Oligopoly

The big picture

The defining feature of oligopoly is what we will call inter-reactivity: each firm's pricing decision directly affects, and is affected by, its rivals' responses. Name it in every essay, because it is the single structural mechanism behind both kinds of behaviour you will be asked about. It drives price behaviour (collusion when coordination is easy, price wars when it is hard) and it drives non-price behaviour (advertising and loyalty-scheme arms races), so one concept organises the whole topic.

The best applied laboratory is the UK supermarket industry, because the same four firms behave in two opposite ways at once. On petrol they have sustained years of near-identical pricing; on groceries they fought a price war from 2014 to 2020. The incentive to collude is constant in oligopoly; what varies is the ability, and that is decided by the conditions of the market, product by product.

The theory

Characteristics and concentration ratios

We measure how concentrated a market is with an n-firm concentration ratio: the combined market share of the largest n firms. The UK grocery Big 4 (Tesco 27%, Sainsbury's 15%, Asda 14%, Morrisons 9%) hold 65%, with Aldi and Lidl on a further 18% combined by 2024, up from around 2% in 2005. The four UK mobile networks hold 94% between them. Alongside concentration, expect high barriers to entry, some product differentiation, price rigidity and heavy non-price competition, though each of these varies by industry and that variation is where evaluation marks live.

Game theory: the payoff matrix

The model to lead with is the prisoner's dilemma, built as a payoff matrix. Two firms each choose to price HIGH or LOW; the figures below are illustrative of the logic rather than audited accounts, a caveat worth writing in your answer too.

Payoffs: Tesco / Sainsbury's	Sainsbury's HIGH	Sainsbury's LOW
Tesco HIGH	£150m / £80m jointly optimal: £230m	£85m / £140m Sainsbury's undercuts
Tesco LOW	£210m / £40m Tesco undercuts	£90m / £65m Nash equilibrium: £155m

The one-shot game predicts mutual undercutting, so why do we observe stable high prices? Because inter-reactivity converts it into a repeated game. Forecourt prices are visible daily, so any cheat is detected at once and the threat of permanent retaliation ("cut once and I hold LOW forever") is credible. On the illustrative figures, cheating gains around £60m once, while permanent punishment costs around £60m every year afterwards: cooperation dominates within two years.

Collusion: overt, tacit, and the five conditions

Overt collusion, an explicit agreement to fix prices or output, is illegal under the Competition Act 1998; the CMA can fine cartels up to 10% of global turnover and directors face personal criminal liability. The 2011 dairy price-fixing case saw the Big 4 fined for coordinating milk and cheese prices, and the BA and Virgin Atlantic fuel-surcharge cartel was fined in 2007 after Virgin blew the whistle for immunity. Because overt collusion is prosecutable, firms reach the same (HIGH, HIGH) outcome through tacit mechanisms: price leadership (Tesco moves first, rivals follow within days), focal points such as round-number prices per litre, price-matching guarantees (which the CMA found in 2013 reduce the incentive to undercut), and information exchange through industry data and even price-comparison apps.

When is tacit collusion stable? Score the market against five conditions, a checklist the CMA itself uses: few firms; a homogeneous product; transparent prices; stable costs; no maverick. Supermarket petrol scores 5/5 (Aldi and Lidl sell no petrol, so there is no external maverick) and delivered roughly seven years of parallel pricing. Groceries score 3/5 (own-brand differentiation, promotional complexity, Aldi and Lidl as mavericks) and delivered a price war. Same firms, opposite conduct: collusion is product-specific.

Price wars, non-price competition, and being hit twice

Where the conditions fail or a maverick is present, the Nash logic takes over. The grocery price war forced by Aldi and Lidl saved consumers around £200 a year on average between 2014 and 2020, on the CMA's estimate. In energy, Bulb's entry in 2017 at around £950 against Big Six bills of roughly £1,200 triggered entry by more than 50 suppliers and cut the Big Six share from 95% to about 68%.

Meanwhile the same inter-reactivity produces a second prisoner's dilemma in non-price competition. If every firm stopped advertising and running loyalty schemes at once, market shares would barely move and all would save money; if only one stops, it loses ground; so defensive spending is the dominant strategy. The Big 4 spend around £625m per firm per year on loyalty schemes, own-brand development, refurbishment and online fulfilment (a directional estimate, treat it as such), while Aldi reached 18% spending roughly a third of that, which suggests around £400m of the spend is economically wasteful. Here is the framing to keep: this spending happens IN ADDITION TO supra-competitive prices, so consumers are hit twice, once through the price and again through costs passed on from an arms race that changes nothing.

The kinked demand curve is an optional extra model of price rigidity: rivals match cuts (demand inelastic below the current price) but ignore rises (elastic above), producing a discontinuity in MR that leaves price stuck even as costs change. It fits cases like overdraft rates clustering near 40% EAR across the big banks, but the payoff matrix explains more with fewer assumptions, so treat kinked demand as background rather than your lead model.

Application

Learn a small canon of industries with dated figures, and practise attaching each fact to a mechanism rather than quoting it as decoration.

• Supermarket petrol: all four raised prices within two weeks of each other repeatedly across 2016–23; the CMA investigated parallel pricing in 2022 and 2024 and estimated consumer detriment of around £900m a year in fuel during 2022–23; prices "rocket" up with costs and "feather" down. In an answer: the 5/5 conditions predict stable coordination, and the £900m figure quantifies the welfare loss your diagram shows.
• Mobile networks: four firms at 94%; contract prices locked between £28 and £35 from 2017 to 2024; consumers overpay by roughly £250m a year; no new network has entered since Three in 2003. Use the locked prices as evidence of tacit coordination, then the entry drought to explain why nothing disturbs it.
• Energy: the Big Six held 95% from 2010 to 2017; entry after 2017 saved consumers around £3bn a year on the CMA's estimate; then the 2021–22 wholesale shock forced 28 supplier exits. Use it both ways: entry destroys coordination, and forced price competition can be destructive in high-fixed-cost markets.
• Airlines: conditions score about 2/5 (Ryanair as an internal maverick pricing 30–50% below legacy carriers, volatile fuel costs, opaque dynamic pricing), worth around €5bn a year to consumers on the EC's estimate. Use it as the counter-case: oligopoly structure with near-competitive conduct.

Building the paragraph

A Level analysis is a chain, with every step applied to one industry. Here is the collusion chain on supermarket petrol, opened by a topic sentence that states mechanism, direction and headline evidence:

Because forecourt petrol satisfies all five conditions for stable coordination, the Big 4 can hold prices above the competitive level without ever communicating, which is why the CMA estimated consumer detriment of around £900m a year in fuel during 2022–23.

1. Define inter-reactivity and establish structure: the Big 4 hold 65% of groceries, all four run forecourts, and prices are displayed on totems and comparison apps, so every move is observed.
2. Build the payoff matrix (this is your diagram step): LOW is each firm's dominant strategy, so the one-shot Nash equilibrium is (LOW, LOW) at £155m joint profit, against a jointly optimal (HIGH, HIGH) at £230m.
3. Inter-reactivity converts the one-shot game into a repeated game: detection is instant, so the threat of permanent retaliation is credible, and a one-off gain of around £60m is outweighed by a £60m-a-year punishment.
4. Score the conditions: petrol is 5/5, few firms, homogeneous fuel, transparent prices, stable costs, and no maverick because Aldi and Lidl sell no petrol.
5. Predict and confirm: parallel pricing sustained for around seven years, with all four raising prices within two weeks of each other across 2016–23, and "rockets and feathers" asymmetry.
6. Close the welfare loop and link back: price above the competitive level transfers consumer surplus to producers and creates a deadweight loss, quantified by the CMA at around £900m a year, so coordination here operates against consumers.

Key terms

Term	Precise definition	Where it earns marks
Inter-reactivity	Each firm's pricing decision directly affects, and is affected by, rivals' responses (the defining feature of oligopoly, usually called interdependence).	Name it in every oligopoly essay; it is the mechanism behind both collusion and arms races.
Concentration ratio	The combined market share of the largest n firms in a market.	Quantifying structure in application (Big 4 at 65%; mobile at 94%).
Nash equilibrium	An outcome where no firm can improve its payoff by changing strategy unilaterally, given rivals' strategies.	Reading the payoff matrix correctly; explaining why (LOW, LOW) persists.
Tacit collusion	Coordination on price or output without any explicit agreement, through devices such as price leadership and focal points.	Explaining legal parallel pricing, and why enforcement struggles to reach it.
Cartel	A formal, illegal agreement between firms to fix prices, restrict output or share markets.	The overt case: Competition Act 1998, fines up to 10% of global turnover.
Maverick firm	A firm whose incentives lead it to defect from coordination, destabilising collusion from inside or outside the coordinating group.	The condition that most often decides the 5/5 scorecard (Aldi and Lidl; Ryanair).