Oligopoly is the awkward middle market — too few firms for perfect competition's tools to work, too many for the clean monopoly model. The fix textbooks reach for is game theory: model each firm as a player who picks a strategy knowing the rival will react. Once you can read a 2×2 payoff matrix, most of an oligopoly chapter falls into place.
What Makes Oligopoly Different
An oligopoly is a market with a small number of large firms — think a few airlines on a route, the big cell-phone carriers, or aircraft manufacturers. Three features set it apart:
- Interdependence. Each firm's profit depends on what the other firms do. A cement maker cutting price by 10% only makes sense if rivals don't match.
- Barriers to entry. Like monopoly, oligopoly is protected by entry costs (capital intensity, brand, regulation, network effects).
- No single simple model. Behavior ranges from cutthroat price wars to tacit price-matching. The same chapter covers Cournot, Bertrand, kinked-demand, and cartels because no one approach captures every case.
Game theory is the unifying language because it focuses on the one thing oligopoly always has: each firm reasoning about a rival's choice.
Reading a Payoff Matrix
A payoff matrix lists each player's strategies along an axis and shows the resulting payoffs in each cell. The convention is (Row payoff, Column payoff).
Consider two carriers, A and B, each choosing High Price or Low Price. Industry profits are highest when both keep prices high; each individual firm earns more by undercutting when the rival keeps prices high.
| B: High | B: Low | |
|---|---|---|
| A: High | (10, 10) | (2, 14) |
| A: Low | (14, 2) | (5, 5) |
To read it: if A plays High and B plays Low, A earns 2 and B earns 14. Each firm picks the row or column that gives it the highest payoff given what the other is doing.
Dominant Strategies and the Nash Equilibrium
A dominant strategy is a choice that yields a higher payoff than the alternative no matter what the rival picks. Check each player in turn.
For A: if B picks High, A earns 10 by playing High versus 14 by playing Low → Low is better. If B picks Low, A earns 2 by playing High versus 5 by playing Low → Low is better. Low is dominant for A. By symmetry, Low is dominant for B.
Both firms play Low and earn (5, 5). That is the Nash equilibrium: no player can improve by unilaterally switching, given the other's choice. And it is worse for both than (10, 10), which both would have gotten by cooperating on High. That gap — the cooperative outcome both prefer, ruined by individually rational defection — is the prisoner's dilemma structure showing up in industrial pricing. The prisoner's dilemma explainer builds the same logic from the classic two-criminal setup.
Not every game has a dominant strategy. When neither player has one, you find Nash equilibria by best-response checking: for each cell, ask whether either player would want to deviate. If neither would, that cell is a Nash equilibrium.
Why Cartels Are Unstable
Cartels — explicit agreements to restrict output and keep prices high — are the cooperative (High, High) cell in the matrix above. They generate cartel-level profits if every member sticks to the quota. The catch is the same as the matrix: each member, given that the others are cooperating, has a private incentive to cheat by producing a little extra at the high price. If one cheats, profits shift to the cheater; if all cheat, the market collapses to the competitive-ish outcome.
OPEC's history shows the pattern: agreed quotas, occasional bursts of discipline, and chronic over-production by members who calculate the (Low, High) payoff is too tempting to forgo. Sustaining cooperation requires either legal enforcement (illegal in most antitrust regimes) or repeated interaction with credible punishment for defectors. Indefinitely repeated games with patient players can support cooperation through strategies like "tit for tat" — defect once, get punished by a rival's defection forever after — but the cooperation is always fragile.
Common Oligopoly Models in One Line Each
Same payoff-matrix logic, different choice variable:
- Cournot competition: firms simultaneously choose quantities, take the rival's quantity as given, and play best-response. Equilibrium output sits between the monopoly and competitive levels.
- Bertrand competition: firms simultaneously choose prices for identical products. With no capacity constraint, the only Nash equilibrium is price = marginal cost — the so-called "Bertrand paradox", a perfectly competitive outcome with just two firms.
- Stackelberg competition: one firm moves first on quantity; the follower best-responds. The leader earns more by exploiting that commitment.
- Kinked-demand model: prices are sticky because each firm believes rivals will match price cuts but not price increases — so neither raises nor lowers price even as costs shift modestly.
Conclusion
Oligopoly and game theory go together because oligopolistic firms cannot ignore each other, and game theory is the cleanest way to model that. Build the payoff matrix, look for a dominant strategy, then check the Nash equilibrium — the cell where no one wants to switch. The recurring lesson is the prisoner's-dilemma trap: cooperation is best for the industry, defection is best for each firm, and self-interested play sends the market to the worse joint outcome.
