Cambridge International AS and A Level Economics: Oligopoly and game theory (9708) – Revision Notes

Short recall notes for the Cambridge International AS and A Level Economics 9708 syllabus (for examination in 2026, 2027 and 2028), part of section 7.6.4. Topics 1 to 6 are AS Level content and topics 7 to 11 are A Level content; this unit is A Level only. For full explanations, start with the study guide. Then test yourself with the practice questions, tick off the checklist, visit the Cambridge Economics hub, or try the free diagnostics.

All figures are invented and illustrative.

Key definitions

Term	Meaning
Contestable market	A market where entry and exit are easy enough that the threat of entry disciplines incumbents
Sunk cost	Spending that cannot be recovered on leaving the market
Hit-and-run entry	Entering to take supernormal profit, then leaving quickly with little or no loss
Limit pricing	Setting price low enough to make entry unprofitable
Price competition	Rivalry through price: cuts, discounts, price wars, predatory pricing
Non-price competition	Rivalry through advertising, branding, quality, service, innovation and loyalty schemes
Interdependence	One firm’s best choice depends on rivals’ reactions
Collusion	Coordination between rivals to limit competition: overt (formal) or tacit (unspoken)
Dominant strategy	The best choice for a player whatever the rival does
Prisoner’s Dilemma	A game where each player’s rational choice leaves both worse off than cooperation would
Allocative efficiency	P = MC
Productive efficiency	Lowest possible average cost
X-inefficiency	Costs above the lowest possible because of weak cost control

Contestable markets

Features: low or no entry and exit barriers; little or no sunk cost; entrants can use technology as good as the incumbents’; good information; hit-and-run entry possible.

Implications: even one or two firms may price near average cost and earn normal profit; incumbents keep costs low; concentration is a poor guide to behaviour; policy can focus on lowering barriers.

Weaknesses: few real markets have zero sunk costs; incumbents can respond with price cuts, contracts or advertising; information is imperfect.

Method: sustainable price

1. Find the monopoly output where MR = MC, then price and profit.
2. Set price equal to average cost: revenue = total cost.
3. Solve the quadratic and take the larger output (the smaller root is not where the entrant is deterred).
4. Check that profit is zero at that price.

Worked reminder

Demand P = 30 − Q, marginal cost 10, recoverable fixed cost 36. Monopoly: MR = 30 − 2Q = 10, so Q = 10, P = 20, profit = 10 × 10 − 36 = 64. Limit price: (30 − Q)Q = 10Q + 36, so Q² − 20Q + 36 = 0, giving Q = 18 or Q = 2. Take Q = 18, P = 12. Average cost = 10 + 36/18 = 12. Profit is zero.

Price and non-price competition

Why non-price? Price cuts are copied, so margins fall for all. Non-price competition is slower to copy, builds loyalty and can make demand less elastic.

Costs of non-price competition: spending that may be matched and wasted; may raise prices; brand loyalty can act as a barrier to entry.

Benefits: better quality, choice and innovation.

Worked reminder: price cut with a matched response

If market demand has a price elasticity of −0.5, a matched price cut of 10% raises market sales by 5%. Revenue changes by 0.9 × 1.05 − 1 = −5.5%. Every firm loses revenue, so matched cuts are unattractive when demand is inelastic.

Collusion and the Prisoner’s Dilemma

Collusion is more likely when: few firms; similar products; high barriers; stable demand; easy to see rivals’ prices.

Collusion breaks down when: more firms enter; products differ; demand falls; there are secret discounts; competition law is enforced.

Method: reading a pay-off matrix

1. Cover one column. Which row gives the row player the higher payoff? Repeat for the other column.
2. If the same row wins both times, it is a dominant strategy.
3. Do the same for the other player.
4. If both have dominant strategies, the outcome is where they meet.
5. Add the payoffs. Compare with the cooperative outcome.

Worked reminder: advertising game

Pay-offs are profits in millions (A, B).

	B: no ads	B: advertise
A: no ads	30, 30	22, 34
A: advertise	34, 22	26, 26

If B does not advertise, A earns 30 or 34, so advertising is better. If B advertises, A earns 22 or 26, so advertising is again better. Advertising is dominant for both, and both end at (26, 26) with joint profit 52. If neither advertised, joint profit would be 60. Each ends up 4 worse off (30 − 26) than if neither had advertised.

Ways to escape: repeated play with punishment, agreement not to advertise (which may be illegal), or a price leader that rivals follow.

Efficiency in oligopoly

Type	Test	Oligopoly result
Allocative	P = MC	Usually not met when price exceeds MC
Productive	Lowest average cost	Possible with economies of scale, not guaranteed
X-efficiency	Costs kept at the minimum	At risk where competition is weak
Dynamic	Innovation over time	Possible if supernormal profit is reinvested

Worked reminder: deadweight loss

Demand P = 40 − Q, MC = 10 (Q in thousands). Competitive output: 40 − Q = 10, so Q = 30. Joint profit maximum: MR = 40 − 2Q = 10, so Q = 15 and P = 25. Deadweight loss = ½ × (25 − 10) × (30 − 15) = 112.5.

Short run: supernormal profit is possible; price wars can cause losses. Long run: high barriers keep supernormal profit, while contestability pushes it to normal.

Chains of reasoning to practise

Each chain should end with a judgement, not just a list.

• Low sunk costs, so entry is cheap and exit is easy, so profit above normal draws in entrants, so the incumbent holds price near average cost, so consumers gain. Judgement: depends on how real the entry threat is, and whether the incumbent can fight back.
• Few firms, so each watches rivals, so price cuts are matched, so price competition leaves everyone worse off, so firms turn to advertising and quality. Judgement: consumers gain if quality rises, but lose if the spending only raises prices.
• Dominant strategy to cut price, so the stable outcome is worse for both than cooperation, so firms are tempted to collude, so each is tempted to cheat. Judgement: stability depends on the number of firms, how fast cheating is seen and the strength of the law.
• Price above marginal cost, so output is below the competitive level, so a deadweight loss arises. Judgement: weigh this against lower costs from scale and profits spent on innovation.

Must-know distinctions

• Few firms is not the same as no competition. A contestable market can be efficient with one firm.
• Sunk costs are not all fixed costs.
• Overt collusion is a formal agreement; tacit collusion is unspoken.
• A dominant strategy is the best reply to every rival move. The best reply to one move is not enough.
• Allocative efficiency (P = MC) differs from productive efficiency (lowest cost).

Quick self-test

1. Give two features of a contestable market.
2. What is hit-and-run entry?
3. Why is a sunk cost a barrier but a resalable machine not?
4. Give two examples of non-price competition.
5. A matched 10% price cut raises market sales by 5%. Does revenue rise or fall, and by how much?
6. In the advertising game above, what is A’s dominant strategy?
7. What is the joint profit if both firms advertise, and what would it be if neither did?
8. Why does collusion tend to break down?
9. Give one reason an oligopoly might be productively efficient.
10. Calculate the deadweight loss when P = 40 − Q, MC = 10 and the firms set the joint profit-maximising output.

Answers

1. Low barriers to entry and exit; no or low sunk costs (or the same technology for entrants, or good information).
2. Entering a market to take supernormal profit, then leaving quickly with little loss if the incumbent responds.
3. A sunk cost is lost on exit, so entry is riskier. A resalable machine can be sold, so exit is cheap.
4. Any two of advertising, branding, quality, after-sales service, loyalty schemes or innovation.
5. It falls by 5.5%, since 0.9 × 1.05 = 0.945.
6. Advertise: it pays more whether B advertises or not.
7. 52 if both advertise, 60 if neither does.
8. Each firm earns most by cheating while rivals stick to the agreement, and new entry, product differences and falling demand make monitoring harder.
9. Large-scale production can lower average cost through economies of scale.
10. 112.5 (thousand), from ½ × 15 × 15.

Where marks are usually lost

• Describing a market as contestable because it has many firms, instead of because entry is easy.
• Failing to link a feature to its consequence for price, profit or efficiency.
• Reading a pay-off matrix from one player’s view only.
• Stating the outcome without saying why neither player can gain by changing alone.
• Treating collusion as stable without discussing the incentive to cheat.
• Writing that non-price competition is always good or always wasteful, with no judgement.
• Defining efficiency without applying it to the oligopoly in the question.
• Drawing a conclusion with no criteria, such as time period, number of firms or barriers.

Official syllabus

Cambridge International AS & A Level Economics 9708 syllabus for examination in 2026, 2027 and 2028, section 7.6.4, published by Cambridge University Press & Assessment.