Chapter 17

Game Theory & Strategic Behaviour

College

Chapter 10 mentioned that oligopolists behave "interdependently" and that cartels resemble a prisoner's dilemma. Game theory is the formal language for exactly this: how rational agents should act when the best choice depends on what everyone else chooses too.

At a glance
Core ideaYour best move depends on what your rivals choose to do.
Key termNash equilibrium — mutual best responses; no one gains by deviating alone.
You can…Solve a sequential game by backward induction.
Watch outA Nash outcome need not be the best joint one — the prisoner's dilemma.
Theory

Games, strategies and Nash equilibrium

A game consists of players, the strategies available to each, and the payoffs each combination of strategies produces. A dominant strategy is one that is best for a player no matter what rivals do. A Nash equilibrium is a set of strategies, one per player, such that no player can improve their payoff by unilaterally switching strategy, given what everyone else is doing — it is a state of mutual best responses, not necessarily the best joint outcome.

Firm B: Low priceFirm B: High price
Firm A: Low priceA: 20, B: 20A: 50, B: 5
Firm A: High priceA: 5, B: 50A: 35, B: 35

Here, Low price is dominant for both firms (it beats High price whatever the rival does), so (Low, Low) with payoffs (20, 20) is the Nash equilibrium — even though (High, High) with payoffs (35, 35) makes both firms better off. This gap between individually rational and jointly rational outcomes is the classic prisoner's dilemma, and it is exactly why the OPEC cartel of Chapter 10 is so hard to sustain.

Explanation

Sequential games, credible threats, and repeated cooperation

When players move in sequence rather than simultaneously, the game is drawn as an extensive-form tree, and solved by backward induction: work out the best move at the last decision point first, then reason backward to the start. This often shows that a threat which looks intimidating is not credible, because carrying it out would hurt the threatener too.

A single prisoner's dilemma predicts non-cooperation. But if the same two players interact repeatedly — as real oligopolists and cartel members do — cooperation can become an equilibrium. A simple tit-for-tat or grim-trigger strategy (cooperate first, then punish any defection forever) sustains the jointly better outcome as long as players value the future enough. Formally, cooperation is sustainable when the discount factor δ satisfies:

δ ≥ (Deviation gain) ÷ (Deviation gain + Punishment loss)

This is the game-theoretic version of "the shadow of the future" — why real cartels sometimes hold together, and why trust between repeat trading partners substitutes for formal contracts.

Practical

Worked example: backward induction in a market-entry game

Step 1 — the setupAn Entrant decides whether to Enter a market or Stay Out. If it Enters, the Incumbent then chooses to Fight (a price war) or Accommodate (share the market).
Step 2 — the payoffsEnter→Fight: (Entrant −10, Incumbent −5). Enter→Accommodate: (5, 5). Stay Out: (0, 10).
Step 3 — solve the last move firstIf the Entrant has already entered, the Incumbent compares Fight (−5) to Accommodate (5) — it prefers Accommodate.
Step 4 — fold back to the first moveKnowing the Incumbent will Accommodate rather than Fight, the Entrant compares Enter (payoff 5) to Stay Out (payoff 0) — it should Enter.
Step 5 — the credibility lessonA pre-entry threat of "we'll fight any entrant" is not credible, because Fighting is worse for the Incumbent than Accommodating once entry has actually happened. Rational entrants should see through empty threats — unless the Incumbent can make Fighting genuinely cheaper for itself (e.g. by pre-committing to excess capacity), the threat won't deter entry.
Q&A
Q1Is a Nash equilibrium always the best possible outcome for the players?

No — the prisoner's dilemma is the standard counterexample. The Nash equilibrium (Low, Low) is not Pareto efficient: both players could be simultaneously better off at (High, High). Nash equilibrium describes what rational, self-interested players will actually do given no ability to make binding joint commitments — not what would be collectively best.

Q2Why do some cartels and duopolies avoid price wars in real life, if the dilemma logic predicts they should defect?

Because real interactions are usually repeated, not one-shot. If firms expect to face each other again and again, the threat of future punishment (a price war triggered by today's defection) can make honouring a high-price agreement individually rational — exactly the tit-for-tat / grim-trigger logic. This breaks down when firms discount the future heavily, or expect the interaction to end soon (e.g. a firm about to exit the market).

Q3What makes a threat "credible" in game theory?

A threat is credible only if carrying it out is actually in the threatener's own interest at the moment it must be carried out — i.e. it survives backward induction as part of a subgame-perfect equilibrium. An empty threat that would hurt the threatener more than compliance is rationally ignored by the other player, however loudly it is announced.

Q4How does game theory formalise the informal "kinked demand curve" idea from Chapter 10?

The kinked demand curve assumed rivals match price cuts but ignore price rises — an assumed reaction function. Game theory instead derives what rivals should rationally do as a best response, and asks whether that assumed behaviour is itself an equilibrium. It replaces an assumed rule of thumb with a testable equilibrium concept, which is why modern oligopoly theory is built on game theory rather than the older kinked-curve story.

Entrant Enter Stay Out Incumbent Fight Accommodate (−10, −5) (5, 5) (0, 10) Payoffs = (Entrant, Incumbent)
Backward induction: the Incumbent prefers Accommodate (5, 5) to Fight (−10, −5), so its Fight threat isn't credible — the Entrant should Enter.
Concept mind map

How the ideas connect

Every key idea in this chapter, branching from the core concept — use it to see the whole picture at a glance.

Payoff matrixDominant strategyNash equilibriumPrisoners dilemmaRepeated gamesCredible threatsCooperationGame Theory
Infographic

The key facts, visualised

Nash eq.
no player can gain by changing strategy alone
Dominant
a strategy best no matter what others do
Prisoners
both defect though cooperating would be better
Tit-for-tat
repeat-game rule that can sustain cooperation
Solved examples

Worked problems, step by step

Follow each solution line by line, then try to reproduce it on paper before moving on.

Example 1Two firms choose High or Low price. Low price is a dominant strategy for both. What is the outcome?

  1. A dominant strategy is best regardless of the rival.
  2. If Low dominates for each firm, both play Low.
  3. The Nash equilibrium is both charging Low.

Example 2In a one-shot prisoners dilemma, why do both players defect even though mutual cooperation pays more?

  1. Whatever the rival does, defecting gives a higher individual payoff.
  2. So defect is the dominant strategy for each.
  3. Both defect, reaching a worse outcome than cooperating.
Practice problem set

Now you try

Work each one out first, then tap to reveal the worked answer.

1What is a Nash equilibrium?
A set of strategies where no player can do better by unilaterally changing their own choice.
2Why is the prisoners dilemma important for oligopoly?
It explains why cartels are unstable: each firm has an incentive to cheat on a price agreement.
3How can repetition sustain cooperation?
The threat of future punishment (e.g. tit-for-tat or grim-trigger) can make cooperating today rational.
4What makes a threat credible?
It is credible only if carrying it out is in the threatener's own interest when the moment comes.
5What is a dominant strategy?
A choice that yields the best payoff for a player regardless of what rivals do.
6Why might a first mover gain an advantage?
By committing first, they can shape rivals' best responses in their favour, e.g. deterring entry.