth Paradox is part of Constraint Prime

A binding capacity Constraint below total demand is the strict constituent that discontinuously changes the price-competition best responses.

Hierarchy paths (20) — routes to 9 parentless roots

Edgeworth Paradox → Bertrand Paradox (Economics) → Oligopoly → Barrier to Entry → Access Friction → Boundary

Show alternative paths (19)

Not to Be Confused With¶

Bertrand competition / the Bertrand paradox. Bertrand's result: two firms with identical constant costs undercut each other to marginal cost, so even a duopoly prices competitively. The Edgeworth paradox does not refute this — it locates its hidden assumption (unlimited capacity) and shows that restoring a binding capacity ceiling discontinues the Bertrand best-response and abolishes pure-strategy equilibrium. Tell: does the model assume firms can serve the whole market (Bertrand, clean marginal-cost equilibrium), or does capacity bind below total demand (Edgeworth, cycling)? Edgeworth is Bertrand-with-capacity, not a rejection of Bertrand.
Cournot competition. The rival oligopoly model where firms choose quantities (not prices) and a stable equilibrium exists at a price above marginal cost. Edgeworth is price competition with capacity constraints — and Kreps-Scheinkman showed the two connect: capacity choice followed by price competition recovers the Cournot quantity. Tell: is the strategic variable output/quantity with a clean equilibrium (Cournot), or price with binding capacity yielding no pure-strategy equilibrium (Edgeworth)?
The Edgeworth box. A pure name collision: the Edgeworth box is the general-equilibrium diagram of a two-person, two-good exchange economy depicting the contract curve and Pareto-efficient allocations — a tool of welfare/exchange theory, unrelated to oligopoly price dynamics. Tell: is the object a rectangular diagram of mutually beneficial trades between two consumers (Edgeworth box), or the non-existence of a price equilibrium among capacity-constrained sellers (Edgeworth paradox)? Same eponym, different topic.
Collusion / price-fixing. Coordinated price-setting by firms to sustain supra-competitive prices. Edgeworth cycling is the competitive equilibrium under binding capacity — the predicted mixed-strategy outcome, not a coordination signal — which is exactly why it shifts the antitrust evidentiary burden. But the two can look superficially similar, so the reading requires independently showing capacity binds. Tell: is the oscillation forced by finite capacity and residual-demand skimming (Edgeworth, competitive), or sustained by mutual agreement above the competitive level (collusion)?
Discontinuous-game equilibrium-existence failure (the parent). The substrate-neutral skeleton — a binding constraint added to a game with a clean continuous-action equilibrium destroys pure-strategy existence and forces a mixed-strategy resolution — is the territory of the Nash/Glicksberg/Reny existence theorems, recurring in matching with indivisibilities and war-of-attrition collapses. The Edgeworth paradox is the oligopoly-pricing instance. Tell: outside price competition, ask which assumed-away constraint breaks equilibrium existence (the general apparatus); reserve "Edgeworth paradox" for capacity-constrained price-setting sellers on a shared demand curve. (Treated fully in earlier sections.)

Neighborhood in Abstraction Space¶

Edgeworth Paradox sits in a crowded region of the domain-specific corpus (4^th percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.

Family — Market Structure & Price Equilibrium (25 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-07-12