Contestable markets

The big picture

Examiners keep making the same observation about this topic: candidates count firms, when contestability is about how easily potential competition becomes actual competition. The number of firms currently trading tells you very little. Across every industry in this theme, the height of entry and exit barriers predicts conduct better than concentration does: energy coordination collapsed the moment entry became cheap, while mobile networks, protected by £6–8bn of sunk mast investment, have faced no new entrant since 2003. Contestability, more than enforcement, is what disciplines powerful firms, which is why it deserves its own page rather than a footnote to oligopoly.

The theory

What makes a market contestable

In a perfectly contestable market, a firm can practise hit-and-run entry: enter, take the supernormal profit, and exit without penalty when incumbents respond. Baumol's claim follows: incumbents cannot hold price above average cost, because any margin would invite a raid. High sunk costs are exactly what break this. A failed entrant who cannot recover its outlay faces a penalty for trying, so the raid never comes, and the deterrent works through rational anticipation: entry is deterred even while incumbents visibly earn supernormal profit. Keep sunk costs and barriers to entry distinct in your writing: sunk costs are a barrier to exit whose anticipation deters entry, while entry barriers (scale, patents, brand) block the way in directly. The two mechanisms operate independently and both must be assessed.

How the threat of entry disciplines incumbents

Where a market is even partially contestable, incumbents change behaviour before any entrant appears. They keep costs lean, because X-inefficiency would hand an entrant its opening, and they may practise limit pricing: pricing below the profit-maximising price, low enough to deter new entrants. The trick works because entrants start small: at low volume their average cost sits above the incumbent's (Tesco's roughly £50bn of annual purchasing buys inputs cheaply), so the incumbent can price above its own average cost yet below the entrant's, and the entrant cannot break even. This also repairs the gap in Baumol's claim, which assumes entrants can match incumbent costs; where they cannot, some supernormal profit survives even under threat. The market does not need to be a monopoly to behave like one, and it does not need an entrant to behave competitively.

Application

• Supermarkets, the partial contestability case: full-service entry is blocked by sunk costs, 15–25 year store leases and fit-out (refrigeration, shelving, signage, loyalty-data systems) that a failed entrant cannot recover, and no like-for-like rival has reached scale against the Big 4 in decades. Yet Aldi (entered 1990) and Lidl (1994) grew from around 2% in 2005 to 18% by 2024 with structurally lower sunk costs: roughly 1,500 product lines against 30,000+, far smaller stores, no loyalty infrastructure. Contestability does not require replicating the incumbent; it requires that entry be viable for some business model.
• Energy, the extreme case: Bulb entered in 2017 for around £10m (the licence itself cost about £500k), pricing roughly 25% below Big Six bills; over 50 entrants followed, the Big Six share fell from 95% to about 68%, and consumers saved around £3bn a year on the CMA's estimate. The 28 supplier exits after the 2021–22 wholesale shock add the caution: easy exit means entrants leave in bad times too.
• Banking, contestability by policy: the FCA and PRA lowered barriers deliberately (New Bank Start-up Unit 2014, seven-day switching, Open Banking 2018), and challengers went from 0 to around 15% of the market between 2014 and 2024. Regulators can manufacture contestability where nature withholds it.
• Mobile, the non-contestable benchmark: £6–8bn of sunk mast investment and no new network since Three in 2003. Quote it as the contrast that makes the supermarket case look open.
• The CMA's role: blocking Sainsbury's–Asda in 2019 (a merged share of about 29%) preserved what contestability the grocery market has. The limits matter for evaluation: the CMA cannot force price down to average cost, cannot shrink incumbents' scale advantage, and cannot lower entrants' sunk costs, so intervention protects contestability rather than creating it.

Building the paragraph

Here is the supermarket chain, the one to rehearse for any contestability question. Topic sentence first, carrying mechanism, direction and headline evidence:

Because a failed full-service supermarket cannot recover its 15–25 year store leases or its fit-out, high sunk costs remove the threat of hit-and-run entry, which lets the Big 4 hold prices above average cost even though no law is broken and no rival is excluded by force.

1. Define contestability precisely: what matters is the ease with which potential competition becomes actual competition, so the analysis starts from entry and exit conditions rather than from the number of firms.
2. Identify the sunk costs: leases of 15–25 years, refrigeration, shelving, signage and loyalty-data infrastructure, none of it recoverable on exit.
3. Link to hit-and-run: an entrant cannot enter, take the supernormal profit and leave without penalty, because exit forfeits the sunk outlay; anticipating this, rational firms never try.
4. Draw the limit pricing diagram (this is your diagram step): the incumbent's scale, around £50bn of annual purchasing at Tesco, puts the entrant's average cost above the incumbent's, so a price above the incumbent's AC and below the entrant's makes entry loss-making before it begins.
5. Confirm against the record: no like-for-like full-service entrant has achieved scale against the Big 4 in decades, while incumbents visibly earned supernormal profit.
6. Link back: with the entry threat neutralised twice over, by sunk costs and by limit pricing, incumbents price above average cost, and the gap between the two is the consumer cost of low contestability.

Key terms

Term	Precise definition	Where it earns marks
Contestable market	A market with freedom of entry and exit, where potential competition can readily become actual competition.	The AO1 opener; examiners reward the "potential into actual" phrasing over counting firms.
Sunk costs	Irretrievable costs that a firm cannot recover if it exits the market.	The mechanism that breaks perfect contestability; the term examiners most miss.
Hit-and-run entry	Entering a market to capture supernormal profit and exiting without penalty once incumbents respond; feasible only where sunk costs are near zero.	Mention it explicitly: it is precisely what high sunk costs prevent.
Limit pricing	Pricing below the profit-maximising price, and low enough to deter new entrants.	The conduct evidence that a threat of entry is shaping incumbent behaviour.
Barrier to exit	Any cost of leaving a market, chiefly sunk costs; high exit barriers deter entry because entrants anticipate being trapped.	Separating entry and exit mechanisms shows the examiner two distinct chains.