Market failure: market power

• Describe the four market structures and explain what gives a firm market power.
• Calculate total, average and marginal revenue, average and marginal cost, and profit from a table of data.
• Explain why profit is greatest where MC = MR, and identify abnormal profit, normal profit and a loss on a diagram.
• Draw and explain the short-run and long-run diagrams for perfect competition, monopoly and monopolistic competition.
• Explain allocative efficiency and show the welfare loss of monopoly against the perfectly competitive outcome.
• Explain what makes a monopoly natural and compare the prices a regulator could set for it.
• Analyse collusive and non-collusive behaviour in oligopoly with a payoff matrix, and interpret a concentration ratio.
• Evaluate the advantages and risks of large firms and the three government responses to the abuse of market power.

A market structure describes the conditions under which the firms in an industry compete: how many firms there are and how large they are, whether their products are identical or differentiated, and how easily new firms can enter. Barriers to entry are obstacles that prevent or discourage new firms from entering a market. Four structures are studied, and they differ in how much market power each firm has.

Market power is the ability of a firm to set its price above marginal cost without losing all its customers. A firm has this ability when buyers cannot switch to an identical product from another seller, so the firm faces a downward-sloping demand curve: it can raise its price and still sell some output. Such a firm is a price maker. It chooses its price, and demand then decides how much it sells, so it cannot choose price and quantity independently. A firm with no market power is a price taker: it must accept the price set by the market, because it would lose every customer if it charged more.

• Perfect competition: very many small firms sell an identical (homogeneous) product to very many buyers, there are no barriers to entry or exit, and buyers and sellers have perfect information. No single firm can affect the market price, so every firm is a price taker.
• Monopoly: a single firm, or one dominant firm, supplies a product with no close substitutes, and high barriers to entry keep rivals out. The firm's demand curve is the market demand curve, which gives it the most market power of the four structures.
• Oligopoly: a few large firms supply most of the market behind high barriers to entry. The product may be identical (steel, oil) or differentiated (cars, air travel). Because there are so few firms, each firm's sales depend on its rivals' decisions, so the firms are interdependent.
• Monopolistic competition: many firms compete with no barriers to entry or exit, but each sells a slightly different product. Differentiation gives each firm a small amount of market power, limited by the many close substitutes.

Barriers to entry explain why market power can last. The main ones are economies of scale (a new, small firm has higher average costs than an established large one); legal barriers such as patents, licences and copyright; control of an essential resource; brand loyalty built by heavy advertising; high set-up costs; and network effects, where a product is worth more to each user as more people use it.

Monopoly, oligopoly and monopolistic competition are together called imperfect competition. In all three the firm is a price maker, but the degree of market power differs. It is greater the fewer close substitutes the product has, because demand for the firm's product is then less price elastic and a price rise loses fewer customers. It lasts longer the higher the barriers to entry, because abnormal profit then does not attract new competitors. A monopolist has the most market power, an oligopolist considerable power, and a monopolistically competitive firm only a little.

Market power matters because of what it does to price and output. A price taker expands output until price equals marginal cost. A price maker restricts output so that price stays above marginal cost, which means that some units consumers value at more than their cost of production are not produced. Resources are under-allocated to the good, and this is why market power is a market failure. Perfect competition is the benchmark against which it is judged: no real market meets every assumption of the model, but it shows what the outcome would be if no firm had any market power.

Market power is not the same as size. A very large firm that faces many rivals, or imports, or the threat of new entrants may have little power over its price. A small firm that is the only supplier in an isolated town may have a great deal. What counts is how easily buyers can switch to another seller, now or in the future.

A firm's revenue is what it receives from selling its output. Total revenue (TR) is price multiplied by quantity: TR = P × Q. Average revenue (AR) is total revenue divided by quantity: AR = TR ÷ Q. When every unit is sold at the same price, AR equals the price, so the AR curve is the demand curve the firm faces. Marginal revenue (MR) is the change in total revenue from selling one more unit: MR = ΔTR ÷ ΔQ.

Total cost (TC) is the cost of all the factors of production the firm uses. Average cost (AC) is the cost per unit: AC = TC ÷ Q. Marginal cost (MC) is the change in total cost from producing one more unit: MC = ΔTC ÷ ΔQ. On a diagram AC is U-shaped and MC cuts it at its lowest point. The reason is arithmetic. When the next unit costs less than the average (MC below AC) it pulls the average down, and when it costs more (MC above AC) it pulls the average up, so AC stops falling and starts rising exactly where MC crosses it.

Economists count more as cost than an accountant does. Economic cost includes the opportunity cost of the resources the owners supply themselves: the wage they could earn elsewhere and the return their capital could earn in its next best use. Normal profit is the minimum return needed to keep those resources in the firm, and it is included in the cost curves. A firm that earns normal profit therefore covers all its costs, including a return to its owners that is just enough for them to stay in the industry.

Profit is total revenue minus total cost: profit = TR − TC. Dividing both terms by quantity gives the form that is read from a diagram: profit = (AR − AC) × Q. The bracket is the profit on each unit. Multiplying it by quantity gives a rectangle whose height is AR − AC and whose width is Q.

• Abnormal profit: AR is greater than AC, so TR is greater than TC. The firm earns more than the minimum needed to keep it in the industry. Abnormal profit is a signal that attracts new firms, if they are able to enter.
• Normal profit: AR equals AC, so TR equals TC. The firm covers every cost, including the owners' opportunity cost. It has no reason to leave the industry, and other firms have no reason to enter.
• Loss: AR is less than AC, so TR is less than TC. The firm earns less than normal profit, which means its resources would earn more in another use. In the long run it leaves the industry unless the price rises or its costs fall.

A firm that aims at profit maximisation produces the output at which marginal cost equals marginal revenue: MC = MR. The reason comes from comparing the two for each extra unit. If MR is greater than MC, one more unit adds more to revenue than to cost, so producing it raises profit and the firm should expand. If MR is less than MC, the last unit added more to cost than to revenue, so not producing it raises profit and the firm should contract. Profit can rise no further only where the two are equal. At that output MC must be rising and cutting MR from below; if MC were falling through MR, the next units would add more to revenue than to cost and the firm would gain by expanding.

The rule fixes the quantity, not the price. The price is whatever buyers will pay for that quantity, which is read from the demand (AR) curve. Profit is then found by comparing AR with AC at that quantity. Every revenue and cost diagram in this topic is read in the same three steps.

1. Find the output at which MC = MR.
2. Go up (or down) to the AR curve at that output and read the price.
3. Compare AR with AC at that output: AR above AC is abnormal profit, AR equal to AC is normal profit, and AR below AC is a loss. The profit or loss rectangle has height AR − AC and width Q.

A data question gives a table of output with price and total cost, or with total revenue and total cost, and asks for some of the six quantities below. Each is one line of arithmetic, so the marks are for choosing the right two rows of the table.

• Total revenue = price × quantity.
• Average revenue = total revenue ÷ quantity. It equals the price.
• Marginal revenue = change in total revenue ÷ change in quantity.
• Average cost = total cost ÷ quantity.
• Marginal cost = change in total cost ÷ change in quantity.
• Profit = total revenue − total cost, or (average revenue − average cost) × quantity.

• The marginal revenue or marginal cost of the nth unit compares row n with row n − 1. If output rises by more than one unit between rows, divide the change by the change in quantity.
• Total cost at zero output is the firm's fixed cost. Total revenue at zero output is zero.
• With whole units MC may never exactly equal MR. The firm produces every unit whose MR is at least as great as its MC and stops before the first unit whose MC is greater than its MR.
• Round average cost to two decimal places when it does not divide exactly, but carry the unrounded value into later steps.

The table also shows two results that the diagrams rely on. Marginal revenue is below the price at every output after the first, because the lower price needed to sell one more unit applies to all the units. And average cost falls while marginal cost is below it (outputs 1 to 4) and rises once marginal cost is above it (outputs 5 and 6).

In perfect competition the price is set by the market, where market demand meets market supply. Each firm supplies a tiny share of an identical product, so it cannot affect that price. It can sell any amount at the market price, nothing at a higher price (buyers would go to another firm), and it has no reason to charge less. The firm's demand curve is therefore a horizontal line at the market price: demand for its output is perfectly elastic.

Along that line price, average revenue and marginal revenue are all equal: P = D = AR = MR. Average revenue equals the price because every unit is sold at the same price. Marginal revenue equals the price because the firm does not have to lower its price to sell one more unit, so each extra unit adds exactly the price to total revenue.

The firm cannot choose its price, so its only decision is how much to produce. It maximises profit at the output where MC = MR, and since MR is the market price this means the output at which marginal cost has risen to equal the price. In the short run the number of firms in the industry is fixed, because firms have not had time to enter or leave. What the firm earns depends on where the market price lies in relation to its average cost at the chosen output.

• Price above average cost: the firm earns abnormal profit equal to (P − AC) × Q.
• Price equal to the lowest average cost: the firm earns normal profit. The price line just touches the bottom of the AC curve, which is also where MC crosses it.
• Price below average cost: the firm makes a loss equal to (AC − P) × Q. Producing where MC = MR does not remove the loss, but any other output would make it larger.

The abnormal profit and the loss are short-run outcomes. Neither can last, because nothing stops firms from entering a profitable industry or leaving a loss-making one. The normal-profit case is the long-run equilibrium, and its diagram is built in the next section.

In the long run firms can enter or leave the industry, and in perfect competition nothing prevents them. Suppose firms are earning abnormal profit. The profit is a signal to firms outside the industry. New firms enter, so market supply shifts to the right and the market price falls. Each firm's horizontal AR = MR line moves down with the price, and each firm reduces its output along its MC curve. Entry continues for as long as any abnormal profit remains. It stops when the price has fallen to the lowest point of the AC curve, where every firm earns normal profit.

The adjustment from losses is the mirror image. Some firms leave, so market supply shifts to the left and the market price rises. The loss of each remaining firm shrinks as its price line moves up. Exit stops when the price has risen to the lowest point of AC and the remaining firms earn normal profit.

Long-run equilibrium is therefore P = MR = MC = minimum AC. Only normal profit is possible in the long run, because free entry competes away abnormal profit and free exit removes losses. The firm also produces at the lowest point of its AC curve, so the good is made at the lowest possible cost per unit.

Allocative efficiency means that resources are allocated to the goods that consumers value most: the quantity produced is the one at which the benefit consumers place on the last unit equals the cost of producing it. The condition is P = MC, or equivalently marginal benefit (MB) = MC. Price measures marginal benefit because it is what consumers are just willing to pay for the last unit. If P were above MC, another unit would be worth more to consumers than it costs to make, so too little is being produced; if P were below MC, too much is. At P = MC social surplus, the sum of consumer surplus and producer surplus, is at its maximum.

Perfect competition reaches this outcome because of its assumptions, so the conditions for allocative efficiency should be stated with it. Firms must be price takers, so that P = MR and the profit rule MC = MR delivers P = MC. There must be no externalities, so that the private benefit and cost in the market are also the benefit and cost to society. Buyers and sellers must have perfect information, so that the price consumers pay reflects the true benefit to them.

A monopolist is the only supplier of a product with no close substitutes, so the firm is the industry. Its demand curve is the market demand curve, which slopes downward: D = AR. High barriers to entry keep it in that position. Unlike a price taker it chooses its price, but the demand curve then fixes the quantity it can sell at that price.

For a firm with a downward-sloping demand curve, marginal revenue is less than price, so the MR curve lies below the AR curve. To sell one more unit the firm must lower its price, and the lower price applies to every unit, not only the extra one. Suppose (with invented figures) it sells 3 units at $16 and must cut the price to $14 to sell a 4th. It gains $14 from the 4th unit but loses $2 on each of the 3 units it was already selling, so marginal revenue is 14 − 6 = $8, well below the new price of $14.

For a straight-line demand curve, MR starts at the same point on the vertical axis and falls twice as steeply, so it meets the quantity axis halfway to where demand does. Marginal revenue is zero at the output where demand is unit elastic (PED = 1), and negative at larger outputs, where demand is price inelastic. Since marginal cost is positive, MC = MR can only hold where MR is positive, so a profit-maximising monopolist never produces on the price-inelastic part of its demand curve.

The monopolist maximises profit in the same three steps as any firm. Output is Qm, where MC = MR. The price is Pm, read from the demand curve above that output. Because MR lies below AR, the price at Qm is above marginal cost. This gap between price and marginal cost is how market power appears on a diagram: AR is greater than MC at the chosen output.

The gap is also the reason market power is a market failure. Price measures what consumers are willing to pay for the last unit, so Pm greater than MC means that consumers value one more unit at more than it would cost to produce. A competitive industry would go on producing until price equalled marginal cost. The monopolist does not, because selling more would force the price down on all its output. Output is therefore below the allocatively efficient level and resources are under-allocated to the good: the monopoly is allocatively inefficient.

Whether the monopolist earns abnormal profit depends on average cost at Qm, exactly as for any firm. If AR is above AC at Qm it earns abnormal profit. If the AC curve just touches the AR curve at Qm it earns normal profit. If AC lies above AR at Qm it makes a loss: being the only seller does not help if demand is too weak or costs are too high.

The long run is where monopoly differs most from perfect competition. Abnormal profit still signals that resources would earn more in this industry, but barriers to entry stop new firms from responding. Supply does not increase, the price does not fall, and the monopolist can go on earning abnormal profit indefinitely.

The cost of monopoly to society is measured by comparing it with a perfectly competitive industry that has the same demand and the same costs. Both are drawn on one diagram with market demand (D = AR), the monopolist's MR, and MC. If the industry were perfectly competitive, the MC curve would be the industry's supply curve, because each price-taking firm supplies along its own marginal cost curve.

• Competitive outcome: where D = AR meets MC, at quantity Qpc and price Ppc. Here P = MC, so the allocation is efficient.
• Monopoly outcome: where MR meets MC, at quantity Qm, with the price Pm read from the demand curve. Qm is less than Qpc and Pm is higher than Ppc.
• Consumers: they buy less and pay more, so consumer surplus shrinks. Part of what they lose is paid to the monopolist as a higher price on the Qm units still bought. That part is a transfer from consumers to the firm, not a loss to society.
• Society: the units between Qm and Qpc are no longer produced, although consumers value each of them at more than its marginal cost. The surplus that would have been earned on them goes to nobody. This triangle, between the demand curve and MC from Qm to Qpc, is the welfare loss.

A natural monopoly is an industry in which average cost keeps falling over the whole range of output that the market demands. It arises where fixed costs are very large compared with the marginal cost of serving one more customer, as in networks of water pipes, gas pipes, electricity cables and railway track. The more output the fixed cost is spread over, the lower the average cost, so economies of scale are never exhausted within the size of the market. One firm can therefore supply the whole market at a lower average cost than two or more firms could, each with its own network and only a share of the customers.

The barrier to entry here is the cost structure itself. A new entrant would start with a small share of the market and a much higher average cost than the established firm, so it could not match its price. Competition is not only absent: it would be wasteful, because it would duplicate the network.

1. Unregulated: the firm produces where MC = MR and reads its price from the demand curve. This gives the lowest output and the highest price of the three, abnormal profit and a welfare loss.
2. Price equal to marginal cost: the allocatively efficient output, where the demand curve crosses MC. Because AC is falling, MC is below AC, so this price is below average cost and the firm makes a loss. It could continue only with a subsidy from the government.
3. Price equal to average cost: where the demand curve crosses AC. The firm earns normal profit and needs no subsidy. Output is between the other two, so some welfare loss remains, but it is smaller than without regulation.

This creates a dilemma for policy. Breaking a natural monopoly into competing firms would raise average cost for all of them, so the usual remedy for market power does not apply. The choices are to regulate the firm's price, to subsidise marginal-cost pricing, or to own the firm publicly. Each is evaluated in the section on government intervention.

The defining feature of oligopoly is interdependence. With only a few large firms, each one's sales and profit depend on the prices, output and advertising of its rivals. A firm cannot decide its own price without predicting how the others will respond, so firms behave strategically. This makes the outcome uncertain: the same market can show high stable prices, a price war or heavy advertising, depending on how the firms expect each other to behave.

Collusion is cooperation between firms to limit competition between them, usually by agreeing on price or output. A formal agreement is a cartel: the members fix a common price or share out output quotas. Cartels between firms are illegal in most countries, so they operate in secret. Tacit collusion involves no agreement at all. Firms coordinate through their behaviour, most often by price leadership, in which one firm (usually the largest) changes its price and the others follow.

The aim of collusion is to act together as one monopoly. The firms restrict their combined output to the level at which the industry's MC = MR, charge the monopoly price read from the industry demand curve, and share the abnormal profit. The diagram is the monopoly diagram applied to the industry as a whole.

A collusive agreement is hard to keep, because each member has an incentive to cheat. At the cartel price, price is well above each firm's marginal cost. A firm that secretly sells more than its quota, or shades its price to win customers from the others, adds more to its own revenue than to its own cost. The gain goes to the firm that cheats, while the cost, a slightly lower market price, is spread across all the members. If one firm cheats it gains at the others' expense. If all do, output rises, the price falls towards the competitive level and the cartel collapses.

Collusion is easier to sustain when there are few firms, their products and costs are similar, demand is stable, entry is difficult and cheating is quickly detected. It is harder to sustain with more firms, differing costs, differentiated products, falling demand (each firm is tempted to cut its price to keep its sales), easy entry, and competition authorities that investigate and fine cartels.

In a non-collusive oligopoly firms do not cooperate, but they still watch each other. Each firm expects that a price cut would be matched by its rivals, because they will not accept losing customers, so nobody would gain market share and all would earn less. It also expects that a price rise would not be matched, so it would lose customers to the firms that kept their prices. Either change looks unprofitable, so prices tend to stay where they are, even when costs change. This is known as price rigidity.

The risk in competing on price is a price war: one firm cuts its price, rivals retaliate, and successive cuts follow. Consumers gain from lower prices in the short run. Firms' profits fall, the weakest firms may be forced out, and the market can end up more concentrated than before, with less competition in the long run.

For this reason oligopolists often prefer non-price competition. They compete through branding and advertising, product development, quality, customer service, loyalty schemes and packaging. A successful brand makes demand for the firm's product less price elastic, and it takes rivals longer to copy than a price cut, which can be matched in a day. Price competition still occurs, through discounts and special offers. It is most likely where products are nearly identical, so that price is the only way to win customers, or where one firm believes its costs are lower than its rivals' and expects to survive a price war.

Oligopoly is allocatively inefficient. Whether firms collude or not, each has market power and sets its price above marginal cost, so output is below the efficient level. The inefficiency is greatest under successful collusion, where the industry reproduces the monopoly outcome and its welfare loss. It is smallest during a price war, when price is driven down towards marginal cost.

Game theory is the study of decisions in which the result of each participant's choice depends on the choices of the others. That is the position of firms in an oligopoly, so a simple game shows why they are pulled towards collusion and towards cheating at the same time. The tool is a payoff matrix: a table showing each firm's profit for every combination of choices.

• Firm A's reasoning: if B sets a high price, A earns 14 from a low price and 10 from a high price, so low is better. If B sets a low price, A earns 6 from a low price and 4 from a high price, so low is better again. A low price is A's best choice whatever B does. A choice like this is called a dominant strategy.
• Firm B faces the same numbers, so it reasons in the same way. Both choose a low price and each earns 6.
• Incentive to collude: if both set a high price each would earn 10, so both would be better off than with 6. Their joint profit of 20 is the highest in the table.
• Incentive to cheat: if the rival keeps to a high price, undercutting it raises a firm's profit from 10 to 14. An agreement on high prices therefore gives each firm a reason to break it.
• Result: choices that are rational for each firm separately lead to the outcome that is worse for both. This is why collusion is unstable and why price wars are a risk. It is also why firms that expect to face each other year after year may keep prices high without any agreement: each fears that a cut now would be answered by cuts from its rival later.

A concentration ratio measures how far a market is dominated by its largest firms. The n-firm concentration ratio, CRn, is the combined market share of the n largest firms, where a firm's market share is its sales as a percentage of total sales in the market. The four-firm ratio, CR4, is the most widely used.

Suppose seven firms have market shares of 35%, 20%, 8%, 15%, 10%, 7% and 5% (invented figures). The four largest shares are 35, 20, 15 and 10, so CR4 = 35 + 20 + 15 + 10 = 80%. The shares must be ranked before they are added: the four largest are not always the first four listed.

The higher the ratio, the more concentrated the market. A CR4 close to zero indicates many small firms, as in perfect or monopolistic competition. A high CR4 indicates that a few firms supply most of the market, which is the structure of an oligopoly. A ratio of 100% from a single firm is a pure monopoly. No single figure marks the boundary between structures, so a ratio is used to compare markets, or one market over time, and not to classify a market by a fixed threshold.

• It does not show how the share is divided among the largest firms. Four firms with 20% each and one firm with 77% beside three with 1% each both give CR4 = 80%, but the second market is close to monopoly.
• It depends on how the market is defined. A firm may have a small share of a national market and a very large share of a local one, and the ratio changes with which products are counted as part of the market.
• It usually ignores imports and firms that could enter. A concentrated domestic industry that faces foreign competition has less market power than its ratio suggests.
• It describes structure, not behaviour. Two markets with the same ratio may differ completely if the firms in one collude and those in the other compete.

Competition authorities use concentration as a first screen, for example when judging whether a merger would create too much market power. Because of these limits a ratio is where an investigation starts, not its conclusion.

In monopolistic competition many firms sell products that are close but not perfect substitutes. Product differentiation means making a product different from rivals' products in real or perceived ways: design, quality, location, service or brand image. Because its product is a little different, each firm keeps some customers when it raises its price, so it faces a downward-sloping demand curve and is a price maker. Because many close substitutes exist, a price rise loses a large proportion of its customers, so that demand curve is relatively price elastic. It is flatter than the demand curve of a monopolist, whose product has no close substitutes.

In the short run the firm behaves like a small monopolist. It produces where MC = MR and reads its price from its AR curve. The diagram has the same shape as the monopoly diagram, with flatter AR and MR curves. Depending on where AC lies at that output, the firm earns abnormal profit or normal profit, or makes a loss.

The short-run result does not last, because there are no barriers to entry. If firms earn abnormal profit, new firms enter with similar products. The market's customers are now shared among more firms, so each existing firm's demand curve shifts to the left, and with more substitutes available it also becomes more elastic. Entry continues until each firm's AR curve just touches its AC curve at the profit-maximising output, where the firm earns normal profit. If firms make losses the process runs the other way: some leave, the remaining firms gain their customers, and their demand curves shift to the right until they earn normal profit.

Long-run equilibrium therefore has P = AC, as in perfect competition, but P is still above MC, because the demand curve slopes downward. Monopolistic competition is allocatively inefficient: consumers value another unit at more than its marginal cost, yet it is not produced. The point where AR touches AC is on the falling part of the AC curve, so each firm also produces less than the output at which its average cost would be lowest. Firms have excess capacity, and average cost is higher than it could be.

The inefficiency is smaller than under monopoly. The more elastic a firm's demand, the smaller the gap it can open between price and marginal cost, and free entry removes abnormal profit in the long run. In return for the inefficiency that remains, consumers get product variety: many versions of the product, in many places and styles, from which each can choose the one that suits them best. The trade-off favours monopolistic competition where tastes differ widely, as with restaurants or clothing, because variety is then worth more to consumers than the slightly higher price. It is less favourable where products differ in little but branding, because the higher cost then buys consumers very little.

Competition in this structure is mainly non-price competition. Firms advertise and keep adjusting their products to stay distinct, because a firm whose product is seen as identical to its rivals' loses its market power. Advertising adds to average cost. It can benefit consumers when it informs them, and it is a waste of resources from society's point of view when it only moves customers from one similar product to another.

Market power is a market failure, but the large firms that hold it can also bring benefits that many small firms could not. A judgement about a dominant firm has to weigh the two, and the weights differ from one industry to another.

• Economies of scale: as a firm's scale of output grows, its long-run average cost falls. It can buy inputs in bulk, use specialised machinery and labour, borrow more cheaply, and spread fixed costs such as research and marketing over more units. A large firm may therefore produce at a lower cost per unit than many small firms could, and if its costs are low enough its price can be below the price that a competitive industry of small firms would charge. In a natural monopoly the saving is greatest, because a single firm avoids duplicating the network.
• Research and development: abnormal profit is a source of funds for developing new products and cheaper methods of production, and the prospect of keeping that profit is the incentive to take the risk. Firms in perfect competition earn only normal profit in the long run, so they have neither the funds nor any protection from imitators. Patents rest on the same reasoning: a temporary monopoly is granted as the reward for innovation.

• Output and price: a firm with market power restricts output and charges a price above marginal cost. Consumer surplus falls, and the units that are no longer produced are a welfare loss.
• Consumer choice and quality: with few or no rivals there are fewer alternatives, and less pressure on the firm to keep quality up and costs down. Costs can drift above their minimum because nothing forces the firm to be efficient.
• Innovation may not follow: a firm protected by high barriers to entry can keep its profit without innovating, and may use its position to exclude or buy rivals that have better products.
• Distribution: abnormal profit goes to the owners of the firm, while the higher prices are paid by all consumers. Since owners of firms tend to have higher incomes than the average consumer, market power tends to widen inequality.
• Influence: large firms can press smaller rivals and suppliers for terms that favour them, and can lobby governments for rules that protect them from competition.

Having market power is not in itself illegal in most countries. Governments act against its abuse: conduct by a dominant firm, or by firms acting together, that restricts output, raises prices or keeps rivals out by means other than offering a better or cheaper product. Three responses are studied: legislation and regulation, government ownership, and fines. Each suits some sources of market power better than others.

Legislation and regulation. Competition policy rests on competition law (also called anti-monopoly law). The law prohibits agreements between rivals to fix prices or share out markets, and prohibits a dominant firm from abusing its position, for example by tying one product to another or by pricing below cost to drive out a rival. Competition authorities also review mergers in advance and can block one, or allow it only on conditions, if it would remove an important competitor. In extreme cases a firm can be ordered to split into separate companies. Regulation goes further where competition is not possible: for a natural monopoly a regulator sets a maximum price, or the total revenue the firm may collect, together with standards of quality. All of these are command-and-control measures: rules backed by monitoring and penalties.

• Strength: the law acts directly on the conduct that causes the harm, and the threat of action deters other firms. Banning an exclusionary practice lowers barriers to entry. As rivals enter, demand for the dominant firm's product becomes more price elastic, and its price moves down towards marginal cost.
• Strength: merger control prevents market power from forming, which is quicker and cheaper than removing it afterwards, and price regulation protects consumers where competition cannot exist.
• Limit, information: the regulator knows less than the firm about its costs and technology. A price cap may be set too high, leaving abnormal profit, or too low, discouraging the investment the network needs, and the firm has an incentive to overstate its costs.
• Limit, time and definition: investigations and appeals take years, during which the conduct may continue and the market may change. What counts as abuse, and how widely to define the market in which a firm is dominant, is disputed in every case.
• Limit, reach and capture: large firms operate across borders while most authorities are national, and a regulator that deals with one industry for many years may come to share its point of view (regulatory capture).

Government ownership (nationalisation) means that the state owns and runs the firm. It is used mainly for natural monopolies and for industries regarded as essential, such as water, railways and electricity grids. A state-owned firm does not have to maximise profit, so the government can instruct it to set its price equal to average cost, or equal to marginal cost with the loss covered from tax revenue, and to pursue social aims such as serving remote areas at the same price as cities.

• Strength: ownership removes the incentive to restrict output, and it removes the information gap between regulator and firm, because the government is the owner.
• Limit, efficiency: without the profit motive, or the threat of takeover or bankruptcy, managers are under less pressure to control costs. Average cost may rise, which can offset the benefit of a lower mark-up over cost.
• Limit, politics and funding: prices and investment may be set for electoral reasons and not to cover costs. Losses fall on taxpayers, and the firm's investment has to compete with schools and hospitals for public money, which is its opportunity cost.

Fines are financial penalties imposed on firms that have formed a cartel or abused a dominant position. Their purpose is deterrence: a firm deciding whether to break the law compares the profit from doing so with the penalty it expects to pay.

• Strength: fines are cheap for a government to administer compared with ownership or continuing regulation, they raise revenue, and a large, well-publicised fine is a warning to other firms.
• Limit, size: a fine deters only if it is greater than the profit the conduct earns, allowing for the chance of never being caught. A firm that expects a fine smaller than its gain can treat it as a cost of doing business, and may pass part of it on to consumers in higher prices.
• Limit, delay and remedy: cases are appealed for years, and a fine paid long afterwards does not bring back the rivals that were excluded or return the overcharge to consumers.
• Detection: a cartel is fined only if it is discovered. Leniency programmes offer immunity or a reduced fine to the first member that reports a cartel. They work by sharpening each member's incentive to cheat on the others, the incentive that the payoff matrix shows.

• Reading the price at the point where MC crosses MR. That point fixes the quantity. The price is on the AR curve at that quantity.
• Mixing up the two conditions. P = MC is allocative efficiency. P = AC is normal profit. A monopolistically competitive firm in the long run has the second without the first.
• Drawing the long-run tangency of monopolistic competition at the lowest point of AC. With a downward-sloping AR curve the tangency is on the falling part of AC.
• Shifting cost curves to show entry or exit. In perfect competition entry shifts market supply and so moves the firm's price line. In monopolistic competition it shifts the firm's demand curve.
• Shading the abnormal-profit rectangle as the welfare loss. The rectangle is a transfer from consumers to the firm. The welfare loss is the triangle between demand and MC over the units that are no longer produced.
• Treating every oligopoly as a cartel. Collusion is one possible outcome. Without it, interdependence leads to stable prices, non-price competition and the risk of a price war.
• Calling any large firm a natural monopoly. A natural monopoly needs average cost to be still falling at the output that meets the whole of market demand.
• Evaluating market power without a condition. Whether a dominant firm harms consumers turns on barriers to entry and on whether lower costs and profits are passed on or reinvested.